Page 1

http://www.sv-jme.eu

59 (2013) 9

Strojniški vestnik Journal of Mechanical Engineering

Since 1955

Papers

499

Saut Gurning, Stephen Cahoon, Branislav Dragovic, Hong-Oanh Nguyen: Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain

Dimitrios Lyridis, Panayotis Zacharioudakis, Stylianos Iordanis, Sophia Daleziou: Freight-Forward Agreement Time series Modelling Based on Artificial Neural Network Models

517

Francesco Longo, Aida Huerta, Letizia Nicoletti: Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation

526

Davorin Kofjač, Maja Škurić, Branislav Dragović, Andrej Škraba: Traffic Modelling and Performance Evaluation in the Kotor Cruise Port

536

Bo Lu, Nam Kyu Park: Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals

547

Andrija Vujičić, Nenad Zrnić, Boris Jerman: Ports Sustainability: A Life Cycle Assessment of Zero Emission Cargo Handling Equipment

556

Eugeniusz Rusiński, Przemysław Moczko, Damian Pietrusiak, Grzegorz Przybyłek: Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure

564

Nikola Marković, Paul Schonfeld: Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals

511

Journal of Mechanical Engineering - Strojniški vestnik

Contents

9 year 2013 volume 59 no.


Strojniški vestnik – Journal of Mechanical Engineering (SV-JME) Aim and Scope The international journal publishes original and (mini)review articles covering the concepts of materials science, mechanics, kinematics, thermodynamics, energy and environment, mechatronics and robotics, fluid mechanics, tribology, cybernetics, industrial engineering and structural analysis. The journal follows new trends and progress proven practice in the mechanical engineering and also in the closely related sciences as are electrical, civil and process engineering, medicine, microbiology, ecology, agriculture, transport systems, aviation, and others, thus creating a unique forum for interdisciplinary or multidisciplinary dialogue. The international conferences selected papers are welcome for publishing as a special issue of SV-JME with invited co-editor(s). Editor in Chief Vincenc Butala University of Ljubljana Faculty of Mechanical Engineering, Slovenia Technical Editor Pika Škraba University of Ljubljana Faculty of Mechanical Engineering, Slovenia Editorial Office University of Ljubljana (UL) Faculty of Mechanical Engineering SV-JME Aškerčeva 6, SI-1000 Ljubljana, Slovenia Phone: 386-(0)1-4771 137 Fax: 386-(0)1-2518 567 E-mail: info@sv-jme.eu, http://www.sv-jme.eu Print DZS, printed in 450 copies Founders and Publishers University of Ljubljana (UL) Faculty of Mechanical Engineering, Slovenia University of Maribor (UM) Faculty of Mechanical Engineering, Slovenia Association of Mechanical Engineers of Slovenia Chamber of Commerce and Industry of Slovenia Metal Processing Industry Association Cover: The Luka Koper - Port of Koper is a multipurpose seaport in Slovenia, with core activities focused on handling and warehousing of a variety of goods. Luka Koper operates the largest container terminal in the Adriatic and is a major automotive hub in the Mediterranean, handling almost half a million cars annually. With its excellent geographical position, modern infrastructure and reliable hinterland connections Luka Koper is becoming the leading port operator serving the countries of Central and Eastern Europe. Image Courtesy: Jaka Jeraša, Luka Koper

International Editorial Board Koshi Adachi, Graduate School of Engineering,Tohoku University, Japan Bikramjit Basu, Indian Institute of Technology, Kanpur, India Anton Bergant, Litostroj Power, Slovenia Franci Čuš, UM, Faculty of Mech. Engineering, Slovenia Narendra B. Dahotre, University of Tennessee, Knoxville, USA Matija Fajdiga, UL, Faculty of Mech. Engineering, Slovenia Imre Felde, Obuda University, Faculty of Informatics, Hungary Jože Flašker, UM, Faculty of Mech. Engineering, Slovenia Bernard Franković, Faculty of Engineering Rijeka, Croatia Janez Grum, UL, Faculty of Mech. Engineering, Slovenia Imre Horvath, Delft University of Technology, Netherlands Julius Kaplunov, Brunel University, West London, UK Milan Kljajin, J.J. Strossmayer University of Osijek, Croatia Janez Kopač, UL, Faculty of Mech. Engineering, Slovenia Franc Kosel, UL, Faculty of Mech. Engineering, Slovenia Thomas Lübben, University of Bremen, Germany Janez Možina, UL, Faculty of Mech. Engineering, Slovenia Miroslav Plančak, University of Novi Sad, Serbia Brian Prasad, California Institute of Technology, Pasadena, USA Bernd Sauer, University of Kaiserlautern, Germany Brane Širok, UL, Faculty of Mech. Engineering, Slovenia Leopold Škerget, UM, Faculty of Mech. Engineering, Slovenia George E. Totten, Portland State University, USA Nikos C. Tsourveloudis, Technical University of Crete, Greece Toma Udiljak, University of Zagreb, Croatia Arkady Voloshin, Lehigh University, Bethlehem, USA President of Publishing Council Jože Duhovnik UL, Faculty of Mechanical Engineering, Slovenia General information Strojniški vestnik – Journal of Mechanical Engineering is published in 11 issues per year (July and August is a double issue). Institutional prices include print & online access: institutional subscription price and foreign subscription €100,00 (the price of a single issue is €10,00); general public subscription and student subscription €50,00 (the price of a single issue is €5,00). Prices are exclusive of tax. Delivery is included in the price. The recipient is responsible for paying any import duties or taxes. Legal title passes to the customer on dispatch by our distributor. Single issues from current and recent volumes are available at the current single-issue price. To order the journal, please complete the form on our website. For submissions, subscriptions and all other information please visit: http://en.sv-jme.eu/. You can advertise on the inner and outer side of the back cover of the magazine. The authors of the published papers are invited to send photos or pictures with short explanation for cover content. We would like to thank the reviewers who have taken part in the peerreview process.

ISSN 0039-2480 © 2013 Strojniški vestnik - Journal of Mechanical Engineering. All rights reserved. SV-JME is indexed / abstracted in: SCI-Expanded, Compendex, Inspec, ProQuest-CSA, SCOPUS, TEMA. The list of the remaining bases, in which SV-JME is indexed, is available on the website.

The journal is subsidized by Slovenian Book Agency. Strojniški vestnik - Journal of Mechanical Engineering is also available on http://www.sv-jme.eu, where you access also to papers’ supplements, such as simulations, etc.

Instructions for Authors All manuscripts must be in English. Pages should be numbered sequentially. The maximum length of contributions is 10 pages. Longer contributions will only be accepted if authors provide justification in a cover letter. Short manuscripts should be less than 4 pages. For full instructions see the Authors Guideline section on the journal’s website: http://en.sv-jme.eu/. Please note that file size limit at the journal’s website is 8Mb. Announcement: The authors are kindly invited to submitt the paper through our web site: http://ojs.sv-jme.eu. Please note that file size limit at the journal’s website is 8Mb. The Author is also able to accompany the paper with Supplementary Files in the form of Cover Letter, data sets, research instruments, source texts, etc. The Author is able to track the submission through the editorial process - as well as participate in the copyediting and proofreading of submissions accepted for publication - by logging in, and using the username and password provided. Please provide a cover letter stating the following information about the submitted paper: 1. Paper title, list of authors and affiliations. 2. The type of your paper: original scientific paper (1.01), review scientific paper (1.02) or short scientific paper (1.03). 3. A declaration that your paper is unpublished work, not considered elsewhere for publication. 4. State the value of the paper or its practical, theoretical and scientific implications. What is new in the paper with respect to the state-of-the-art in the published papers? 5. We kindly ask you to suggest at least two reviewers for your paper and give us their names and contact information (email). Every manuscript submitted to the SV-JME undergoes the course of the peer-review process. THE FORMAT OF THE MANUSCRIPT The manuscript should be written in the following format: - A Title, which adequately describes the content of the manuscript. - An Abstract should not exceed 250 words. The Abstract should state the principal objectives and the scope of the investigation, as well as the methodology employed. It should summarize the results and state the principal conclusions. - 6 significant key words should follow the abstract to aid indexing. - An Introduction, which should provide a review of recent literature and sufficient background information to allow the results of the article to be understood and evaluated. - A Theory or experimental methods used. - An Experimental section, which should provide details of the experimental set-up and the methods used for obtaining the results. - A Results section, which should clearly and concisely present the data using figures and tables where appropriate. - A Discussion section, which should describe the relationships and generalizations shown by the results and discuss the significance of the results making comparisons with previously published work. (It may be appropriate to combine the Results and Discussion sections into a single section to improve the clarity). - Conclusions, which should present one or more conclusions that have been drawn from the results and subsequent discussion and do not duplicate the Abstract. - References, which must be cited consecutively in the text using square brackets [1] and collected together in a reference list at the end of the manuscript. Units - standard SI symbols and abbreviations should be used. Symbols for physical quantities in the text should be written in italics (e.g. v, T, n, etc.). Symbols for units that consist of letters should be in plain text (e.g. ms-1, K, min, mm, etc.) Abbreviations should be spelt out in full on first appearance, e.g., variable time geometry (VTG). Meaning of symbols and units belonging to symbols should be explained in each case or quoted in a special table at the end of the manuscript before References. Figures must be cited in a consecutive numerical order in the text and referred to in both the text and the caption as Fig. 1, Fig. 2, etc. Figures should be prepared without borders and on white grounding and should be sent separately in their original formats. Pictures may be saved in resolution good enough for printing in any common format, e.g. BMP, GIF or JPG. However, graphs and line drawings should be prepared as vector images, e.g. CDR, AI. When labeling axes, physical quantities, e.g. t, v, m, etc. should be used whenever possible to minimize the need to label the axes in two languages. Multi-curve graphs should have individual curves marked with a symbol. The meaning of the symbol should be explained in the figure caption. Tables should carry separate titles and must be numbered in consecutive numerical order in the text and referred to in both the text and the caption as

Table 1, Table 2, etc. In addition to the physical quantity, e.g. t (in italics), units (normal text), should be added in square brackets. The tables should each have a heading. Tables should not duplicate data found elsewhere in the manuscript. Acknowledgement of collaboration or preparation assistance may be included before References. Please note the source of funding for the research. REFERENCES A reference list must be included using the following information as a guide. Only cited text references are included. Each reference is referred to in the text by a number enclosed in a square bracket (i.e., [3] or [2] to [6] for more references). No reference to the author is necessary. References must be numbered and ordered according to where they are first mentioned in the paper, not alphabetically. All references must be complete and accurate. All non-English or. non-German titles must be translated into English with the added note (in language) at the end of reference. Examples follow. Journal Papers: Surname 1, Initials, Surname 2, Initials (year). Title. Journal, volume, number, pages, DOI code. [1] Hackenschmidt, R., Alber-Laukant, B., Rieg, F. (2010). Simulating nonlinear materials under centrifugal forces by using intelligent crosslinked simulations. Strojniški vestnik - Journal of Mechanical Engineering, vol. 57, no. 7-8, p. 531-538, DOI:10.5545/sv-jme.2011.013. Journal titles should not be abbreviated. Note that journal title is set in italics. Please add DOI code when available and link it to the web site. Books: Surname 1, Initials, Surname 2, Initials (year). Title. Publisher, place of publication. [2] Groover, M.P. (2007). Fundamentals of Modern Manufacturing. John Wiley & Sons, Hoboken. Note that the title of the book is italicized. Chapters in Books: Surname 1, Initials, Surname 2, Initials (year). Chapter title. Editor(s) of book, book title. Publisher, place of publication, pages. [3] Carbone, G., Ceccarelli, M. (2005). Legged robotic systems. Kordić, V., Lazinica, A., Merdan, M. (Eds.), Cutting Edge Robotics. Pro literatur Verlag, Mammendorf, p. 553-576. Proceedings Papers: Surname 1, Initials, Surname 2, Initials (year). Paper title. Proceedings title, pages. [4] Štefanić, N., Martinčević-Mikić, S., Tošanović, N. (2009). Applied Lean System in Process Industry. MOTSP 2009 Conference Proceedings, p. 422-427. Standards: Standard-Code (year). Title. Organisation. Place. [5] ISO/DIS 16000-6.2:2002. Indoor Air – Part 6: Determination of Volatile Organic Compounds in Indoor and Chamber Air by Active Sampling on TENAX TA Sorbent, Thermal Desorption and Gas Chromatography using MSD/FID. International Organization for Standardization. Geneva. www pages: Surname, Initials or Company name. Title, from http://address, date of access. [6] Rockwell Automation. Arena, from http://www.arenasimulation.com, accessed on 2009-09-07. EXTENDED ABSTRACT By the time the paper is accepted for publishing, the authors are requested to send the extended abstract (approx. one A4 page or 3.500 to 4.000 characters). The instructions for writing the extended abstract are published on the web page http://www.sv-jme.eu/ information-for-authors/. COPYRIGHT Authors submitting a manuscript do so on the understanding that the work has not been published before, is not being considered for publication elsewhere and has been read and approved by all authors. The submission of the manuscript by the authors means that the authors automatically agree to transfer copyright to SV-JME and when the manuscript is accepted for publication. All accepted manuscripts must be accompanied by a Copyright Transfer Agreement, which should be sent to the editor. The work should be original by the authors and not be published elsewhere in any language without the written consent of the publisher. The proof will be sent to the author showing the final layout of the article. Proof correction must be minimal and fast. Thus it is essential that manuscripts are accurate when submitted. Authors can track the status of their accepted articles on http://en.svjme.eu/. PUBLICATION FEE For all articles authors will be asked to pay a publication fee prior to the article appearing in the journal. However, this fee only needs to be paid after the article has been accepted for publishing. The fee is 300.00 EUR (for articles with maximum of 10 pages), 20.00 EUR for each addition page. Additional costs for a color page is 90.00 EUR.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9 Contents

Contents Strojniški vestnik - Journal of Mechanical Engineering volume 59, (2013), number 9 Ljubljana, September 2013 ISSN 0039-2480 Published monthly

Editorial

497

Papers Saut Gurning, Stephen Cahoon, Branislav Dragovic, Hong-Oanh Nguyen: Modelling of MultiMitigation Strategies for Maritime Disruptions in the Wheat Supply Chain 499 Dimitrios Lyridis, Panayotis Zacharioudakis, Stylianos Iordanis, Sophia Daleziou: Freight-Forward Agreement Time series Modelling Based on Artificial Neural Network Models 511 Francesco Longo, Aida Huerta, Letizia Nicoletti: Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation 517 Davorin Kofjač, Maja Škurić, Branislav Dragović, Andrej Škraba: Traffic Modelling and Performance Evaluation in the Kotor Cruise Port 526 Bo Lu, Nam Kyu Park: Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals 536 Andrija Vujičić, Nenad Zrnić, Boris Jerman: Ports Sustainability: A life cycle assessment of Zero Emission Cargo Handling Equipment 547 Eugeniusz Rusiński, Przemysław Moczko, Damian Pietrusiak, Grzegorz Przybyłek: Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure 556 Nikola Marković, Paul Schonfeld: Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals 564


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9 Guest Editorial

Guest Editorial Special Issue: Maritime and Port Logistics The International Conference on Material Handling, Constructions and Logistics 2012 was organized for academics from diverse backgrounds and interests to meet and to discuss critical and challenging issues under the themes of material handling and conveyance equipment and systems, construction and mining equipment (CME), constructions and design engineering, technical logistics, as well as maritime and port logistics (MPL). The guest editors of this special issue have selected six papers to address the themes relevant to MPL and one paper to address CME. An interesting paper that was previously accepted for publication within the Journal’s review process has also been included in this special issue. The focus of the selected papers is outlined in the following paragraphs. The first paper Modelling of multi-mitigation strategies for maritime disruptions in the wheat supply chain written by Saut Gurning et al. focuses on four major mitigation strategies to determine their suitability for managing potential disruptions in wheat supply chains. Markovian processes are applied to model supply chain (SC) risks, which are then incorporated into a framework for SC discruption strategies. The authors found that multi- disruption management scenarios across the SC network make a strong contribution to the stochastic approach. The second paper Freight forward agreement timeseries modeling based on artificial neural network models by Dimitrios Lyridiset et al. employs an artificial neural network (ANN) in order to forecast the future price of freight derivatives. Drawing on historical data for the period between January 2005 and March 2009, an ANN is built and trained, and the resulting model indicates to the investor which position to take in the derivatives market (short for sale of agreements and long for the purchase of agreements). The third paper Performance analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation by Francesco Longo et al. introduces a simulation model to recreate the complexity of a medium-sized Mediterranean seaport and analyses the performance evolution of such system with particular reference to the turnaround time of the ships. The verification and validation of the design of the experiments were carried out in order to evaluate how some critical factors (i.e. inter-arrival times, loading/unloading times, number of cars and trucks to be unloaded/ loaded) affect the performance of the seaport. The fourth paper in this Special Issue is Traffic modelling and performance evaluation in the Kotor Cruise Port written by Davorin Kofjač et al. Unlike previous research regarding cruise port logistics, this paper presents a simulation modeling the traffic in the Kotor cruise port, which is used for the port’s performance evaluation and to optimize operations. Furthermore, a complex cruise port revenue criteria function is introduced in order to maximize the port’s total revenue. Several simulations are performed, where the scenarios of an extended main berth and of increased traffic intensity are evaluated. The fifth selected paper is Sensitivity analysis for identifying the critical productivity factors of container terminals written by Bo Lu and Nam Kyu Park. The main purpose of this paper is to identify the various productivity factors in order to achieve a more positive influence on container terminal productivity. Sensitivity analysis provides a more appropriate benchmark for identifying the critical factors for productivity improvement. The efficiency results of 28 East Asian major container terminals were compared using data envelopment analysis and regression analysis. The sixth paper Ports sustainability: A life cycle assessment of zero emission cargo handling equipment written by Andrija Vujičić et al. investigates the zero emission concept for ports. In this study, the conventional and zero emission versions of the two most common machines found at container terminals, the rubber tired gantry crane and utility tractor rig, are compared. Although the transition from diesel to electric handling equipment can be regarded as a step forward, certain sustainability issues common with energy source transitions were necessary to rule out in order to draw conclusions. The seventh paper of this Special Issue is Experimental and numerical investigations of the jaw crusher supporting structure fatigue by Eugeniusz Rusiński et al., which experimentally and numerically investigates 497


reasons for the serious fatigue failures of jaw crusher supporting structures. A new design was investigated for dynamic and fatigue behaviour using the finite element method modified model. This modified structure was tested using a vibrations measuring device and the results confirmed proper behaviour during normal operation and absence of resonance problems. The final paper Scheduling for a single-terminal intermodal system recovery with Poisson arrivals written by Nikola Marković and Paul Schonfeld investigates a comprehensive model for the recovery of a single-terminal intermodal freight system from a disruption. A model that optimizes the schedule of vehicles on main routes assuming Poisson arrivals on feeder routes was developed. A genetic algorithm was used to optimize several case studies and sensitivity analysis confirmed the anticipated tradeoff in types of cost. We would like to take this opportunity to thank the editorial staff of Strojniški vestnik – Journal of Mechanical Engineering as well as the conference proceedings editors for their strong support and encouragement of this special issue. We would also like to express our appreciation to all the contributors and the anonymous reviewers of the papers for their time and effort. Guest Editors, Prof. Dr. Nam-Kyu Park Prof. Dr. Branislav Dragović Assist. Prof. Dr. Boris Jerman

498


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2012.941 Special Issue, Original Scientific Paper

Received for review: 2012-12-30 Received revised form: 2013-05-06 Accepted for publication: 2013-05-17

Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O. Saut Gurning1 – Stephen Cahoon2,* – Branislav Dragovic3 – Hong-Oanh Nguyen2 1 Sepuluh

Nopember Institute of Technology, Indonesia of Tasmania, Australian Maritime College, Australia 3 University of Montenegro, Maritime Faculty, Montenegro

2 University

This paper assesses potential mitigation strategies for maritime disruptions occurring during the containerised transport of wheat in wheat supply chains (WSC). The assessment focuses on four major mitigation strategies (inventory and sourcing mitigation, contingency rerouting, recovery planning, and business continuity planning) in order to determine their suitability for managing potential disruptions in the WSC. In this paper, a Markovian-based methodology is the prime means used to evaluate the WSC mitigation strategies. The results of the continuous time period of the Markov chain application suggest that optimised mitigation strategies include the measurement and prediction of WSC costs and time functions over a one-year period. Keywords: containerised wheat supply chains, multi-mitigation strategies modelling, maritime disruptions

0 INTRODUCTION There is a variety of operations involved in loading and transporting a container of wheat. One method is to base the process on the existing system used by wheat-grade producers or farmers: wheat is harvested, then cleaned, separated (by size or other characteristics), graded, and bagged either on the farm or at a nearby facility. The bags of wheat are then stockpiled in a warehouse or shipped directly via containers to the facility. During the shipping of containerised wheat, changes in the performance of wheat supply chains (WSC) may be identified beyond the assumptions predicted in a generic planning stage. These changes include factors such as changing demand, the origin of supply sources, distances, and lead times, all which may fluctuate widely resulting in a variety of at times, unexpected transportation costs. Supply chain optimisation models have traditionally treated the WSC with certainty and frequently ignore unpredicted events such as disruptions and disasters (for example [1]). In reality, operational parameter estimations may be inaccurate due to poor forecasts, measurement errors, changing demand patterns of the wheat commodity, inadequate sea transport infrastructure and managerial problems, all of which may vary substantially depending on the destination of the wheat. Moreover, even if all the variables of the WSC could be known with certainty, only some may be identified as causing disruptions. A major cause of disruptions tends to stem from the maritime leg of the WSC; for example, in the case of wheat and its derivative products, these may be inclement weather, sea terminal congestion, the requirements imposed by agencies within the marketing systems and a shortage

in the dry-bulk fleet [2]. Therefore, significant attention to potential maritime disruptions in a WSC is required, particularly as the wheat industry is more vertically integrated than in the past, and its supply chains are increasingly global, thus often necessitating a maritime segment [3]. In contrast, in [4], the lot size optimisation problem in supply chain management is solved using the artificial neural networks method. An integrated protocol for research and development-marketing integration, based on the theoretical framework for new product development is considered in [5]. An intelligent decision support system for the design process is developed in [6]. In [7], a combined method of a simulated annealing algorithm and the best priority rules for solving the problem of scheduling in multi-projects is presented. The objective of this paper is to provide a mitigation framework for both maritime service operators and users that can respond to various maritime disruptive events when managing a containerised WSC. This paper applies a maritime disruption model constructed for use in an AustralianIndonesian WSC context, which incorporates a WSC simulation, available data, and judgments from practitioners in order to quantify a disruption level arising from the contribution of situational attributes to maritime disruptions. The rest of the paper is organised as follows. The problem statement and maritime disruption strategies in a WSC are discussed in Section 2. Section 3 presents the mathematical model, followed by the Australia-Indonesia WSC computational experiments in Section 4. Conclusions are presented in Section 5.

*Corr. Author’s Address: Australian Maritime College, University of Tasmania, Locked Bag 1397, Launceston Tasmania 7250 Australia, S.Cahoon@amc.edu.au

499


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

1 PROBLEM STATEMENT AND MARITIME DISRUPTION IN A WHEAT SUPPLY CHAIN 1.1 The Stages of Maritime Disruption Understanding the stages of disruptive maritime events in the broad perspective of a WSC operation can be derived from the general logical approach of a common disruption framework. In general terms, some studies describe the stages of a disruption risk by the operational outcomes of disruptions and their effects on supply chain performance [8]. In terms of the scale of disruptions, these stages may experience a recurrent risk of individual events, ranging from delays to disaster events that demolish the service platform of a supply chain [9]. In total, there are four stages of disruption: delay, deviation, stoppages, and the loss of the service platform. The delay stage is the first stage through which a maritime risk passes. During this stage, the focus is on the recurrent changes in the performance of a service operation in the supply chain resulting in the cancelation of pre-determined planning by managers. During the deviation stage, planners of a service operation need to re-evaluate their external and internal service plan to account for the more significant changes to current operations resulting in forecasted objectives and service levels not being attained. The reduced level of operations can be contrasted against the stoppages stage, in which some existing services become unavailable due to direct and indirect factors interrupting the services’ provisions. The final stage is the loss of service platform, whereby the service platform is damaged, and as a consequence, service operators in the supply chain are unable to provide their services for a substantially longer period than in the third stage. The end result may be the continued unavailability of transport facilities and the shutting down of particular service points in the supply chain [10]. Within the four stages are events that may trigger maritime disruptions. These causal factors are referred to here as stimulators [11]. Potential stimulators include security threats, political riots or wars, lack of facilities and management at ports, lengthy customs and quarantine processes, severe weather conditions and earthquakes, electrical outages, a lack of maintenance, a shortage of ships, an insufficient number of empty containers, uncertainty of bunkering costs, communication failures, and a lack of inland accessibility [12] and [13]. The outcomes of a delay, deviation, stoppage, and/ or loss of service platform may generate unavoidable 500

divergences from the original plans of supply chain operations. In addition, those outcomes may also be regarded as stages through which disruptions evolve, starting from a delay to the loss of service platform or a disaster when unwanted internal or external changes occur in a supply chain. 1.2 A Review of Maritime Disruption Strategies The goal of mitigating maritime disruptions in the WSC is to alleviate the consequences of disruptions and risks or, simply put, to increase the robustness of a WSC through the maritime leg. However, there is little evidence of qualitative approaches being used to mitigate maritime disruptions, particularly during specific periods such as pre-disruption, at time disruption, and post-disruption. For example, the majority of supply disruption papers (as shown in Table 1 [12]) focus on the combination of contingency rerouting and inventory/ sourcing mitigation strategies in response to maritime disruptions. The research further finds that the dominant approaches by maritime users in managing the WSC during a disruptive event is to adapt to a new route on the maritime leg, use strategic stock (when no alternative source is available), utilise backup systems, and/or to implement business continuity actions (see Table 1). Table 1. Mitigation strategies for a WSC Mitigation

Inventory and Sourcing

Contingency Rerouting

Business Continuity Planning Recovery Planning

Strategies Inventory polling at ports Utilising agency service Apply other chain links Optimum ordering policy Postponement delays Supply flexibility Reserves routes Critical nodes mapping Applies other chain links Formal assessment Changes to work practices Max. allowable interruption Develop warning system Risk impact monitoring Apply discovery responses Apply recovery actions Network & proc. redesign

Literature [2] [14] [15] [14] [16] [17] [16] [9] [3] [18] [19] [20] [9] [21] [22] [22] [23]

Choices such as inventory pooling at ports, changes to working practices, applying other chain links, postponement delays, formal assessment of risk, determining the maximum allowable interruption, risk

Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O.


StrojniĹĄki vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

Fig. 1. A framework for maritime disruption strategies

impact monitoring, and re-evaluating contingency plans are considered as strategies that occur in the postdisruption stage. These choices, including constraints, are assessed in the context of five different maritime consequences as conditional events, such as normal, delays, deviation, stoppages, and loss of port services. Any input data included in calculating parametric variables may vary between scenario objectives. However, the way in which parametric variables are calculated from historical data and the way they are applied in the estimating process should be consistent within individual estimating systems. Input data was gathered from a maritime disruption survey which was collected both quantitatively and qualitatively and combined with the triangulation method. Fig. 1 shows a framework of maritime disruption strategies that was developed from the literature, which also reflects information from the maritime disruption survey and scenario assessments at various observations of maritime disruptive events [24]. The assessment of a maritime disruption management scenario comprises a component expressing the preparedness and strategic choices in the three stages of disruption: pre-disruption, at-time disruption and post-disruption [9], [12], [15] and [17]. The assessments specify the decision framework of senior managers’ action choices, such as changing variable values, copying strategies, or accepting or continuing the previous actions. However, the

performance is also dependent on instigating factors, inter-dependent factors and the combination of both as decider factors [9] and [23]. The above strategies appear to be the common steps in disruption risk management and contingency planning addressed by the literature (Table 1); however, these strategies tend not to consider the sequence of maritime disruptions that occur before, during, and after the disruption. The current study therefore differs in that it explores the optimised strategies during the various stages of maritime disruptions by applying qualitative and quantitative approaches to managing the delay, deviation, stop, and loss as consequences occurring during the sequence of maritime disruptions. Quantitative and qualitative input data were gathered from a maritime disruption survey in 2009/2010 from senior managers involved in the Australian-Indonesian WSC and triangulated with findings from previous studies to provide scenario assessments at various observations of maritime disruptive events (see Table 2). There are 16 disruption management scenarios listed in Table 2, all of which have ranges of costs attached to them. For example, S11 is the cost using inventory pooling ports (i = 1) for a delay risk (j = 1). The ranges of costs from S11 to S416 exist across all entities in the Australian-Indonesian WSC. The disruption management scenarios that can

Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain

501


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

be applied by senior managers when maritime disruptions occur are inventory pooling, agency utilisation, using other chain links, applying optimum ordering, postponing delays, using supply flexibility strategies, using reserved routes, mapping out the critical nodes, containerised shipment (as one of the business continuity responses), changing work practices, enabling allowable interruptions, applying warning systems, using implication monitoring, and developing an insurance package. In addition, senior managers can also set up a risk preparedness strategy or contingencies for operations management in both countries, which are used to detect and reduce the potential for maritime disruptions or commercial issues between sellers and buyers [25]. Table 2. The i-scenarios of disruption management and j-consequences indicators Scenario i i=1 i=2 i=3 i=4 i=5 i=6 i=7 i=8 i=9 i = 10 i = 11 i = 12 i = 13 i = 14 i = 15 i = 16

Type of DMS Inventory pooling ports Utilising agency service Control access to load Optimum ordering policy Postponement delays Supply flexibility Reserves routes Critical nodes mapping Apply other chain links Changes of working practices Max. allowable interruption Develop warning system Implication monitoring Insurance arrangement Re-evaluating contingency plan Network and procedure redesign

Consequences indicators J=1 J=2 J=3 J=4 S11 S21 S31 S41 S12 S22 S32 S42 S13 S23 S33 S43 S14

S24

S34

S44

S15 S16 S17 S18 S19

S25 S26 S27 S28 S29

S35 S36 S37 S38 S39

S45 S46 S47 S48 S49

S110

S210

S310

S410

S111

S211

S311

S411

S112

S212

S312

S412

S113 S114

S213 S214

S313 S314

S413 S414

S115

S215

S315

S415

S116

S216

S316

S416

Note: J=1 (Delay); J=2 (Deviation); J=3 (Stop) and J=4 (Loss)

2 MODELLING OF DISRUPTION MANAGEMENT ASSESSMENT The objective of modelling the disruption management process is to determine an optimal disruption strategy for events that recurrently and severely impact maritime services and the WSC process. The roles of the senior managers of WSC entities surveyed during the maritime disruptions study provided 502

input for a disruption management assessment. The subjective perspectives of respondents related to the flexibility factors and real-time responses to maritime disruptions have also been appraised in terms of total costs and time. The maritime disruptions occurring in the Australian-Indonesian WSC vary in both frequency and severity, from high probability and low consequence disruptions to low probability and high consequence disruptions. To account for this variability, a maritime disruption management (MDM) framework is proposed, incorporating a WSC simulation with available data and specific judgments by quantifying disruption levels and estimating the contribution of situational attributes to MDM. In estimating the level of maritime disruptive risks quantitatively, three questions need to be considered: i) What are the driving factors of MDM events?; ii) How likely are they to occur?; and iii) If they do occur, what are the consequences? To answer these questions, the scenarios listed in Table 2 will be used. Let si be the ith scenario, and pi and xi be its probability and consequence (either in costs and time), respectively. In the case of MDM, disruption risks can be defined as unexpected events that can be driven from instigating, interdependent, and decision maker factors resulting in four progressive factors, i.e. delays, deviations, stoppages, and loss of maritime services. The risk level of a maritime disruption scenario i is defined as follows:

Ri = pi (xi), (1)

where pi is a disruption probability and xi a consequence impact. A Markov chain of a WSC has a set of states in the WSC process denoted as S = (s1, s2, ..., sn). The process starts in one of these states and moves successively from one state (such as a farmer) to another (until the final consumers). Each move is called a step. If the WSC is currently in state si, then it moves to state sj in the next step at probability pij. This probability does not depend upon which state the chain was in before the current state. The probabilities pij are called transition probabilities (TP). The probability of remaining in the same state i is pi. An initial probability distribution, defined as S, specifies the starting state. In general, if a Markov chain has r states, then the following general theorem is easy to prove by using the above observation and induction. For a Markov Decision Process (MDP) defined by a finite set of states, S, a finite set of actions, A, and a transition function, T : S × A × S ' → [0,1] ,

Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

to N), which may occur from a normal state to failure where T ( s, a, s ') = Pr( s ' s, a ) , where Pr( s ' s, a ) ≡ Pr( st +1 = s ' st = s, at = a ) mode (λ) and may be recovered again due to responses Pr( s ' s, a ) ≡ Pr( st +1 = s ' st = s, at = a ) is the probability that or proper MDM strategies. action a in state s at time t will lead to state s' at time t + 1. A reward function R can also be defined P P 0 0 0 0 0 0 0 0 0 0 0  P as T : S × A → ℜ+ where ℜ+ is the set of positive P P P 0 0 0 0 0 0 0 0 0 0 0   real numbers. In addition, a disruption management P P P P P 0 0 0 0 0 0 0 0 0    0 0 0 0 0 0 0  0 P 0 P 0 P  0 policy is a function π : S → A and its expectation  0 0 0 0 0 0 0 0 0 0 0  P P P   as expected cumulative reward value function 0 0 0 P P P 0 0 0 0 0 0 0   0 π  0 P P P 0 0 0 0 0 0 0 0 0 0  S : V → ℜ, where ℜ is the set of real numbers.   0 0 0 0 P P P P P 0 0 0 0   0 The transition function as defined in a MDP is   0 0 0 0 0 0 0 0 0 0 0 P P P   Markovian, i.e. the probability of reaching the next  0 0 0 0 0 0 0 0 0 0 0 0  P P   0 0 0 0 0 0 0 0 0 0 P P 0 0   state depends only on the current state and action,  0 0 0 0 0 0 0 0 0 0 P 0 P 0    and not on the history of earlier states. Inclusion of 0 0 0 0 0 0 0 0 0 0 P 0 P   0  0 0 0 0 0 0 0 0 0 0 0 0 P P  the transition function allows MDPs to model and  reason with non-deterministic (uncertain) actions. Fig. 3. Markov transition matrix in the WSC Furthermore, the horizon may be either finite or infinite. If a MDP is solved over a finite horizon, then A MDM framework can help manage the the resulting policy is non-stationary, since the best disruptive events as a single strategy (λ1) or multiaction to perform may depend on the remaining time. disruption management strategy (λ1+ λ2+…+ λN). With If the horizon is infinite, then the resulting policy is analysis of the relevant data, the initial probability stationary. vector is calculated as: The four methods of MDM can be implemented V ji = X i DM ji , (2) for one particular WSC facing one maritime disruptive event both to reduce the likelihood of occurrence of a where Vji is a strategy value index for type j disruptive primary disruptive event, and to lower maritime risk event for scenario i; Xi probability of occurrence of after being in a normal (initial) stage, μ0. To depict scenario i; DMji type managed consequences of type the different approaches of MDM measures and its j related to the scenario i. processes, a diagram of MDM formalism is used in The mitigation functions are combined to simplify disruption management assessment and is shown in the evaluation of mitigation measures that typically Fig. 2. couple detection and recovery functions. Each decision node has a set of conditional probabilities that describe the probability of occurrence of each branch, conditional upon the previous states. The overall likelihood of each outcome is determined by multiplying conditional probabilities through the branch, and the risk level is aggregated along potential consequences in different branches as shown in Eq. (2). If, in addition, Rij is denoted as the reward corresponding to the transition from stage i to stage j, then: 11

12

21

22

13

23

31

32

33

34

43

35

45

53

54

411

56

65

67

66

76

77

78

86

87

88

89

810

98

99

910

108

1011

1112

1110

1211

1213

1312

1314

1413

Fig. 2. MDP from disruptive events to normal state

Fig. 3 shows the sequence of MDM events and the process from normal (initial) state to their possible consequences. MDM measures denoted by (λ) that reduce the probability of entering a disruptive stage are referred to as single or multi-disruption scenarios. Maritime stages such as port and shipping operations have more than one probable disruptive event (from 1

1414

Vi = ∑ j =1 pijxi ( Rij + β v j ) for i = 1,..., N , N

where the Vi is the value of each disruption strategy at stage i [12]. Note that the above formula can be written as: Vi = β ∑ j =1 pijxi v j + ∑ j =1 pijxi Rij N

N

Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain

for i = 1,..., N . (3)

503


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

If we denote by qixi the expected reward for the next transition if the current stage is i, then N qixi = ∑ j =1 pijxi Rij . Therefore, Eq. (3) becomes:

Vi = β ∑ j =1 pijxi v j + qixi N

for i = 1,..., N . (4)

Managed consequences are evaluated from the potential consequence that is mitigated by the susceptibility and coping capacity of the strategy. The consequences of MDM j relative to the scenario i (Dji), which is calculated through the Eq. (4), is the sum of all consequences referred to as the intensity threshold value m. Vjim is the vulnerability related to intensity m of the j type of consequence and related to the scenario i, DXjim is the potential consequence type j related to scenario i to consequences j and to intensity m [8] and [12].

as follows; λ32 the TP from stage 3 to stage 2; λ31 the TP from stage 3 to stage 1; λ30 the TP from stage 3 to stage 0; λ21 the TP from stage 2 to stage 1; λ20 the TP from stage 2 to stage 0; λ10 the TP from stage 1 to stage 0 and m the TP from state 0 to state 3.

DM j ,i = ∑ m ( DX j ,i ,m ⋅ V j ,i ,m ). (5)

2.1 Transition States of WSC The detailed interaction of 14 entities (as discussed in Section 3.1) upstream and downstream in the WSC including the disruption strategy options is explained by Fig. 3. Maritime stages such as port and shipping operations have more than one probable disruptive event (from 1 to N), which may occur from one state to another, including strategies that are also considered as transition probabilistic states (λ) and may recover again to a normal state due to responses or MDM strategies.

Fig. 4. Maritime disruption stages and scenario

The probabilities of internal stages for each risk event are further approached by using four different MDM stages; these are indicated by states 3, 2, 1 and 0. A description of each stage is as follows: Stage3: ensuring the normal level of the maritime service availability as planned; Stage 2: the occurrence of delays along maritime services, which creates a less efficient level of service performance; Stage 1: the events of deviations as the results of further divergences of maritime services; Stage 0: the conditions in which disruptions occur due to variable factors that result in various maritime services being unavailable. These four stages, as part of the Markov chain process, can also be defined as the transitions of maritime disruptions of which the value will depend on the combinations of actions. Assume that the probability functions for the maritime disruptive stages that are defined as states 3, 2, 1 and 0 for any operational period t that is continuously changing with t are F3(t), F2(t), F1(t), F0(t) respectively, where: F3 (t ) + F2 (t ) + F1 ( t ) + F0 (t ) = 1. (8)

2.2 Multi-State Scenarios of MDM

Let PC(t) be the state probability of X(t) at time t, so the probability of event at time t can be defined as:

Thus, to find the probability function for each stage, a system of differential equations based on MDP is applied on the assumption that the transition rates are relatively constant and can be estimated from historical records. With reference to these scenarios, each stage may be formulated as:

pc (t ) = P [ X (t ) ≥ C ] where C ∈ X , (6)

where C is an acceptable level of costs or time. The following system of differential equations for finding the four states’ probabilities (normal, delay, deviation, and loss of service) pt(t) for the Markov process can be written as:

dpi (t ) v v = ∑ j =1 p j ( t ) λ ji (t ) − pi (t )∑ j =1 λij ( t ). (7) dt

These four states or probability levels as shown in Fig. 4, where λij is the TP from stage of disruption i to stage of disruption j, can be explained in detail 504

dF3 (t ) / dt = − ( λ32 + λ31 + λ30 ) F3 ( t ) + µ F0 ( t ) , dF2 (t ) / dt = − ( λ21 + λ20 ) F2 ( t ) + λ32 F3 ( t ) , dF1 (t ) / dt = −λ10 F1 ( t ) + λ21 F2 ( t ) + λ31 F3 ( t ) ,

(9)

dF0 (t ) / dt = − µ F0 ( t ) + λ30 F3 ( t ) + λ20 F2 ( t ) + λ10 F1 ( t ) . Therefore, M(t) or total probabilities of some disrupted stages is equal to:

Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

M (t ) = F2 (t ) + F1 ( t ) + F0 (t ). (10)

From Eqs. (8) and (10), it follows that:

2.3 Optimised MDM Strategies The assessment framework for MDM strategies provides a preliminary reference to potential disruption management strategies derived from the MDP that meets the objectives and requirements of company disruption responses. However, in this model, it is possible to select the lowest costs or values that may occur for each strategy and the combination of the three disruption strategies taken. To do so, first, the value or the cost vi of each disruption strategy at state i is formulated by the steady-state condition of MDP:

Vi* − β ∑ j =1 pijxi v j = qixi N

for i = 1,..., N , (12)

Vi*

is the value or the cost for the policy where solution (see Eq. (4)). The process of disruption policy evaluation is initially started by solving the steady state cost equation using Xk as is the optimum disruption management policy. If vi* = vi for all i, then the calculation process that is finished as Xk has been achieved. Otherwise, increase k by 1 and let vi = vi* and do the iteration process of Vi * − vi ≤ ε , where ε = C (acceptable cost level). Therefore, the iteration process will stop if the condition of Vi * − vi ≤ tolerance value is achieved. In a steady-state condition, the value vi is equal to: vi = lim Vi (k ) for i = 1,..., N . (13)

k →∞

Then Eq. (13) becomes: Vi = qixi + β ∑ j =1 pijxi v j or Vi − β ∑ j =1 pijxi v j = qixi N

N

for i = 1,..., N .

(14)

Therefore, the steady-state disruption probability is π i = lim π i (k ) and then continued to: k →∞

π i = ∑ j =1 pijxi π j N

N j =1

pijxi π j − π i = 0 for i = 1,..., N − 1,

where ∑ j =1π j = 1. Recall that: Xi is a disruption management policy implemented in disruption stage i where Pi ∈ Ai; P = (X1, X2, …, XN) disruption management scenario; Vi(n) cost value of stage i at period n, n = 0, 1, 2, …, N; V(n) is {V1(n), V2(n),…, VN(n)} as vector of costs; πi(n) probabilities of disruption stage i at period n, n = 0, 1, 2, …, N; (n) is {π1(n), π2(n), …, πN(n)} as a vector of probabilities. The initial cost of the disruption management policy n in disruption stage i is Vi(n). By using the iteration solution, let k = 1 combined with the value of disruption management costs, then the minimum consequences can be calculated. The management formulation to address the minimum consequences (e.g. costs) at stage i is given here as: N

F3 (t ) = 1 − M (t ). (11)

for i = 1,..., N . (15)

As the set of equations has one redundant equation, then:

N   Vi ∗ = min qixi + β ∑ pijxi v j  for i = 1,..., N . (16) xi ∈Ai   j =1

where Ai is a set of disruption strategy ith action. From Eq. (16), let the optimum disruption management policy be Xk as the disruption management costs for minimisation can be estimated with Vi*. 2.4 Algorithm for Optimised Mitigation Scenario In Table 3, the algorithm for the optimised mitigation scenario is proposed. The solution method involves the following calculation: given a set of mitigation scenarios, in the first step, the number of entities in the WSC is incorporated into S (line 1) including the four categories of disruption stages and transitions λ (line 2) and mitigation scenarios Dji (line 3). Next, the probability of the mitigation scenario based on the mitigation policy of entities in the chain is applied (line 4). This is what is also assessed in the next algorithm step in order to estimate the probability at time t and acceptable level of costs C (lines 6 and 7). Due to the mitigation scenario taken, then the TP and transition costs can be calculated (lines 10 and 11). The objective level of multi-mitigation scenarios is identified in 13: Vi* correspondents to the minimal costs or values among various scenarios within available mitigation scenarios. If Vi* is undefined because no contradiction could be obtained (line 14), the level of costs or values is equal to Vi represents feasible disruption consequences (line 15). Otherwise, a recursive process is started (lines 16 to 22): Vi* is initialised (line 16) before the level of consequence (reward) of a scenario applied is identified (line

Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain

505


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

17). For each i scenario of a disruption j (line 18), the steady-state of disruption probability may be estimated (line 19 and 20). The set of all identified feasible scenarios is finally returned as the result of the algorithm (line 23). Table 3. Algorithm of multi-mitigation scenarios Line

Process

1

S ← Set up the number of entities in the chain (s1, …, sN) λ ← Apply four different stage of disruptions (λ3, λ2, λ1, λ0) Dji ← Set up mitigation scenario i with disruption j

2 3 4

Pr( s ' s, a ) ← Include probabilities of each actions C ∈ X, do Pc(t) ← P [x(t) > C] Mt ← F2(t) + F1(t) + F0(t)

5 for each 6 7

8 end for

9 for each i and

j ∈ N, do pij ← Transition probabilities DMji ← Transition costs

10 11

12 end for 13

N   Vi ∗ ← min qixi + β ∑ pijxi v j  j =1  

14 If

Vt*

is undefined then

20

N   Vi ∗ = qixi + β ∑ pijxi v j  j =1   * Vi ← Φ Vji ← V(Xi, DMji), V ∈ R (reward) for each i = 1, ..., N, do πj ← probabilities of disruption j at period n πi ← π (Pij, πj)

21

end for

15 16 else, 17 18 19

22 end if 23 return

S 3 EXPERIMENTAL STRATEGY

3.1 Australian-Indonesian Wheat Supply Chain On the basis of the previously implemented disruption management strategies, it was determined that a business continuity concept was implemented by wheat chain senior managers in the AustralianIndonesian WSC. In order to complete the disruption management assessment, it is necessary to determine which assumptions can be presented, particularly in unit costs needed, to apply the various strategy options. The assumption is relatively straightforward based on a monthly wheat shipment of one parcel volume of 8,201 tons. This is calculated from the basis 506

of 98,420 tons of wheat shipment in one year from ports on the east coast of Australia to either the Port of Tanjung Priok, Jakarta or Tanjung Emas, Semarang. In the maritime disruption assessment, attributes contributing to disruption occurrence from stages A to M (as shown in Fig. 5 [12]) are quantified in order to estimate the future risk level. Therefore, in this study, maritime disruption risks are quantified based on the maritime disruption survey, including professional judgment, elicitation and disruption management assessment of the wheat supply chain in the Australian-Indonesian trade link. The disruption risk at various stages of the wheat supply chain is denoted as s; Rs is calculated based on the snapshot of the traffic in that stage every time a wheat consignment enters it. The orientation of wheat cargo starts from local farmers in Australia and then enters the stages in Indonesia’s direction. The observed wheat flow, when entering the stages, first (A) calculates its own contribution to the stages disruption and then may contribute to the geometric mean of disruption value of that stage after being accumulated with other wheat at the same stage. Table 4 shows the results of the assessment of multi-disruption management scenarios of the study. When mitigating maritime disruptions on the Australian-Indonesian WSC, MDP proposes strategies (in the column of action name) that may be implemented by entities (state name column). State value is a contractual cost required by each entity when a disruption occurs to handle the total wheat shipment (line 15, Table 3). The value is calculated from the total tonnage of monthly wheat shipment (8,201 tons) and contractual costs allocated by entities. The value of the contractual costs was collected from the 2009 maritime disruption survey interview of 34 senior managers in the AustralianIndonesian WSC. Final cost is a maximum acceptable cost (C) of all companies (line 6, Table 3). The level of this cost is obtained from the value of state costs and the sensitivity factors ranging from 0.1 to 1%. In this case, 0.5% is applied. The step value is actually the minimum costs that will be occurred from 14 alternatives of scenarios given (see line 13 to 21, Table 3). The function of this value depends on action costs, decision value, probabilities, decision index, and discount value of costs applied across a one-year period, i.e. 365 days. Five policy scenarios are recommended: 1) inventory pooling; 2) postponement delays; 3) containerised shipment; 4) implication monitoring and 5) other chain links. Of these alternatives, containerised shipment (CS) is the most essential

Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

Fig. 5. The wheat supply chain process Table 4. Multi-mitigation scenarios Dynamic programming solver Type Title States Goal Actions Action/state Events Events/action Iteration type Policy steps Stop Difference Value error Probability error Time measure

MDP 2006-2007 Min 14 14 14 5 5 Policy 365 1.00E-06 3.00E-06 6.30E-09 Days

Index 1 2 3 4 5 6 7 8 9 10 11 12 13 14

State name Farmers Handlers Processors Australian shippers Australian forwarders Australian. shipping Australian ports Indonesian shipping Indonesian ports Indonesian forwarders Consignees Wholesalers Retailers Final consumers

State cost [US$] 1,766,596 124,995 124,995 1,833,260 1,999,920 999,960 666,640 708,305 374,985 1,666,600 2,083,250 2,124,915 2,166,580 2,208,245

Final cost [US$] 1,775,429 125,620 125,620 1,842,426 2,009,920 1,004,960 669,973 711,847 376,860 1,674,993 2,093,666 2,135,540 2,177,413 2,219,286

Step value [US$] 2,868,913 1,227,312 1,096,495 2,449,760 3,102,237 1,124,955 1,768,957 1,533,300 1,199,980 2,491,595 2,908,245 3,024,915 2,588,897 2,616,562

Action name IP IP PD PD IP CS OCL CS CS CS CS IM IP IP

Last probability [%] 0.215 0.110 0.099 0.054 0.042 0.039 0.011 0.110 0.046 0.031 0.024 0.004 0.001 0.215

Note: Inventory pooling (IP); Postponement delays (PD); Containerised shipment (CS); Other chain links (OCL); Implication monitoring (IM).

scenario as it may generate state and final costs more efficiently, particularly for entities such as maritime service providers. The cost reduction as a reward in applying CS may gain significant maximum step values gained by Australian shipping operators, Indonesian shipping and port operators, including Indonesian forwarders. The MDP also estimates that farmers and final consumers are the entities that may experience a higher likelihood of maritime disruptions of less than 22% compared to retailers (0.1%). 3.2 Simulation Results Reconfiguring other links in the supply chain is recommended to shift shipping services from dry bulk

operations to CS. This also consequently requires the use of a container terminal rather than a grain terminal. The outputs of probabilistic levels of each entity from the scenario assessment (see Fig. 6a to c) indicate that farmers and final consumers are entities with the same consequences at the end of the disruption period (for about 365 days). Another issue is that the higher probabilistic level of commercial and operational consequences due to maritime disruptions may severely impact Australian handlers and wheat processors and Indonesian shipping companies. It is estimated that retailers in Indonesia will have no commercial consequences because the risks affecting this entity are likely to be passed on to final consumers. In general, by applying multi-strategy

Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain

507


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

a)

b)

c) Fig. 6. The transition probabilities of one-year WSC; a) from day 0 to day 20, b) from initial period to day 300, c) from initial period to day 365

508

Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

scenarios, business entities on the maritime leg may be affected with relatively low levels of risk probabilities ranging from 3 to 5%; farmers and final consumers may experience maritime risk probabilities in the range of 21 to 22%. This is different from handlers and processors who may have 10 to 11% of maritime disruption risks. The outputs of TPs of the case are presented for 365 days. From day zero to the 20th day (as shown in Fig. 6a), farmers are estimated to have transition probabilistic levels ranging from 18 to 55% as an entity in the chain that may experience the highest level of risk consequences. Similarly, handlers and processors in the chain may achieve 18 to 25% probabilistic levels; processors in Australia and Indonesian shipping companies and wholesalers have a TP level of 0.4%. If the maritime disruption event is continuing, in the ranges of 21 to 40 and 41 to 60 days, the amplitude of TPs is decreased for all entities (except wholesalers, whose amplitude remains at 0.4%) to 18 to 21% for farmers, and 10% for handlers, processors and Indonesian shipping companies. After the period of the 60th day, the TPs pattern takes a similar configuration with a relatively constant level of values. Thus, TP outputs from days 341 to 365 have a constant level for all entities (as shown in Fig. 6c). Furthermore, in the constant level of TP, there are three groups of TPs. One is the group of two entities (i.e. farmers and final consumers) that may be clustered in a similar probability value. In practice, however, entities in the wheat supply chain have various barriers to implementing mitigation strategies due to financial and source limitations. In addition, it was found that entities with good coordination levels along the wheat supply chain may have a mitigation outcome due to its visibility and monitoring capacity in managing maritime disruptions. 4 CONCLUSIONS In this paper, the multi-mitigation strategy modelling of maritime disruptions in a WSC have been investigated. The level of service is proposed to represent the TP of satisfying the farmers and final consumers in the WSC, and it has been formulated as MDP. The application of MDM scenarios using the MDP for maritime disruptions may minimise the consequences of the risks in the containerised WSC. An experimental strategy has been carried out for the model assessment and impact analysis of the

confidence parameters in the containerised AustralianIndonesian WSC. The contribution of the paper is threefold. The first is a new approach that combines the advantages of MDP. An analytical model has been presented and the simulation based on implemented MDM strategies. The second and third contributions are a proposed model and suggestion that containerised WSC are able to generate competitive solutions quickly, even compared with traditional planning approaches that are much more time consuming. MDM represents not only mitigation responses but also a set of adaptation actions and feedback from the experiences of facing the disruptive events. It can be implemented to rectify both the consequences of disruption impacts and the probability of unwanted internal and external factors that may recurrently and severely change the stability of maritime services in the considered WSC. If the set of indicators could be applied to another territorial situation, the expression of their relevance will need to be discussed with stakeholders of this new region. It is in this context that exploring longer term macro-environmental events, such as the impact of climate change on the WSC may provide further insights into other related adaptation practices that may assist with strengthening the resilience of the WSC. As a recommendation, entities with better mitigation responses and coordination capacity may gain more significant results for their maritime disruption mitigation strategies. Therefore, a coordination factor should be included in the development of any future research model on maritime disruption. In regards to potential limitations in the study, the participants were self-selected, highly educated, mostly having a supply chain background, and reporting high levels of intentions toward wheat transport as well as the impact of maritime disruptions and its consequences in the wheat trade. It is possible that the managers who volunteered to participate in this research are not representative of the general population. The managers in this study are more likely to be concerned about detailed disruption management strategies that are more likely to be executed in relation to various maritime disruptive events. In any future research, the participation of executives in the real case modelling may contribute to further detailed implementation procedures of maritime disruption risk mitigations.

Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain

509


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 499-510

5 REFERENCES [1] Kennett, J., Fulton, M., Molder, P., Brooks, H. (1998). Supply chain management: The case of a UK baker preserving the identity of Canadian milling wheat. Supply Chain Management, vol. 3, no. 3, p. 157-166, DOI:10.1108/13598549810230912. [2] Young, L.M., Hobbs, J.E. (2002). Vertical linkages in agri-food supply chains: changing roles for producers, commodity groups, and government policy. Review of Agricultural Economics, vol. 24, no. 2, p. 428-441, DOI:10.1111/1467-9353.00107. [3] Wilson, W.W., Carlson, D.C.E., Dahl, B.L. (2004). Logistics and supply chain strategies in grain exporting. Agribusiness, vol. 20, no. 4, p. 449-465, DOI:10.1002/ agr.20026. [4] Hachicha, W. (2011). A simulation metamodelling based neural networks for lot-sizing problem in MTO sector. International Journal of Simulation Modelling, vol. 10, no. 4, p. 191-203, DOI:10.2507/ IJSIMM10(4)3.188. [5] Fain, N., Kline, M., Duhovnik, J. (2011). Integrating R&D and marketing in new product development. Strojniški vestnik - Journal of Mechanical Engineering, vol. 57, no. 7-8, p. 599-609, DOI:10.5545/svjme.2011.004. [6] Kaljun, J., Dolsak, B. (2012). Improving products’ ergonomic value using intelligent decision support system. Strojniški vestnik - Journal of Mechanical Engineering, vol. 58, no. 4, p. 271-280, DOI:10.5545/ sv-jme.2011.193. [7] Dalfard, V.M., Ranjbar, V. (2012). Multi-projects scheduling with resource constraints and priority rules by the use of simulated annealing algorithm. Technical Gazette, vol. 19, no. 3, p. 493-499. [8] Yu, G., Qi, X. (2004). Disruption Management. World Scientific Publishing Co. Pte. Ltd, Singapore. [9] Craighead, C.W., Blackhurst, J., Rungtusanatham, M.J., Handfield, R.B. (2007). The severity of supply chain disruptions: design characteristics and mitigation capabilities. Decision Sciences, vol. 38, no. 1, p. 131156, DOI: 10.1111/j.1540-5915.2007.00151.x. [10] Paul, J.A., Maloni, M. (2010). Modeling the effects of port disasters. Maritime Economics and Logistics, vol. 12, no. 2, p. 127-146, DOI:10.1057/mel.2010.2. [11] Merrick, J.R.W., Van-Dorp, J.R., Mazzuchi, T., Har, J.R. (2002). The Prince William sound risk assessment. Interfaces, vol. 32, no. 6, p. 25-40, DOI:10.1287/ inte.32.6.25.6474. [12] Gurning, S., Cahoon, S. (2011). Analysis of multimitigation scenarios on maritime disruptions. Maritime Policy & Management, vol. 38, no. 3, p. 251-268, DOI: 10.1080/03088839.2011.572701. [13] Gurning, S., Cahoon, S., Nguyen, H.O., Achmadi, T. (2011). Mitigating maritime disruptions: Evidence

510

from the Australian-Indonesian wheat supply chain. International Journal of Shipping and Transport Logistics, vol. 3, no. 4, p. 406-429, DOI:10.1504/ IJSTL.2011.041135. [14] Tomlin, B. (2009). Disruption-management strategies for short life-cycle products. Naval Research Logistics, vol. 56, no. 4, p. 318-347, DOI:10.1002/nav.20344. [15] Kleindorfer, P.R., Saad, G.H. (2005) Managing disruption risks in supply chains. Production & Operations Management, vol. 14, no. 1, p. 53-68, DOI:10.1111/j.1937-5956.2005.tb00009.x. [16] Tang, C.S. (2006). Robust strategies for mitigating supply chain disruptions. International Journal of Logistics Research and Applications, vol. 9, no. 1, p. 33-45, DOI:10.1080/13675560500405584. [17] Hou, J., Zeng, A.Z., Zhao, L. (2010). Coordination with backup supplier through buy-back contract under supply disruption. Transportation Research E, vol. 46, no. 6, p. 881-895, DOI:10.1016/j.tre.2010.03.004. [18] Phillips, P.W.B., Smyth, S. (2007). Grounding the management of liabilities in the risk analysis framework. Bulletin of Science Technology Society, vol. 27, no. 4, p. 274-285, DOI:10.1177/0270467607300639. [19] Skelton, P. (2007). Business continuity and supply chain management: How to manage logistical operations in the event of an interruption or emergency. Journal of Business Continuity & Emergency Planning, vol. 2, no. 1, p. 13-20. [20] Haque, C.E., Burton, I. (2004). Adaptation options strategies for hazards and vulnerability mitigation: An international perspective. Mitigation and Adaptation Strategies for Global Change, vol. 10, no. 3, p. 335353, DOI:10.1007/1-4020-4514-X_1. [21] Howick, S., Eden, C. (2001). The impact of disruption and delay when compressing large projects: Going for incentives?. The Journal of the Operational Research Society, vol. 52, no. 1, p. 26-34. [22] Carpignano, A., Golia, E., Di Mauro, C., Bouchon, S., Nordvik, J.P. (2009). A methodological approach for the definition of multi-risk maps at regional level: First application. Journal of Risk Research, vol. 12, no. 3-4, p. 513-534, DOI:10.1080/13669870903050269. [23] Handfield, R.B., McCormack, K. (eds.) (2008). Supply Chain Risk Management: Minimizing Disruptions in Global Sourcing. Auerbach Publications, New York. [24] Gurning, S. (2011). Maritime disruptions in the Australian-Indonesian wheat supply chain: An analysis of risk assessment and mitigation strategies. PhD thesis. Australian Maritime College, University of Tasmania, Launceston. [25] McKelvey, B., Andriani, P. (2010). Avoiding extreme risk before it occurs: A complexity science approach to incubation. Risk Management, vol. 12, no. 1, p. 54-82, DOI:10.1057/rm.2009.14.

Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 511-516 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2013.947 Special Issue, Original Scientific Paper

Received for review: 2013-01-02 Received revised form: 2013-06-06 Accepted for publication: 2013-06-12

Freight-Forward Agreement Time series Modelling Based on Artificial Neural Network Models

Lyridis, D. – Zacharioudakis, P. – Iordanis, S. – Daleziou, S. Dimitrios Lyridis1,* – Panayotis Zacharioudakis1 – Stylianos Iordanis1 – Sophia Daleziou2 2 National

1 National Technical University Athens, Greece Technical University of Athens, School of Applied Mathematical and Physical Sciences, Greece

Over the last thirty years, there has been an extraordinary growth in the financial derivatives market, in the field of shipping. This can be attributed to the fact that financial derivatives are contracts that allow all players participating in the shipping market to reduce their exposure to fluctuations in freight rates, bunker prices, interest rates, foreign exchange rates and vessel values. This paper employs an artificial neural network (ANN) in order to forecast the future price of freight derivatives. More specifically, drawing on historical data for the period between January 2005 and March 2009, an ANN is built and trained, and its estimates lead to two individual results. The resulting model indicates to the investor which position to take in the derivatives market (short for sale of agreements and long for the purchase of agreements). Keywords: freight rates, trading strategy, artificial neural networks, shipping market modelling, freight rate forecasting

0 INTRODUCTION Artificial neural networks are a technology that has been adopted in many disciplines, including neuroscience, mathematics, statistics, physics, computer science and engineering. Their ability to learn from data has endowed them with powerful properties and has made them invaluable tools in financial forecasting. This is a challenge for researchers worldwide: much effort has been made using ANNs to design a trading system for profitable movements in the market. However, few people would dare invest their money according to the forecasts of a neural network that has been trained using historical data, as it is widely believed that is impossible to predict the dynamics of an economy [1]. Here, it should be noted that any attempt to create a neural network about financial variables will necessarily draw on historical time series data. The main purpose of this process is to draw conclusions about the future evolution of the variable being studied, and the most ambitious goal is to detect predictable patterns for future values. An important advantage offered by ANNs is that they can constantly enrich their knowledge with new information resulting from new market conditions. The decisive factor is the volatility of the market being studied, as the bigger the market, the shorter the period during which the new data will be valid. In other words, the success rates of an ANN forecasting market trends for a specific time period are inversely analogous to the market’s growth rates. However, training an ANN with the use of time series data has been proven to be successful in such applications; therefore, it may be possible to build

a system to successfully forecast the evolution of economic indicators. Extensive literature in the shipping market and shipping derivatives [2] to [4] describe in detail the statistical and stochastic background of shipping derivatives, along with their uses regarding their pricing, and time series properties of the underlying dynamics governing the fluctuations of freight rates. Risk management, i.e. hedging, may be the primary theoretical use of freight derivatives, and this topic is the main focus of the related bibliography. However, speculation, i.e. directional trading, is the most popular application of shipping and other derivatives; the related academic research and literature is, however, not as rich. Li and Parsons [5] were the first to attempt applying neural network modelling within the context of crude oil freight rate prediction for the Mediterranean line (Med-Med) using data from 1980 to 1995 and three variables: the actual rate time series, demand index for tankers and Drewry’s total active tanker capacity. To compare the results, the researchers also developed two parallel auto regressive moving average (ARMA) models. The study showed that the prediction with neural networks proved much more capable of dealing with the excessive discontinuities of the time series. The ANN performance was better than the ARMA models in all cases, and this was even clearer for longer-term forecasts. Alizadeh and Nomikos [3] also present a technical analysis framework for freight trading strategies based on technical trading rules using moving average (MA) crossover divergence / stochastic oscillators, moving average envelopes, Bollinger bands, momentum

*Corr. Author’s Address: National Technical University Athens, Laboratory for Maritime Transport, dsvlr@mail.ntua,gr

511


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 511-516

trading model, or spread trading in freight forward agreement markets. However, if we compare the two approaches, neural networks are a vastly more powerful methodology than technical heuristics, as they incorporate artificial intelligence potential and universal function approximator properties. The following study focuses on freight futures, in terms of the type of the derivative product. It is based on three-month agreements for a specific dry freight transportation route issued by IMAREX, according to the Baltic Exchange indexes. Hence, the system to be designed should forecast fluctuations in the prices of freight futures, in relation to changes in freight rates at the spot market, and thus create forecasts that lead to profitable investment actions. Lastly, it will be possible to acquire new forecast results generated on a daily basis, after the necessary data, similar to those used to train the network, have been entered. The rest of the paper is organized as follows: The role of derivatives in the shipping market is presented in Section 1. Section 2 provides the criteria for selecting the data, the implementation of the methodology and the results. In the last section, the conclusions of the paper are discussed. 1 DERIVATIVES AND SHIPPING Freight derivatives or freight forward agreements (FFAs) were developed. aiming at managing (in a more efficient, cost-effective and flexible manner) risks resulting from fluctuations in freight rates, the cost of storage, ship prices, scrap prices, interest rates, and foreign exchange rates [2]. FFAs can be defined as future agreements whereby the two parties undertake to buy or sell the transportation of bulk cargo from one location to another at a price established at the time of closing the agreement. The settlement is in cash, so there is no physical delivery. Payment is made on a specific date agreed upon by the two parties. For the settlement of FFAs, the London Baltic Exchange Baltic Dry Index (BDI) and Baltic International Tanker Routes (BITR) are taken into account. The first FFAs appeared in 1985, when the Baltic International Freight Futures Exchange (BIFFEX) agreement was created. It was an agreement with underlying average index the Baltic Freight Index (BFI) of the Baltic Exchange. Following that, and due to market segmentation, many individual BFIs made 512

their appearance. They made their first appearance as private agreements between two parties traded over the counter in 1992. The increased capital liquidity prevailing in the shipping market, the ease of creating standardized agreements offered by sea transportation (regarding bulk cargos, routes and ship sizes), the fact that the shipping market is subject to a common valuation structure / methodology, and generally increased transparency are all factors that have contributed to the success of derivatives in shipping. Such products are already traded over the counter (OTC) at the clearing houses of London and Oslo. Shipping companies, as well as energy companies, use freight derivatives in order to cope with changes in freight rates and these contracts are now considered to be the most rapidly growing area of the shipping industry. FFAs, which should be mentioned for the understanding of this paper, fall under the following categories: Freight Forward Agreements: Private agreements entered into mainly by and between ship owners and shippers, and aim at hedging against the volatility of the freight market. In this case, the ship owner takes a short position and sells FFAs, whereas the shipper is the buyer (long position). Freight Futures: The principle behind this instrument is the same as above; however, they are traded in organized markets, such as IMAREX and NYMEX. Hence, like all derivatives, FFAs are also used for hedging, speculative and arbitrage purposes. 2 METHODOLOGY AND RESULTS The artificial neural network to be built will be based on Dry Bulk Routes. Table 1 shows the routes of IMAREX for the issue of dry bulk routes, and the route basket selected for use in the research part of this paper (Routes are grouped in baskets based on the type of the ship to which each route refers. Hence, the names of the BCI, BPI, BHI and BSI indexes come from Capesize, Panamax, Handymax and Supramax, respectively). Data about quarterly freight futures prices and the daily spot prices for the specific route basket have been collected. Such figures are given in time charter (T/C) in $/day. The data range is the period between 4 January 2005 and 13 March 2009. For each day, we have: • The asset spot price (freight), • The asset future price for the next quarter.

Lyridis, D. – Zacharioudakis, P. – Iordanis, S. – Daleziou, S.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 511-516

Table 1. IMAREX Dry Bulk Routes IMAREX listed single route and basket of T/C dry bulk futures Cargo Size Routes Sector Route Description (tons) C4 Capesize Richards Bay – Rotterdam 150000 C7 Capesize Bolivar – Rotterdam 150000 P2A Panamax T/C Skaw Gibraltar – Far East 74000 T/C South Korea – Japan P3A Panamax 74000 Pacific R/V Panel B: listed basket of T/C dry-bulk futures CS4TC Capesize T/C Average PM4TC Panamax T/C Average HM6TC Handymax T/C Average

n/a n/a n/a

For each of the days in the database, the investor’s profit or loss is calculated as follows: • If Χ denotes the trading day and trade price (TP) the price of the three-month future agreement on day Χ; • If Avg1, Avg2, Avg3 denote the average of the spot prices for the 1st, 2nd, and 3rd month respectively; • If the investor takes a short position (by selling FFAs). Then, the trading signal function is defined as:

more simply their correlation with the desired variable (profit). The correlation coefficient is used to measure the strength of the relationship between two variables x, y. Given n pairs of observations (xi,yi) we can compute the correlation coefficient r as:

sxy sx ⋅ s y

,

where sxy = Cov (x,y), the covariance between the variables x, y, sx, sy is the standard deviation. As a result, n

n

sxy =

∑ ( x − x) ⋅ ( y − y ) ∑ x y − n ⋅ x ⋅ y i =1

i

i

=

n −1

i

i =1

i

n −1

,

where

sx =

1 n ( xi − x) 2 , ∑ n − 1 i =1

sy =

1 n ∑ ( yi − y)2 . n − 1 i =1

and

Profit = (TP – Avg1)×31 + (TP – Avg2)×30 + + (TP – Avg3)×31.

If Profit > 0, taking a short position is a correct trading signal. If Profit < 0, taking a long position is a correct trading signal. For the ANN to be trained, the largest possible number of time series is needed, which will be directly or indirectly correlated to freight rate fluctuations being the input variables, whereas profit is the output (desired) variable. These time series as well as all data used in this paper were taken from Clarkson Research Services [6]. Initially, the time series data were monthly, weekly or daily, depending on their fluctuation rate. The first processing of such time series involved the reduction to daily time series as a whole by linear interpolation of new data. Such conversion is necessary, as the desired results are given for each day of operation of the stock market, and in order to train the ANN with the same input data. Forty-seven time series were taken from Clarkson Research Services. Out of these, only six were selected to train the ANN, otherwise the problem would have been exceedingly complex. Hence, a decisive factor in the selection of the suitable time series to be used to train the system is the correlation coefficient, or

r=

Finally, n

r=

sxy

∑ ( x − x) ⋅ ( y − y )

=

sx ⋅ s y

i =1

i

n

∑ (x − x ) i

i =1

2

=

n

∑(y − y )

i =1

i

2

n

=

∑x y − n⋅ x⋅ y i =1

n

∑x i =1

i

i

i

2 2

− n⋅ x ⋅

n

∑y i =1

i

2

. − n⋅ y

2

The correlation coefficient lies between –1 and +1. r > 0 indicates a positive linear relationship between the two variables and r < 0 a negative linear relationship between x, y. r = 0 indicates no linear relationship between x, y. Lastly, the prerequisites that need to be met by the final time series to be used in training the network are summarized as: • Coefficient of correlation of the ΤS time series with the output time series > 0.20. • Coefficient of correlation among ΤSi time series < 0.80. The rationale behind these two heuristics is that we keep mathematically irrelevant candidate

Freight-Forward Agreement Time series Modelling Based on Artificial Neural Network Models

513


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 511-516

explanatory variables out of the model, i.e. when the coefficient of correlation with the target variable is too low, but we also exclude any highly correlated explanatory variables in order to eliminate colinearity issues in the modelling process. The time series considered to be the most important are: • capesize bulker orderbook number, • capesize bulk carrier fleet development, • capesize scrap value, • one-year time charter rate 170,000 dwt bulk carrier, • capesize bulker contracting number, • capesize sales, • capesize bulk carrier deliveries number, • capesize bulk carrier new building prices. Hence, after a first filtering of the data using calculations for their correlation with desired output, these are next tested in various combinations as to their contribution in the performance of a neural network. The first step is then to create the network. The network was created using the software package NeuroSolutions v.5, which allows entering data directly from Excel worksheets without particular processing, and, of course, aids in the design of an ANN [7].

Fig. 2. The success rates of the four modular topologies

Fig. 3. The MSE values of the four modular topologies

Fig. 4. Network’s Topology

Fig. 1. The four alternative ANN possible feed-forward modular connection topologies in the case of 2 hidden layers

This paper employs modular neural networks. These networks are a special class of multi-layer perceptron (MLP) feed-forward artificial neural network model corresponding to input data maps. These networks process their input using several parallel MLPs, and then recombine the results. Such networks are more suitable for time series prediction of non-linear economic fundamentals. A more technical discussion on ANNs can be found in [8] and 514

Fig. 5. The success rates using 1,2,3 and 4 hidden layers

Fig. 6. The MSE values using 1,2,3 and 4 hidden layers

Lyridis, D. – Zacharioudakis, P. – Iordanis, S. – Daleziou, S.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 511-516

[9]. More recent application of ANN can also be found in [10] and [11] with applications in many disciplines. There are four modular topologies concerning the interaction between the layers, as shown in Fig. 1. Experimentations were carried out with all potential combinations of the above time series and modular topologies. The correlation coefficient between the employed time series and the output data is shown in Table 2. The initial step for designing any neural network is to collect the training data; the selection of testing and cross validation data follows. The testing set is used to test the performance of the network, and cross validation is used as a highly recommended method for stopping the network training. This method monitors the error of the independent set of data and stops training when this error begins to increase. This is considered to be the optimal point for generalization. The appreciation of a network is not based only on its success rates but also on the mean square error 1 N (MSE) of the results. The MSE ( MSE = ∑ et2 ) N t =1 of a prediction is one of the ways of quantifying the difference between the predicted and the true value of the variable being estimated [12]. Based on the above, four modular topologies were initially used with a default number of layers (two), in order to find which of these topologies have the highest success rates and the minimum error. The results are illustrated in Figs. 2 and 3.

It is obvious that the second network provides the best results with the minimum error. The topology of this network is given in Fig. 4. The results of this specific topology were tested using different number of hidden layers (1, 2, 3, 4) as set out in Figs. 5 and 6. It is the network with the four hidden layers that seems to be the most successful. The graphic representation of the final network results is given in Figs. 7 and 8. The 85% success rate of the testing set refers to the percentage at which the two lines are above or below the x-axis at the same time. 3 CONCLUSION It is a fact that neural networks employed for forecasting the evolution of freight derivatives can become an invaluable tool, capable of leading to successful investments, as also explained in the literature (for additional cases and discussion, please also see [13] and [14]). However, the forecast must always play an auxiliary part and be used carefully. It should not be forgotten that such models are actually self-learned, based on the past in order to forecast the future, which could hide unexpected events. Ideally, their use is recommended in conjunction with empirical knowledge and human judgment, especially during periods when fundamental structural breaks take place. The quantitative findings of the models developed above have shown that the applied Connectionist models fit well to the underlying dynamics of the time

Table 2. Correlation between the input and output variables TS r

FUTURE –0.380

SPOT –0.406

TS1 –0.215

TS2 –0.264

TS3 0.430

TS4 0.318

TS5 –0.168

TS6 –0.550

Fig. 7. The network output (in $) of the training set; desired output and actual network output Freight-Forward Agreement Time series Modelling Based on Artificial Neural Network Models

515


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 511-516

Fig. 8. The network outputs of the test set

series, yielding satisfactory accuracy capable of a high success rate when implementing the models within a trading strategy context. The methodology can easily be expanded to multivariate models forecasting vector time series, i.e. it can be generalized to model multiple system variables of the shipping market and also be adapted to forecast the price and time of the market for other financial instruments. MLP architecture is definitely qualified to handle any such task if applied with the appropriate level of complexity regarding the number of layers and given sufficient computational power, and if we attempt to enhance the model with more exogenous explanatory variables and dependent variables. The whole model architecture will be tested in future research for modelling composite indices consisting of equally weighted equity price time series of returns of various shipping companies. Possible practical applications can be found within the context of commodity trading advisor (CTA) funds, commodity trading and shipping companies; companies seeking to outperform classic heuristic technical methods as practised by the majority of market participants, given that there is currently an “arms race” in sophistication of quantitative tools of traders. 4 REFERENCES [1] Haykin, S. (1999). Neural Networks: A Comprehensive Foundation. 2nd ed. Prentice Hall, Upper Saddle River. [2] Hunt, P.J., Kennedy, J.E. (2004). Financial Derivatives in Theory and Practice. Revised ed. John Wiley & Sons, Chichester, DOI:10.1002/0470863617.fmatter. [3] Alizadeh, A.H., Nomikos, N.K. (2009). Shipping derivatives and risk management. Palgrave Macmillan, New York, DOI:10.1057/9780230235809. [4] Helyette, G. (ed.) (2008). Risk Management in Commodity Markets: From Shipping to Agriculturals

516

and Energy. John Wiley & Sons Inc., Hoboken Clarkson Research Services, from: http://www.crsl. com, accessed on 2012-12-1. [5] Li, J., Parsons, M.J. (1997). Forecasting tanker freight rates using neural networks. Maritime Policy & Management, vol. 24, no. 1, p. 9-30, DOI:10.1080/03088839700000053. [6] Principe, C.J., Euliano, R.N., Lefebvre, W.C. (2000). Neural and Adaptive Systems: Fundamentals through Simulations. Wiley & Sons Inc., Chichester. [7] Uran, S., Safaric, R. (2012). Neural-network estimation of the variable plant for adaptive sliding-mode controller. Strojniški vestnik - Journal of Mechanical Engineering, vol. 58, no. 2, p. 93-101, DOI:10.5545/ sv-jme.2011.098. [8] Florjanic, B., Govekar, E., Kuzman, K. (2012). Neural network-based model for supporting the expert driven project estimation process in mold manufacturing. Strojniški vestnik - Journal of Mechanical Engineering, vol. 59, no. 1, p. 3-13, DOI:10.5545/sv-jme.2012.747. [9] Li, J., Burke, E.K., Qu, R. (2011). Integrating neural networks and logistic regression to underpin hyperheuristic search. Knowledge-Based Systems, vol. 24, no. 2, p. 322-330, DOI:10.1016/j.knosys.2010.10.004. [10] Casta-o, A., Fernández-Navarro, F., Gutiérrez, P.A., Hervás-Martínez, C. (2012). Permanent disability classification by combining evolutionary Generalized Radial Basis Function and logistic regression methods. Expert Systems with Applications, vol. 39, no. 9, p. 8350-8355, DOI:10.1016/j.eswa.2012.01.186. [11] Kleinbaum, D.G., Kupper, L.L., Nizam, A., Muller, K.E. (2007). Applied Regression Analysis and Other Multivariate Methods. 4th ed. Thomson Learning, Inc., Belmont. [12] Marose, R.A. (1990). A financial neural network application. AI Expert, vol. 5, p. 50-53. [13] White, H. (1988). Economic prediction using neural networks: The case of IBM daily stock returns. IEEE International Conference on Neural Networks, vol. 2, p. 451-458, DOI:10.1109/ICNN.1988.23959.

Lyridis, D. – Zacharioudakis, P. – Iordanis, S. – Daleziou, S.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2013.963 Special Issue, Original Scientific Paper

Received for review: 2013-01-09 Received revised form: 2013-06-05 Accepted for publication: 2013-07-05

Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation Longo, F. – Huerta, A. – Nicoletti, L. Francesco Longo1,* – Aida Huerta2 – Letizia Nicoletti1

1 University

of Calabria, Department of Mechanical, Energetic and Management Engineering, Italy University of Mexico, Department of Systems Engineering, Mexico

2 National

Modeling & Simulation (M&S) has proved to be a day-to-day highly indispensable tool for complex systems design, management and monitoring. Therefore, the proposed research study aims to develop a simulation model to recreate the complexity of a medium-sized Mediterranean seaport and analyse the performance evolution of such system with particular reference to the ship turnaround time. After the input data analysis, simulation model development, verification and validation, a design of experiments (according to a 24 factorial experimental design) was carried out in order to evaluate how some critical factors (i.e. inter-arrival times, loading/unloading times, number of cars and trucks to be unloaded/loaded) may affect the seaport’s performance. To this end an analysis of variance is performed and an analytical input-output meta-model was created to evaluate the system’s performance. Keywords: logistics, marine ports, supply chain, modeling & simulation

0 INTRODUCTION Despite the economic and financial crisis, statistics show that merchant fleets are growing: more than 138,000 seagoing commercial ships are currently in service [1]. These data prove the crucial role of seaports for both national and international trade. In this framework, the Mediterranean Sea represents one of the most strategic areas with 15% of global shipping activities taking place at mainly western and central Mediterranean ports [2]. Therefore decision making in seaports requires the support of powerful tools allowing performance measurement and analysis. To this end, the main goal of this research work is to propose a simulation-based tool of a mediumsize Mediterranean seaport that could be used by the main port administrators (i.e. port managers, the port authority, etc.) to support decision making and process management. Indeed, all port stakeholders need to monitor their performance taking into account quantitative and qualitative aspects, determining whether their strategies produce the desired outcomes, and correcting any misallocations and malfunctions [3]. Performance measurements can be considered in terms of three kinds of indicators: key result indicators (KRIs), performance indicators (PIs), and key performance indicators (KPIs) [4]. As reported in [3], port performance measurements – usually recommended to the port community stakeholders – are divided into five categories: market trends and structure; socio-economic impact; environmental performance; logistic chains; and operational performance and governance. In such a context, modeling and simulation (M&S) has proved to be a

valuable methodology for performance assessment, as well as enhancement and monitoring activities in complex systems. Indeed its first applications in industry and logistics dates back to 1980 [5] to [7]. Among others, discrete-event simulation (DES) is a leading simulation paradigm used to study operational and planning processes within domains like industry [8], logistics and supply chains and more specifically within seaports. In effect, DES is able to capture and recreate the highly dynamic evolutionary processes that are typical of complex real systems. A simulation model of an inland port can be found in [9], whereas [10] and [11] propose simulation models devoted to evaluating logistical and operational processes in marine terminals. Considering the port terminal capacity, [12] and [13] propose simulation models to investigate the capacity increment that can be achieved through new management strategies applied to the devices and equipment available, while avoiding additional capital costs. With reference to security issues in marine ports, namely inspection procedures within container terminals, it has been proved that simulation can be an effective tool for supporting decision, which can be easily integrated into the day-to-day container terminal operations [14] and [15]. Furthermore, in this field, simulation has been successfully applied in conjunction with artificial intelligence techniques for systems performance optimization [16] to [19]. Although it has been proven that M&S is able to support seaport management at various levels (even when combined with agent based approaches [20] such as multiagent systems [21] or when the simulation model is designed to reproduce the microscopic, stochastic, real-time environment of a part of the container

*Corr. Author’s Address: University of Calabria, Via P. Bucci, Cube 44C, 87036 Rende (CS), Italy, f.longo@unical.it

517


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

terminal i.e. a multiple-berth facility [22]), it is also important to consider that decision-making processes imply different planning horizons. To this end, [23] this study develops different microscopic DES models for a container terminal, focusing on the best approach to simulate container handling activities and their time duration. In addition, it is worth mentioning that when applied to marine ports M&S is also a very powerful methodology for supporting advanced and cooperative training, exercise and education [24] and [25]. The above brief state of the art overview shows that M&S can provide valuable support within seaports. Therefore the main aim of this research work is to develop an advanced simulation model able to recreate the main processes and activities of a medium size Mediterranean seaport in order to analyse the evolution of the turnaround time of certain types of ships (i.e. Roll-on Roll-off, Ro-Ro ships) when the ships inter arrival time, the duration of the loading/ unloading activities and the number of cars and trucks loaded/unloaded change. The paper is structured as follows: Section 1 presents the simulation methodology, Section 2 describes the simulation model development, Section 3 deals with the validation of the simulation model, Sections 4 and 5 propose the design of experiments and the simulation results analysis, while the last section summarizes the conclusions. 1 SIMULATION METHODOLOGY As stated before, DES is one of the most common simulation paradigms used to investigate real complex systems. Through DES, in fact, it is possible to model the system behaviour over time as a series of events that change the system status [25]. Therefore the methodological approach applied in this research work relies on M&S fundamentals in order to explore the operating procedures in a seaport, as well as to obtain a greater understanding and to analyse their performance evolution over the time. To this end, the key processes of this simulation study comply with the approach well-established in the literature (see [26] to [29]): • objectives and overall project plan setting; • model conceptualization; • data collection and input data analysis; • simulation model development); • simulation model validation and optimal run length; • design of experiments; • simulation results analysis. 518

1.1 Port Conceptual Model It is widely recognized that conceptual models are used to document those aspects of a real system that need to be represented in computerized models as well as the ones that need to be omitted. The system studied in this research work is a medium-sized Mediterranean seaport (the port of Salerno) with a strategic logistic position in the middle of the Mediterranean sea and a primary role in trade exchanges involving southern Italy. Its strategic position led us to consider this seaport as one of the main Italian ports of call for the ‘motorways of the Sea’ that are part of the transEuropean transport network (TEN-T) and Marco Polo II program. In this study, the commercial seaport has been considered as a system where seven physical zones are distinguished: the West Pier, the Red Quay, the Trapezio Pier, the Ligea Quay, the 3 January Pier, the Manfredi Pier, and the Levanter Pier; the layout of the seaport is shown in Fig. 1. The seaport offers regular Ro-Ro and passengers (RoRo/Pax) connections to the ports of Valencia, Malta, Tunis, Messina, Palermo, Tripoli, Termini Imerese, and Cagliari. The Ro-Ro/Pax macro-activities, the container handling operations, ferry and merchandise operations have been included in the conceptual model. Some of the activities’ conceptual flows are depicted in Figs. 2 and 3. After arrival the ship waits outside the port until a berth position becomes available; mooring operations are in many cases performed with the help of tugboats (above all for large vessels that cannot use the side thrusters close to the berth). After unloading/loading operations the vessels leave the port. It is important to outline that due to the physical size of the seaport, ships can only enter or exit the seaport one by one. In addition, exit has priority over entry.

Fig. 1. Seaport layout

1.2 Data Collection and Input Data Analysis Data collection plays a crucial role in simulation studies and affects the development and use of

Longo, F. – Huerta, A. – Nicoletti, L.


StrojniĹĄki vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

Fig. 2. Flow chart of the arrival/departure process

It is worth stating that the random behaviour of some variables makes a seaport a stochastic system. As reported in [26], for each element in a system being modelled, the simulation analyst must decide on a way to represent the associated variables. The data collection step takes care of collecting data for all port processes and activities as well as finding the most suitable computer representation for such data. Usually there are three different choices: (i) data are deterministic or data are considered to be deterministic, (ii) a distribution probability is fitted to the empirical data and (iii) the empirical distribution of the data is directly used in the simulation model [30] to [31]. In our case the second and the third choices have been used. For the purposes of this study, data on actual ship arrivals from January, 1st, 2010 to December, 31st, 2011 and from January 1th, 2012 to May 14th, 2012 were collected and used in the simulation model. In the case of stochastic variables and distribution fitting, the procedure for input data analysis is the classical procedure proposed by many statistics handbooks as well as implemented in numerous commercial software applications: (i) starting from a histogram of the data, one or more candidate distributions are hypothesized, (ii) for each distribution the characterizing parameters are estimated, (iii) a goodness of fit test is performed, and (iv) lastly, the best distribution is chosen. For any additional information on input data analysis for simulation studies please refer to [26] and [33]. The K intervals for each input parameter of the model were obtained taking into consideration equal-width intervals and using Scottsâ&#x20AC;&#x2122; formula [32] (see Eq. (1)).

K = 1.15 N1/3 , N â&#x2030;Ľ 25 . (1)

Then, the N data collected were fitted into the probability density functions based on the K intervals. As an example, the simulation input values for the RoRo/Pax ships is presented in Fig. 4.

Fig. 3. Flow chart of the quay crane allocation process

simulation models. The accuracy of the simulation is affected by the quality of the input data, which is why special attention should be paid to this step. In effect, if the data used to design and populate the model are inaccurate, the results of the model will be inaccurate as well.

Fig. 4. Interarrival time of Ro-Ro/Pax ships

Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation

519


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

2 SIMULATION MODEL DEVELOPMENT The translation of the conceptual model into a computerized model is one of the key steps in a simulation study. In this research work the simulation model was implemented using the Anylogic® (6.4) simulation software. It is a Java-based and general purpose simulation tool that takes advantage of the power of Java in any part of the model or in any library. These features make Anylogic® a suitable tool for simulating complex systems such as seaports. Furthermore it allows the simulation of various domains using different approaches, and also provides animation, which can be useful in supporting the decision-making processes of seaport planners. The authors have extensive experience in developing simulation models in the field of supply chains [34] as well as in seaports and container terminals [14]. Therefore the approach used for the development of the simulation model proposed in this paper follows the same logics and rules already used in other research works by the same authors. For the sake of clarity a description of the port simulation model development is reported in the remaining part of this section (according to the same structure used in [14]).

The container terminal simulation model is in four parts: (i) the flow chart that recreates the main port activities; (ii) the transportation networks that allows the recreation of entities and resources movements within the port area; (iii) the graphic user interface and output section for scenarios definition and performance measures monitoring, respectively; (iv) the animation main frame recreating a 2D animation of all the port operations including vessel arrivals and departures, as well as vessel unloading and loading operations. Basically, at the beginning of the flow chart there are different source objects used to generate different types of vessels. Each vessel is then re-directed into one of the flow chart branches according to the type of vessel and berth position. In each branch the following operations are simulated: mooring operations, unloading and loading operations, and detachment operations according to the resources available. Finally after unloading and loading operations, vessel detachment operations are performed and the vessel can leave the port area. As part of the flow chart, the terminal resources used for executing the operations described above are of three different types: autonomous resources (i.e. forklift, quay cranes, tugboat, trucks etc.), support resources

Fig. 5. Simulation model animation

520

Longo, F. – Huerta, A. – Nicoletti, L.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

(i.e. a chassis for a truck, a spreader for a forklift, etc.), or static resources (i.e. a quay crane position, a parking position, etc.). Transportation networks (built on the port layout) are used to handle entities and resources movements while simultaneously defining the animation main frame of the simulation model. In effect, defining these networks requires inserting class objects both into the simulation model flow chart and into the simulation model animation. Within the simulation model flow chart, a network is made up of several objects that allow entry into the network, taking resources from the network, moving within the network, freeing the resources, and exiting the network. For example, consider the case of a forklift that is supposed to move a container from the quay crane area to the yard area. The container enters the network, takes the resource forklift, waits until the resource forklift becomes available (then the forklift performs the movement from the quay crane area to the yard area), frees the resource forklift, and, if needed, leaves the network. Within the simulation model animation a network is made up of rectangles and lines (respectively, resource locations and trajectories between locations). A rectangle represents a network entry or exit point, the idle position for some resource, or a destination point in the port. A line is the path followed by an entity moving among rectangles. For example, consider again the case of the container to be moved from the quay crane area to the yard area. During the animation, the container appears in a rectangle located in the quay crane area, it waits there until the forklift becomes available and then the forklift moves the container to the yard area by following the path specified by the lines connecting the quay crane area rectangles and the yard area rectangles. Fig. 5 shows the final animation of the simulation model based on the transportation network. In order to develop an advanced interactive tool for testing scenarios, the most important variables, analysis and problems solving, are defined as parameters; each parameter has specific range values and statistical distribution forms and its values can be changed by using a dedicated graphic user interface. Similarly a dedicated output section shows the results of the simulations, i.e. the simulation model including information (for each vessel type) about the actual port traffic, the number of ships as a function of the arrival time, the empirical distributions (as histograms) for the average waiting time for a berth position, the average waiting time for mooring operations, the average service time, and the average turn-around time.

3 SIMULATION MODEL VALIDATION AND OPTIMAL RUN LENGTH The simulation model validation aims at establishing whether the simulation model is providing valid and reliable outputs (close to the real system outputs). In addition, a seaport is a non-terminating system; this implies that the duration of a simulation run is not a-priori fixed therefore the optimal length of a simulation run has to be defined. For this purpose a mean square pure error analysis (MSpE) was applied. The MSpE is a characteristic of the simulation model; it is related to the overall stochasticity of the real system that is represented within the simulation model [35]. The MSpE allows the simulation run length to be chosen because it is an unbiased estimator of the error affecting the simulation model results. One should note that the MSpE is an intrinsic characteristic of the simulation model, therefore in this case it has been evaluated in relation to the service time of the ships arriving at the seaport. This simulation output has been analysed based on confidence intervals rather than point estimators, thus providing a realistic analytical framework (a confidence interval with a confidence level of 99.75% has been calculated for the service time but also for an additional performance measure, the number of ships per week). The MSpE evolution for the Ro-Ro/Pax ships arriving at the seaport is a suitable knee curve, as illustrated in Fig. 6 and the values for the confidence intervals (in terms of the lower confidence limit (LCL) and the upper confidence limit (UCL)) are summarized in Table 1. After a simulation time of 60 weeks, the value of the MSpE was quite small. At that time, the effect of the noise produced by a casual overlapping of the probability density functions related to the input parameters on the simulation results was minimized considerably. In addition, the comparisons between the real values and the confidence intervals calculated by the simulation model suggest that the simulation model is able to recreate the real system with satisfactory accuracy. Under these circumstances the DES model can be considered a valid support for decision-making processes. Table 1. Confidence interval (99.75%) Ro-Ro/Pax ships

Real value

Ships per week Service time [h]

3.46 7.42

Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation

Simulated confidence interval [LCL,UCL] [3.32, 4.30] [6.58, 7.47]

521


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

the real empirical distribution of the inter-arrival times, respectively.

Fig. 6. MSpE analysis (Ro-Ro passenger ships per week)

4 DESIGN OF EXPERIMENTS The design of experiments was used to plan and execute the simulation runs. In particular, a 24 factorial experimental design was chosen to analyse the evolution of Ro-Ro/Pax turnaround time given some input parameter changes. The 24 factorial experimental design implies that 16 simulation runs are needed to study 4 factors (input parameters) each one having two different levels (a minimum and maximum level). In other words, the simulation runs are made for all the possible factor level combinations. Factorial experimental designs are particularly useful in the early stages of an experimental work [33] to select or screen out the few important main effects from the less important ones. Each simulation run of 60 weeks’ length has been replicated 5 times. The input parameters and their two levels are presented in Table 2. Table 2. Parameters for the 24 experiments design Factor Inter-arrival time of ships [h] Unloading/Loading time of Ro-Ro /Pax [min] Numbers of cars Numbers of trucks

Var.

Level [-1]

Level [1]

x1

42

60

x2

0.5

1.5

x3 x4

50 50

100 100

As only two levels for each input factor were taken into consideration, only a linear regression can be carried out to express the response as a function of the input factors. The simulation results have been analysed using the Minitab® software taking into account the influence of the inter-arrival time of the ships, the loading/unloading time of Ro-Ro/Pax and the number of cars and trucks over the turnaround time. For the levels of the inter-arrival times please refer to the empirical distributions shown in Figs. 7 and 8. In other words, Figs. 7 and 8 represent two histograms that are 20% lower and 20% greater than 522

Fig. 7. Inter arrival time of Ro-Ro/Pax ships (20% lower than the real empirical distribution of the inter-arrival times)

Fig. 8. Inter arrival time of Ro-Ro/Pax ships (20% greater than the real empirical distribution of the inter-arrival times)

5 SIMULATION RESULTS ANALYSIS Since two levels of each input factor have been taken into account, it is interesting to compare how changes in these levels impact the turnaround time using an analysis of variance (ANOVA). The ANOVA partitions the total variability of the turnaround time into different components due to the influence of the change in the inter-arrival time, the Ro-Ro/Pax unloading/loading time, the number of cars and the numbers of trucks. Table 3 and Fig. 9 present the ANOVA results obtained using the Minitab® software. The column DF indicates the degrees of freedom of each input parameter. The adjusted mean squares is a statistical indicator that allows the total variability for each factor to be evaluated. Column P is the probability of accepting as true the hypothesis that a factor has no impact on the response considered (shown here for turnaround time: in other words the column P pertains to the probability of error in accepting the alternative hypothesis that a factor has no impact on the response considered). The ANOVA results show that the factors taken into account as well as their interactions significantly affect the turnaround time and therefore cannot be neglected.

Longo, F. – Huerta, A. – Nicoletti, L.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

Fig. 9. Input factors versus output

Table 3. Analysis of variance for the turnaround time Source x1, x2, x3, x4 x1x2, x1x3, x1x4, x2x3, x2x4, x3x4 x1x2x3, x2x3x4, x1x3x4, x1x3x4 x1x2x3x4 Total

Degrees of Freedom (DF) 4

Adjusted Mean Squares (Adj MS) 0.42

0.00

6

8.25

0.00

4

8.78

0.00

1 15

2.83

0.00

P

100, the turnaround time increases from 9.5 to 12 hours; • and, if the unloading/loading time of the cars and of the trucks increases from 0.5 to 1.5 minutes, the turnaround time will decrease from 11 to 10 hours. ANOVA allows us to evaluate a meta-model of the simulation model that expresses the response as an analytical function of the factors considered. Consequently, the general form of the linear model that relates the turn-around time (Y) to the four input factors (x1, x2, x3, x4) is expressed by Eq. (2). j =4

i =3 j = 4

Y = ∑ β j x j + ∑ ∑ β ij xi x j + j =1

i =2

i =1 j =i +1

j =3

+∑ ∑

k =4

∑β

i =1 j =i +1 k = j +1

Fig. 10. Analysis of variance for the turn-round time

Based on the ANOVA results, shown graphically in Fig. 10, and on the mean value of the turnaround time (8.42 hours), it is possible to conclude that: • if the inter-arrival time increases, the turnaround time of the ship Ro-Ro/Pax will decrease from 11 to 10.5 hours; • if the number of cars unloaded/loaded increases from 50 to 100, the turnaround time will increase from 8.8 to 12.5 hours, while increasing the number of trucks unloaded/loaded from 50 to

i =1 j = 2

k =3

x x j xk +

ijk i

n=4

∑∑ ∑ ∑ β i =1 j =i +1 k = j +1 n = k +1

x x j xk xn + ε . (2)

ijkn i

Table 4 reports the main effects and interaction effects (up to order 4) as well as the coefficents of Eq. (2). The main effects and interaction effects express the variation in the average turnaround time when the factors change from their minimum levels to their maximum levels. From Table 4 we can observe that the three largest effects on the turnaround time are the interactions between x1x3, x2x3, x1x2x3, that correspond to the interactions between inter-arrival time and the number of cars, unloading/loading time and the number of cars, inter-arrival time, unloading/loading time and number of cars, respectively. In addition, Table 4 shows the βi and βij coefficient to be inserted in Eq. (2). Therefore, by using such coefficients the

Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation

523


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

analytical meta-model can be expressed in explicit form (see Eq. (3)). Table 4. Residuals from the 24 design Source Constant x1 x2 x3 x4 x1x 2 x1x 3 x 1x 4 x 2x 3 x 2x 4 x 3x 4 x 1x 2x 3 x 1x 2x 4 x 1x 3x 4 x 2x 3x 4 x 1x 2x 3x 4

Effect 0.12 0.49 -0.24 -0.34 -1.35 2.49 0.79 1.41 0.92 -0.94 2.04 -1.93 -0.44 -0.82 0.84

Coefficient 10.70 0.06 0.24 -0.12 -0.17 -0.68 1.25 0.39 0.70 0.46 -0.47 1.02 -0.97 -0.22 -0.41 0.42

6 CONCLUSIONS The developed DES model allows the macro activities carried out in a Mediterrenean seaport to be explored. The reliability of the simulation model was ensured by applying the MSpE technique for validation purposes. The validation process was a crucial step in the simulation study and was indispensable for considering the seaport DES model as a valid support tool for decisions-making. Preliminary simulation experiments were carried out following a 24 factorial experimental design and the simulation outputs were investigated by applying the ANOVA. This methodology has clearly shown that inter-arrival time, Ro-Ro/Pax unloading/loading time, number of cars, and number of trucks are crucial factors for seaport performance in terms of turnaround time. In addition, an analytical meta-model relating the turn-around time to the input factors was evaluated. The meta-model is an additional tool that can be used apart from (or even jointly with) the simulation model to investigate how the input factors affect the seaport’s behaviour.

Y = 0.06 x1 + 0.24 x2 − 0.12 x3 − 0.17 x4 −

7 REFERENCES

−0.68 x1 x2 + 1.25 x1 x3 + 0.39 x1 x4 + +0.70 x2 x3 + 0.46 x2 x4 − 0.47 x3 x4 + +1.02 x1 x2 x3 − 0.97 x1 x2 x4 − 0.22 x1 x3 x4 −

−0.41x2 x3 x4 + 0.42 x1 x2 x3 x4 + 10.70 . (3)

Eq. (3) can be seen as an additional outcome of the simulation model; indeed the estimated values that have been used to build this metamodel are the simulation outputs. It can be used (apart from the simulation model) for evaluating the port performance (in terms of vessel turn-around time) depending on the values of some well-established critical factors. Therefore it is a decision support tool that port managers can use to understand seaport behaviour or to investigate different operative scenarios when the factors under consideration change. It is worth pointing out that the analysis proposed above is only an example of an application in order to highlight the simulation model potential as a tool for supporting the port manager’s decision making process. In fact, the simulation model can be used to carry out additonal analysis considering multiple performance measures (i.e. average time spent by all the ships in the port areas, container handling equipment efficiency, etc.) and all those input factors that may have an inpact on the port’s performance. 524

[1] UNCTAD (2012). Merchant fleet by flag of registration and by type of ship, annual, 1980-2012, from http:// unctadstat.unctad.org/TableViewer/tableView. aspx?ReportId=93, accessed on 2013-01-08. [2] Lloyd’s Marine Intelligence Unit (2008). Study of maritime traffic flows in the Mediterranean Sea. Final Report - Unrestricted Version of the European Union financed MEDA regional project “Euromed co-operation on Maritime Safety and Prevention of Pollution from Ships – SAFEMED”, from http://80.33.141.76/esatdor/attachments/article/69/ Maritime_Traffic_Flows.pdf accessed on 2013-01-08. [3] ESPO (European Sea Ports Organisation). (2012). Port Performance Indicators Selection and Measurements. Project executive Report, from http://pprism.espo.be/LinkClick.aspx?fileticket=sv7LYCzIJs%3D&tabid=3555, accessed on 2013-0108. [4] Parmenter, D. (2007). Key Performance Indicators, Developing, Implementing, and Using Winning KPIs, 2nd ed., John Wiley and Sons, New Jersey. [5] Bruzzone, A.G., Longo, F. (2010). An advanced system for supporting the decision process within large-scale retail stores. Simulation, vol. 86, no. 12, p. 742-762, DOI:10.1177/0037549709348801. [6] Woo, S.H., Pettit, S.J., Kwak, D.W. (2011). Seaport research, a structured literature review on methodological issues since the 1980s. Transportation Research Part A, vol. 45, no. 7, p. 667-685, DOI:10.1016/j.tra.2011.04.014.

Longo, F. – Huerta, A. – Nicoletti, L.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 517-525

[7] Cortés, P., Munuzuri, J., Ibanez, J.N., Guadix, J. (2007). Simulation of freight traffic in the Seville inland port. Simulation Modelling Practice and Theory, vol. 15, no. 3, p. 256-271, DOI:10.1016/j.simpat.2006.11.004. [8] Yazdani, R., Schmidt, F.A., Merkuryev, Y., Kamperman, F. (2005). The generic animation and simulation tool for ports and terminals. International Journal of Simulation Systems, Science & Technology, vol. 6, no. 7-8. p. 61-69. [9] Bugaric, U., Petrovic, D. (2007). Increasing the capacity of terminal for bulk cargo unloading. Simulation Modelling Practice and Theory, vol. 15, no. 10 p. 1366-1381, DOI:10.1016/j.simpat.2007.09.006. [10] Parola, F., Sciomachen, A. (2009). Modal split evaluation of a maritime container terminal. Maritime Economics & Logistics, vol. 11, no. 1, p. 77-97, DOI:10.1057/mel.2008.22. [11] Bruzzone, A., Longo, F., Nicoletti, L., Bottani, E., Montanari, R. (2012). Simulation, analysis and optimization of container terminals processes. International Journal of Modeling, Simulation, and Scientific Computing, vol. 3, no. 4, art. 1240006, DOI:10.1142/S1793962312400065. [12] Bruzzone, A., Signorile, R. (1998). Simulation and genetic algorithms for ship planning and shipyard layout. Simulation, vol. 71, no. 2, p. 74-83, DOI:10.1177/003754979807100202. [13] Bruzzone, A. (2004). Preface to modeling and simulation methodologies for logistics and manufacturing optimization. Simulation, vol. 80, no. 3 p. 119-120, DOI:10.1177/0037549704045812. [14] Bruzzone, A., Fancello, G., Fadda, P., Bocca, E., D’Errico, G., Massei, M. (2010). Virtual world and biometrics as strongholds for the development of innovative port interoperable simulators for supporting both training and R&D. International Journal of Simulation and Process Modeling, vol. 6, no. 1, p. 89102, DOI:10.1504/IJSPM.2010.032661. [15] Garro, A., Russo, W. (2010). easyABMS: a domainexpert oriented methodology for agent based modeling and simulation. Simulation, Modeling Practice and Theory, vol. 18, no. 10 p. 1453-1467, DOI:10.1016/j. simpat.2010.04.004.

[16] Douma, A.M., Van Hillegersberg, J., Schuur, P.C. (2012). Design and evaluation of a simulation game to introduce a multi-agent system for barge handling in a seaport. Decision Support Systems, vol. 53, no. 3 p. 465-472, DOI:10.1016/j.dss.2012.02.013. [17] Petering, E. (2010). Development and simulation analysis of real-time, dual-load yard truck control systems for seaport container transshipment terminals. OR Spectrum, vol. 32, no. 3 , p. 633-661, DOI:10.1007/ s00291-009-0183-6. [18] Cartenì, A., de Luca, S. (2012). Tactical and strategic planning for a container terminal: Modelling issues within a discrete event simulation approach. Simulation Modelling Practice and Theory, vol. 21, no. 1, p. 123145, DOI:10.1016/j.simpat.2011.10.005. [19] Bruzzone, A.G., Longo, F. (2012). 3D simulation as training tool in container terminals: The TRAINPORTS simulator. Journal of Manufacturing Systems, vol. 32, no. 1, p. 85-98, DOI:10.1016/j.jmsy.2012.07.016. [20] Bruzzone, A.G., Fadda, P., Fancello, G., Massei, M., Bocca, E., Tremori, A., Tarone, F., D’Errico, G. (2011). Logistics node simulator as an enabler for supply chain development: innovative portainer simulator as the assessment tool for human factors in port cranes. Simulation, vol. 87, no. 10, p. 857-874, DOI:10.1177/0037549711418688. [21] Banks, J. (1998). Handbook of Simulation. J. Wiley and Sons, New York, DOI:10.1002/9780470172445. [22] Shannon, R.E. (1975). Systems Simulation: The Art and Science. Prentice-Hall, New Jersey. [23] Gordon, G. (1978). System Simulation, 2nd ed. PrenticeHall, New Jersey. [24] Law, A.M., Kelton, W.D. (2006). Simulation Modeling and Analysis, 4th ed., McGraw-Hill, New York. [25] Scott, D.W. (1979). On optimal and data-based histograms. Biometrika, vol. 66, no. 3 , p. 605-610, DOI:10.1093/biomet/66.3.605. [26] Montgomery, D.C. (2002). Applied Statistics and Probability for Engineers, 3rd ed. John Wiley and Sons, New York. [27] Mee, R.W. (2009). A Comprehensive Guide to Factorial Two-Level Experimentation. Springer-Verlag, New York, DOI:10.1007/b105081.

Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation

525


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2012.942 Special Issue, Original Scientific Paper

Received for review: 2012-12-31 Received revised form: 2013-05-24 Accepted for publication: 2013-06-24

Traffic Modelling and Performance Evaluation in the Kotor Cruise Port Kofjač, D. – Škurić, M. – Dragović, B. – Škraba, A. Davorin Kofjač1,* – Maja Škurić2 – Branislav Dragović2 – Andrej Škraba1 1 University

of Maribor, Faculty of Organizational Sciences, Slovenia of Montenegro, Maritime Faculty, Montenegro

2 University

Simulation modelling of traffic in the Kotor cruise port is used for performance evaluation and optimization of the operational policy. During the summer, the traffic intensity of the cruise ships’ arrivals is higher than in other periods of the year. The increased traffic intensity causes congestion at anchorages, which can lead to significant dissatisfaction of the cruise ship operators and passengers. The simulation model, based on the port calling frequency statistics and port tariff charges, is developed. First, the simulation model is validated against the real data. Second, several simulations are performed, where the scenarios of an extended main berth and of increased traffic intensity are evaluated in order to minimize congestion and to maximize revenue. Simulation results indicate significant queue reductions and a higher revenue, thus justifying the intention in the main berth extension. Keywords: traffic modelling, port revenue, operational policy, simulation, model validation, Kotor cruise port

0 INTRODUCTION Over the previous decade, the cruise ship market has become extremely attractive due to increased destinations and leisure activities. Recently, it has become evident that the increased traffic throughput in ports, has led to serious traffic congestion [1]. Therefore, it is in ports’ interest to reduce their queues in order to obtain higher profits. The port of Kotor was introduced as a new cruise destination (which had not previously been investigated in the literature) by Dragović et al. [2]. They described the Boka Kotorska Bay location, the entrance of the port of Kotor, the draft characteristics, how ships approach the port and other performance metrics such as berth length and draft along the quay. A simulation model was developed to investigate the impact of the possible main berth extension on the port’s operational policies. To briefly summarize, the Kotor cruise port consists of one berth for receiving cruise ships, a river berth for smaller cruise ships (which is not taken into account in this study), and two anchorages that are positioned near the port (Fig. 1). If the main berth is taken, larger ships are anchored, i.e. queued; this is a common scenario in the peak season. However, passengers from anchored ships do not have a proper approach for disembarking and sightseeing. The consequence is a lesser revenue for the city of Kotor and its surroundings. Furthermore, anchored ships are subject to a higher probability of collision with other ships because of the narrow passage to the port and the tendering boats, used to take passengers to land, represent a certain danger for tourists. 526

Anchorage

Anchorage

Cruise berth

Fig. 1. Location of berth and anchorages in Kotor Bay

This paper aims to justify an investment in the main berth extension in order to reduce the number of ships in the queue and to increase the port’s revenue. The analysis employs an analytical approach and discrete event simulation to provide a model of the cruise port sufficient to produce valid results. The rest of this paper is organized as follows. In the next subsection, a literature review is given. Section 1 provides port-calling frequency statistics to show the importance of incoming traffic in Kotor cruise port. In Section 2, a problem formulation with extended berth assumption and port revenue analysis is described. Next, Section 3 provides queuing assumptions of the considered problem while in Section 4, the simulation model with input data, variable distribution and model validation is explained. Computational results are given in Section

*Corr. Author’s Address: University of Maribor, Faculty of Organizational Sciences, Kidričeva cesta 55a, 4000 Kranj, Slovenia, davorin.kofjac@fov.uni-mb.si


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

5. The main conclusions and suggestions for future directions are presented in Section 6. 0.1 Literature Review Despite the fact that many investigations related to cruise ships have attempted to find the key influential factors for an effective cruise destination, there are only a few papers that deal with cruise port infrastructure improvements based on revenue analysis. In [3], port technical information, port marketing and development, and cruise ship activity at Leith (northern Europe) with detailed port charges and plans for a new cruise ship terminal are given. A shift-share analysis to adapt the procedure to the cruise business and execute it for the first time in order to describe the change in passenger flows through ports is described in [4]. An example of the passenger head tax charges was introduced to indicate the importance to the port’s revenue stream. In [5], a simple analytical model of cruise and cargo queuing is given. On the basis of summary of marginal costs and benefits, the author supported an automatic priority for cruise ships. In [6], a dynamic programming model to maximize the average revenue rather than the profit from the cruises is proposed. The derived optimal revenue and the corresponding total cost were studied simultaneously. In contrast, considering the queuing theory, in [7] potential risks and their levels as consequences of increased liquefied natural gas (LNG) activities are identified. Using a quantitative approach, collision (probability of collision can be described as a stationary Poisson distributed stochastic process) and grounding risk assessment is obtained through an automatic identification system (AIS). A model for optimizing vehicle schedules under disrupted conditions is developed in [8]. It optimizes the recovery of a single-terminal system with relatively short feeder routes on which vehicle round-trip

times are exponentially distributed, and arrivals at the terminal are Poisson-distributed. In [9], seaports are described via agent-based modelling and system dynamics, showing that the queuing theory can be effectively used in conjunction with system dynamics and statistical distributions. Taking into account the operational policy in the cruise port, simulation modelling of cruise ship traffic in the Boka Kotorska Bay is proposed in [2]. The solution for minimizing the number of ships at anchorage is given, representing a solid foundation for this study. Considering the literature review and especially [2], we can describe the objectives of the paper. The first is to develop a simulation model that is adequate for describing the cruise ship traffic volume in the port of Kotor, regarding only larger ships, unlike [2], in which all ships were taken into account. The second objective is to show the increased port revenue per peak season based on the extended berth assumption and increased traffic volume prediction. Unlike previous research, this paper introduces an analytical approach for the cruise port traffic modelling. Third, the revenue criteria function to maximize the cruise port’s total revenue is introduced. 1 PORT CALLING FREQUENCY STATISTICS The main parameter for calculating the ship traffic volume is the traffic intensity. In this study, we analyze the cruise ships’ arrivals in an eight-month period in 2011 and a seven-month period in 2012 (peak periods of the season) [10] and [11]. Moreover, we have taken into account only cruise ships that went to the main berth or anchorages, excluding those going to the river berth. The ship arrival rate was 0.90 ships/day in 2011 and 1.14 ships/day in 2012. Our database contains the ships’ identification (ID), the date of ship’s call, arrival, and service and departure time [2], [10] and [11]. The frequency of

Fig. 2. Port calling frequency of individual cruise ship Traffic Modelling and Performance Evaluation in the Kotor Cruise Port

527


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

Fig. 3. Berth layout in case of new scenario [2]

port callings is presented in Fig. 2. The methodology was adapted from [12]. Similar to [2], but excluding the ships that went to the river berth, we recorded the port calling frequency statistics: in 2011, 54 cruise ships made 219 port calls, while in 2012, 51 ships made 243 port calls. From June to October in 2011 and 2012, there were two or more port calls in a day, and there were situations when cruise ships arrived and departed almost simultaneously [2], [10], [11] and [13], thus making a queue at the anchorages because only one ship can be moored at the main berth. It should be noted that about 30% of the callers produced approximately 60 to 65% of the total port calls. Moreover, three or more port calls (58% in 2011 and 51% in 2012) were made by the 23 and 27 most frequent callers (21 and 14% of the total number of ships in 2011 and 2012, respectively). This implies that the most frequent callers would occasionally make three or four port calls per day simultaneously [2], thus creating traffic congestion by occupying both the main berth and both anchorages. 2 PROBLEM FORMULATION The company that provides port services in the port of Kotor is the Port of Kotor A.D. (PKAD). It is a public limited company in which the municipality of Kotor is a key stakeholder, with a 56.97% shareholding. In this study, we assume that PKAD might extend the main berth for flexible berth allocation planning, especially in the peak season when the large cruise ships are more frequent. 2.1 New Scenario Assumption In this study, we assume a main berth extension from 380 to 500 m (total quay length) to allow for servicing of two large ships simultaneously. Thus, the guaranteed water depth of 8 m extends approximately 300 m along the quay (Fig. 3, left side), enabling the port to handle larger ships. More than 96% of port calls occupied this length in both 2011 and 2012, while only 1.4% in 2011 and 3.3% in 2012 of total port 528

calls were serviced simultaneously. The additional 200 m of quay length provides service to ships with the maximum allowable draft of 6.5 m (Fig. 3, right side). Currently, the proposed quay length of 500 m is the only possible scenario, because the right side of the quay has the maximum allowable draft of 5 m, while the minimum safety clearance required for a ship making a turn in the harbor basin on the left side does not permit the extension of more than 120 m. To summarize, such a quay extension would ensure more flexible berth allocation planning, especially in the peak season, when the large cruise ships are more frequent. The new scenario assumption consists of two possibilities. In the first, we assume that the existing berth extends to 500 m on the left side of the quay with mooring dolphins without a gangway [2]. The second implies the extension using mooring dolphins but with a gangway or a construction of a floating dock, such as steel pontoon barges. This facility presented in Fig. 3 consists of a system of two mooring/breasting dolphins, where mooring dolphins are connected to the existing main berth with the gangway [2]. Therefore, a simulation is used to propose a new operational policy because in the case of the extended quay, many cruise ships can be transferred from anchorages to the main berth. 2.2 Port Revenue Analysis In port literature, fees and charges are mostly levied as per gross tonnage (GT) which is an essential element of a cruise port’s revenue stream. The cruise port revenue consists of [10], [11] and [13]: • Services to cruise ship (berthing, anchoring, mooring/unmooring); • Services to passengers (ship’s tender and embarkation/disembarkation); • Navigation services (port signalling, pilotage service using a pilotage boat); • Hinterland services (fresh water supply, waste removal, car (bus) fee, inspection and custom fee);

Kofjač, D. – Škurić, M. – Dragović, B. – Škraba, A.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

Fig. 4. Port revenue model

Security and safety services (unfortunately, still not defined in the Kotor cruise port). Berthing fees depend on GT, the length of a cruise ship, fresh water supply, waste removal and charge per passenger on-board on arrival. Furthermore, mooring/ unmooring operations require an additional small boat. In the case of an anchored ship, the parameters for GT and passengers on-board on arrival are taken into account. In this situation, a ship’s tender with lifeboats is used for embarkation/disembarkation and the charge for every boat call at berth is calculated. The pilotage fee is based on the GT of a ship as an extra charge for the pilot boat. This fee applies to: navigating the ship into the port from the entrance of the Boka Kotorska Bay and guiding it to the berth or anchorage; maneuvering the cruise ship within the Kotor Bay and port; unmooring and navigating the ship from the berth or anchorage. Bunker, car (bus) fees, as well as custom and inspection fees are not taken into account in this analysis because of the absence of data [10], [11] and [13]. The port revenue model is illustrated in Fig. 4. With the services specifications, Fig. 4 also shows the equations that provide a basis for the total port revenue calculation per ship with the following parameters: Cb and Ca are revenue for the ship at berth and anchorage, respectively; GT gross tonnage of a ship; pgt and pat price per GT for a ship at berth and anchorage; SL ship’s length; psl price per meter of the ship’s length; WRf initial fixed waste removal only for ships that go to berth; PS number of passengers per ship, for which the price for the passenger is fixed at €1.20. The pilotage services represent the 20% of revenue for services of the cruise ship. The ship’s tender, ST, is calculated on the basis of the number of passengers that are embarking / disembarking from/to

the ship PSe/d, and the number of lifeboats used nt. Finally, waste removal per berth WRb depends on GT and price per one container of waste pwr. There are also some other charges that account for part of the port tariff per ship call that have influence on the total port revenue, such as an additional charge for daily working overtime, night work, working during shifts on Saturdays and Sundays, and holiday work, etc. 3 QUEUING APPROACH IN CRUISE PORT The arrival and service routine using the queuing system in the port is shown in the flow chart in Fig. 5.

Fig. 5. Flow chart for arrival routine based on queuing and revenue analyses

First, it is assumed that the berth has no vacancy and all cruise ships have to go to the queue until the

Traffic Modelling and Performance Evaluation in the Kotor Cruise Port

529


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

main berth becomes available. Second, it must be assumed that most of these ships that went to queue first do not need to go to the berth when it becomes available. These ships rest in the queue (anchorages) for a certain period of time. However, there were rare situations when the berth becomes vacant. In such a case, the ship from the queue changes its position and goes to the main berth. Next, the charges are determined, and the revenue of the port is calculated. Finally, when the ship’s service time runs out, the ship leaves the port. Here, we give the explicit formulae for port with one berth and a finite number of ships in the queue that represent the real situation in the Kotor cruise port. Hence, the corresponding queuing model in Kendall’s notation is M/D/1/m. In [14], the transient analytical solution of M/D/1/m queues based on M/D/1/m as a finite capacity queuing system, with m – 1 waiting spaces in the queue is defined. In our study, we also suppose that cruise ships arrive according to a Poisson process at arrival rate λ. The mean service rate per berth is μ. Therefore, cruise ships that see a full system upon arrival are rejected and do not further influence the system. The traffic intensity for this model is defined θ k −θ e denote the probability as θ = λ / μ. Let α k = k! of k arrivals during a service of the ship. Using a probability transition matrix of the embedded Markov chain [14] and [15]; it is possible to derive the stationary probability distribution of the number of cruise ships in the M/D/1/m queue: 1 p0 = , (1) 1 + θ bm−1

pk =

bk − bk −1 , k = 1, ..., m − 1, 1 + θ bm−1 pm = 1 −

(2)

bm−1 , (3) 1 + θ bm−1

where the coefficients bn are given by b0 = 1 and n (−1) k (n − k ) k e( n −k )θ θ k , ∀n ≥ 1. (4) bn = ∑ k! k =0 According to the second formula of Theorem 2 from [14], for an average waiting time in the M/D/1/m queue, m −1   bk − m  ∑  1  , (5) Wq ( M / D /1/ m ) =  m − 1 − k =0 θ bm−1  µ      530

whence by Little’s law it follows that the average number of cruise ships in the queue is

Lq ( M / D /1/ m )

m −1   bk − m  ∑  λ  , (6) =  m − 1 − k =0 θ bm−1  µ     

which implies m −1

Lq ( M / D /1/ m ) = (m − 1)θ −

∑b k =0

k

−m

bm−1

. (7)

As observed in [16], a more exact approach to the calculation of average waiting time in the M/D/nb/m queue is possible through the M/Ek/nb/m queue with Erlang-k services by making k large enough and number of berths, nb > 1. In [16], it was also observed that many related approximations would yield a new approach for the calculation of the waiting time probabilities in the M/D/nb/m queue. Of course, the M/D/1/m queue is an exceptional case of the M/G/1/m queue for which precise equations of various performance measures have already been derived in the literature. In contrast, also observing that there are no exact solutions for the M/D/nb/m queue, a good heuristic to approximate M/D/nb/m queue with an M/D/1/m queue, the service rate of which is nb times greater than those of M/D/nb/m queue, was proposed quite recently in [17]. Their approach is well verified by related simulation results. Bearing in mind that their idea consists of the obvious heuristic fact that an M/D/nb/m queue can be replaced by a M/D/1/m queue, the service rate of which is increased to balance the reduction from a multi-server system to a single server one [17]. In view of their numerical results based on analytical expressions for the M/D/1/m queue, we believe that their proposed simulation method is much more adequate than the classic Cosmetatostype approximations, which are limited to the infinite capacity queuing system (for a discussion, see [16], subsection 2.3). 4 SIMULATION MODEL In our case, discrete event simulation is used to provide insight into the consequent changes of the operational policy. This simulation utilizes a mathematical/logical model of a physical system that portrays state changes at precise points in simulated time [18] and [19]. The

Kofjač, D. – Škurić, M. – Dragović, B. – Škraba, A.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

proposed simulation model was developed in Flexsim [20], with a screenshot provided in Fig. 6. The main window provides instant insight into some variables. In the upper part of the screenshot, the revenues can be observed: TR is total revenue, Ca+Cb revenue of ships at berth and anchorage, PCa+PCb revenue from pilotage services for ships at berth and anchorage, WRb revenue from waste removal and ST revenue from ships’ tendering services. In the lower part, average berthing (service), approaching and departure time can be observed, together with the total number of ships at the main berth and anchorages.

average GT per ship going to the anchorage increased almost seven times while the same parameter for the berth increased three times. Fig. 7 justifies the object of our study, which is based on change in operational policy in the case of the new scenario. The input data for the simulation model is based on the traffic intensity and actual cruise ship port callings for the eight-month period in 2011 and sevenmonth period in 2012 [10]. This involved 219 ship calls in 2011 and 243 ship calls in 2012. The ships were categorized into the following classes according to the GT of ships (see Fig. 8): up to 5000 GT; 5001– 10000 GT; 10001–20000 GT; 20001–40000 GT; 40001–60000 GT; 60001–80000 GT; 80001–100000 GT and over 100001 GT. Ship arrival probabilities were as follows: 17.35 and 15.23% for first class, 12.33 and 11.11% for second, 13.24 and 16.87% for third, 26.03 and 16.46% for fourth, 19.63 and 24.28% for fifth, 6.85 and 5.35% for sixth, 0.00 and 2.88% for seventh and 4.57 and 7.82% for eighth class of ships for 2011 and 2012, respectively. As can be observed in Fig. 8, the class of ships with the highest traffic is fourth in 2011 while in 2012, the fifth class of ships made the biggest number of port calls.

Fig. 6. A screenshot of the simulation model with intermediate simulation results

The simulation model structure is made in accordance with the flow chart in Fig. 5. All input values for the parameters within the described model are based on the data collected during the study period. The simulation model includes the traffic intensity of ships with different lengths and GT, their arrival and service patterns and their costs in the Boka Kotorska Bay.

Fig. 7. Increase of GT and passenger capacity per ship from 2008 to 2012 [10]

4.1 Input Data After the data analysis, it is evident that the total revenue mostly depends on the GT of the ship and the number of passengers per ship [10]. In Fig. 7, the relation between average GT per ship and average passenger capacity per ship for the peak seasons from 2008 to 2012 is given. Evidently, there is the increased trend of average GT and passenger capacity of ships for both berth and anchorages. This figure affirms the increased trend of cruise ship arrivals from 2008 to 2012. Therefore, the port’s revenues were drastically changed. The average passenger capacity per ship increased five times and three times for ships going to the anchorage and berth, respectively. However, the

Fig. 8. Number of port calls in respect to GT class for considered study period [10]

Traffic Modelling and Performance Evaluation in the Kotor Cruise Port

531


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

4.2 Variable Distributions The schedule of cruise ships in the Kotor Bay depends on arrival times [10]. Here, according to the obtained data, we have fitted inter-arrival times CDF in MATLAB. Obviously, the CDF corresponds to the exponential distribution, thus justifying the Poisson process of arrivals. Curve fittings for 2011 and 2012 are presented in Fig. 9, where probability and variable axis are inverted, because of the limitations of the curve fitting module. Also, the curve exponential functions, with corresponding a, b, c, d coefficients, and R-square values are given in Fig. 9. R-square values of 0.95 for 2011 and 0.96 for 2012, together with sum squared error (SSE) and root mean square error (RMSE) values, indicate a quality fit.

a)

time is known in advance. It is also worth noting that in 2011, 3% of total calls stayed in port during the night while in 2012, that figure was 6%; however, this data did not affect the obtained deterministic average service time of ships. Table 1. Distributions for number of passengers Year 2011 2012

Distribution 55 + WEIB(32.4, 4.94) 54.5 + 46 × BETA(2.61, 0.923)

4.3 Model Validation Model validation is an essential process in the development of a simulation model, whether the systems fulfils the purpose for which it was intended must be determined [21]. For validation purposes, several tests are performed, and several key performance measures are observed to see if they are close to the real data. We have run 100 replications of each simulation scenario, in which one berth and two finite spaces in the queue are assumed, for peak seasons in 2011 and 2012. Validation results are presented in Tables 2 to 4. Simulation results are given with a performance measure mean value and the standard deviation in brackets. The results of the average number of ships (ANS), average number of ships in the queue (ANSQ) and average waiting time in the queue (AWTQ) are given in Table 2. Table 3 shows the revenue comparison between real data and simulation results, while Table 4 provides a comparison of port calls, GT and passenger capacity (PAX) for the main berth (MB) and anchorages (Anc). Table 2. Model validation with regard to ANS, ANSQ and AWTQ

ANS ANSQ AWTQ [h]

b)

Fig. 9. CDF of inter-arrival times with fitted exponential curves, a) 2011 and b) 2012

The number of passengers assigned to a ship follows a Weibull and Beta distributions for 2011 and 2012, respectively (see Table 1). Finally, the real data analysis indicates that the deterministic average service time can be assumed with a value of 10 hours for 2011 and 9 hours for 2012. Typically, cruise ships arrive early in the morning and leave the port late at night. Consequently, their service 532

Real data 2011 2012 9.75 8.50 0.23 0.23 2.48 2.51

Simulation 2011 2012 9.60 (0.62) 8.60 (0.54) 0.24 (0.10) 0.20 (0.15) 2.62 (0.32) 2.59 (0.27)

To confirm the validity of the model, we have performed a one-sample two-tailed t-test. We have tested the null hypothesis, where it is assumed that the sample mean, e.g. average number of ships in the queue, is equal to the specified value of the real data versus the alternative hypothesis, where it is assumed that the sample mean is not equal to the specified value, as proposed in [18]. With a chosen level of significance α = 0.05, we were not able to reject the null hypothesis for each case in Tables 2 to

Kofjač, D. – Škurić, M. – Dragović, B. – Škraba, A.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

4. Therefore, we can tentatively accept the model as valid [18]. Table 3. Model validation with regard to revenue

Revenue Tariff Pilotage Waste removal Pilot and/or mooring boat Ship’s tender Total

Real data 2011 2012 [104 €] [104 €] 89.5 112.1 22.3 28.0 4.1 4.7 4.7 1.7 122.3

5.1

Simulation 2011 2012 [104 €] [104 €] 86.9 (5.1) 109.9 (4.2) 21.7 (1.3) 27.5 (1.1) 3.9 (0.9) 4.5 (0.7) 4.5 (0.8)

4.9 (0.7)

8.7 1.6 (0.5) 8.4 (0.9) 158.7 118.6 (6.3) 155.2 (6.5)

Table 4. Model validation with regard to number of port calls, GT and PAX for main berth and anchorages

MB Calls Anc Calls Total Calls

MB GT Anc GT Ttl GT

MB PAX Anc PAX Ttl PAX

Real data 2011 2012 136 149 83 94 219 243 Real data 2011 2012 [104] [104] 410.1 486.9 313.7 434.9 723.8 921.8 Real data 2011 2012 [103] [103] 102.3 121.4 65.6 100.7 167.9 222.1

Simulation 2011 2012 132 (8.1) 148 (4.0) 82 (3.8) 89 (8.5) 214 (9.1) 237 (11.1) Simulation 2011 2012 [104] [104] 400.2 (19.4) 514.9 (52.2) 301.9 (24.3) 398.5 (63.1) 702.1 (37.7) 913.4 (28.5) Simulation 2011 2012 [103] [103] 95.5 (12.4) 119.0 (6.4) 68.4 (6.4) 98.6 (6.3) 163.9 (9.6) 217.6 (8.8)

By presenting these simulation results, we can conclude that the model reflects the real system accordingly, and it is suitable to make a further ‘whatif’ analysis by varying input parameters. 5 COMPUTATIONAL RESULTS Following the model validation, we have run several simulation scenarios in order to simulate the possible port performance as a consequence of the main berth extension to cope with the increasing traffic. We have assumed that the existing berth (EX) has been extended (EXT) to 500 m, thus being capable of servicing two ships simultaneously if their total length does not exceed the berth length. Further, we have assumed the increase of traffic by shortening the inter-arrival times by 20 and by 40%. Hence, we have run the following scenarios: a) EXT_11, an extended

berth with real inter-arrival times in 2011, b) EXT_12, an extended berth with real inter-arrival times in 2012, c) EX20_12, an existing berth with 20% shorter interarrival times than in 2012, d) EX40_12, an existing berth with 40% shorter inter-arrival times than in 2012, e) EXT20_12, an extended berth with 20% shorter inter-arrival times than in 2012, f) EXT40_12, an extended berth with 40% shorter inter-arrival times than in 2012. Again, we have run 100 replications of each simulation scenario. Table 5 (the left part) presents the results for original incoming traffic in 2011 and 2012, and the increased incoming traffic, in the case of the existing berth. The numbers in brackets represent the standard deviation of replication results. The results indicate that if the incoming traffic had been increased, the ANSQ and AWTQ would also have been increased as expected in both 2011 and 2012. In 2011, ANSQ would increase by 25 and 71% if the incoming traffic would increase by shorter inter-arrival times by 20 and 40%, respectively. AWTQ would increase by 12 and 22%. Similarly, in 2012, the ANSQ would increase by 33 and 76% while AWTQ would increase by 17 and 33%. The right part of Table 5 reflects the scenarios with the extended berth. As expected, the ANSQ and AWTQ would significantly decrease in comparison with the existing berth scenarios, because the extended berth is capable of servicing two ships simultaneously. In 2011, the ANSQ would decrease by 99.5, 93 and 51% for the original and increased incoming traffic due to shorter inter-arrival times by 20 and 40%, respectively. AWTQ would decrease by 99.2, 92 and 59%. In 2012, the ANSQ would decrease by 95, 93 and 87%, and the AWTQ would decrease by 95, 92 and 85%. The results in Table 6 reveal that the extended berth scenarios yield higher revenues than the ones with existing berth real data (RD) if comparing RD_11 vs. EXT_11 (for 4%), RD_12 vs. EXT_12 (for 5%), EX20_12 vs. EXT20_12 (for 2%) and EX40_12 vs. EXT40_12 (for 4%), respectively. The total revenues are increased because of the increased value of tariff (by 3, 8, 4 and 6%), pilotage (by 3, 8, 4 and 6%), waste removal (by 39, 41, 42 and 52%) and pilot and/ or mooring boat (by 22, 16, 16 and 23%). The ship’s tender revenue is decreased (by 51, 66, 68 and 67%), because fewer ships are being anchored, as shown in Table 7. However, this decrease is lower than the increase of the other revenue categories. Furthermore, analysis of simulation results with regard to GT, PAX and number of port calls in Table 7, reveals some more aspects that affect increased revenues in the case of the extended berth. Again,

Traffic Modelling and Performance Evaluation in the Kotor Cruise Port

533


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

Table 5. Simulation results with regard to ANSQ and AWTQ EX_11 EX_12 ANSQ AWTQ [h] ANSQ AWTQ [h] Original incoming traffic 0.24 (0.13) 2.62 (0.53) 0.20 (0.12) 2.59 (0.46) Shorter inter-arrival times by 20% 0.30 (0.14) 2.92 (0.48) 0.27 (0.17) 3.03 (0.50) Shorter inter-arrival times by 40% 0.41 (0.20) 3.18 (0.64) 0.36 (0.21) 3.44 (0.59)

EXT_11 EXT_12 ANSQ AWTQ [h] ANSQ AWTQ [h] 0.001 (0.0032) 0.02 (0.05) 0.01 (0.02) 0.14 (0.21) 0.02 (0.03) 0.22 (0.28) 0.02 (0.03) 0.24 (0.28) 0.20 (0.14) 1.3 (0.90) 0.05 (0.06) 0.5 (0.50)

Table 6. Simulation results with regard to revenue categories Tariff Pilotage Waste removal Pilot and/or mooring boat Ship’s tender Total

RD_11 895458 223864 40640 46750 16714 1223426

EXT_11 922322 230580 56489 57035 8190 1274616

RD_12 1121286 280321 46960 51750 86812 1587129

EXT_12 1210988 302747 66213 60030 29516 1669494

EX20_12 1300691 325172 51656 59512 102438 1839469

EXT20_12 1356756 339189 73727 69345 32988 1872005

EX40_12 1715567 428891 65744 76590 143240 2430032

EXT40_12 1816483 454120 100494 94185 46878 2512160

Table 7. Simulation results with regard to number of port calls, GT and PAX MB Calls Anc Calls Total Calls MB GT Anc GT Total GT MB PAX Anc PAX Total PAX

RD_11 136 83 219 4101223 3137297 7238520 102383 65614 167997

EXT_11 198 10 208 5420367 1600834 7021201 132111 30788 162899

RD_12 149 94 243 4869523 4349352 9218875 121410 100655 222065

we are comparing RD_11 vs. EXT_11, RD_12 vs. EXT_12, EX20_12 vs. EXT20_12 and EX40_12 vs. EXT40_12. The increased share (from 62 to 92%, from 61 to 95%, from 65 to 94% and from 54 to 94%) of ships being berthed rather than being anchored, yields higher Cb because more ships are being charged (SL×psl) and WRf, which is the main difference against Ca. A higher share of ships at MB also increases the revenue of Cb because of the increased share of MB GT against Anc GT (from 57 to 77%, 52 to 83%, 57 to 85% and 5 to 85%) and MB PAX against Anc PAX (from 61 to 81%, 55 to 87%, 59 to 87% and 58 to 88%). Finally, the extended berth significantly reduces the number of ships at anchorages. Again, if comparing RD_11 vs. EXT_11, RD_12 vs. EXT_12, EX20_12 vs. EXT20_12 and EX40_12 vs. EXT40_12, the number of ships at anchorage is reduced by 79, 87, 84 and 83%, respectively. 534

EXT_12 226 12 238 7634041 1500328 9134369 189991 27390 217381

EX20_12 182 99 281 6025535 4490234 10515769 150252 104846 255098

EXT20_12 267 16 283 8896540 1619465 10516005 223749 31673 255422

EX40_12 223 119 342 7764101 5905092 13669193 193676 137395 331071

EXT40_12 322 20 342 11703121 2065257 13768378 292348 40244 332592

6 CONCLUSION This paper proposes analytical and simulation approaches for traffic modelling and performance evaluation, and to reveal the practices to manage the better movement of ships in the Kotor cruise port. The simulation model was developed to analyze the impact of the possible PKAD investment in the main berth extension while taking the existing and predicted increased traffic into account. The model was verified and validated against the real data. On the basis of simulation results, the main conclusions are drawn. First, the extended berth scenarios yield higher revenues than the ones with the existing berth. Second, a higher share of ships at the main berth significantly decreases the number of ships at the anchorages, thus reducing the possibility of accidents while using the ship’s tender service. That would possibly impact the tourism revenues of the city of Kotor. Third, by investing in the main berth extension, the port is capable of coping with the increased incoming traffic even up to 40%.

Kofjač, D. – Škurić, M. – Dragović, B. – Škraba, A.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 526-535

Further, this paper provides a correlation between ship’s GT and PAX that are the most salient parameters for berth extension. In this manner, the study offers many improvements considering the different insights for operational policy of tendering service in the coming years. The proposed methodology may be used for improving various segments of the port, especially with regard to the ship traffic intensity and operational policies, which can be considered as the core components of the system. Future research is possible in several areas. First, the impact of the increased number of tourists disembarking to the main berth on the economy of local area is vital. Second, the construction of the passenger terminal building might lead to a better PKAD and municipality economy. Therefore, developing security and safety services in future is necessary in order to attract more operators to the Kotor cruise port. Finally, gas emissions and ballast water management strategies might be investigated. 6 ACKNOWLEDGMENT This work was partially supported by the Ministry of Higher Education, Science and Technology of the Republic of Slovenia; Program No. P5-0018. 7 REFERENCES [1] Almaz, O.A., Altiok, T. (2012). Simulation modeling of the vessel traffic in Delaware River: Impact of deepening on port performance. Simulation Modelling Practice and Theory, vol. 22, no. 3, p. 146-165, DOI:10.1016/j.simpat.2011.12.004. [2] Dragović, B., Škurić, M., Kofjač, D. (2012). A proposed simulation based operational policy for cruise ships in the Port of Kotor. Maritime Policy & Management, accepted for publication on February 4th, 2013. [3] Baird, A.J. (1997). An investigation into the suitability of an enclosed seaport for cruise ships the case of Leith. Maritime Policy & Management, vol. 24, no. 1, p. 3143, DOI:10.1080/03088839700000054. [4] Marti, B.E., Cartaya, S. (1996). Caribbean cruising: Competition among US homeports. Maritime Policy & Management, vol. 23, no. 1, p. 15-25, DOI:10.1080/03088839600000049. [5] Wood, T.W. (1982). The economics of mixed cargo and cruise ship traffic in a port. Journal of Transport Economics and Policy, vol. 16, no. 1, p. 43-53. [6] Hersh, M., Ladany, S.P. (1989). Optimal scheduling of ocean cruises. INFOR, vol. 27, no. 1, p. 48-57. [7] Perkovic, M., Gucma, L., Przywarty, M., Gucma, M., Petelin, S., Vidmar, P. (2012). Nautical risk assessment for LNG operations at the Port of Koper. Strojniški

vestnik - Journal of Mechanical Engineering, vol. 58, no. 10, p. 607-613, DOI:10.5545/sv-jme.2010.265. [8] Markovic, N., Schonfeld, P. (2013). Scheduling for a single-terminal intermodal system recovery with Poisson arrivals. Strojniški vestnik - Journal of Mechanical Engineering, In press, DOI:10.5545/svjme.2010.268. [9] Latilla, L. (2011). Modelling seaports with agent-based modelling and system dynamics. International Journal of Logistics Systems and Management, vol. 10, no. 1, p. 90-109, DOI:10.1504/IJLSM.2011.042055. [10] Port of Kotor A.D. Statistic information (2012), from http://www.portofkotor.co.me/en/O-luci/statisticinformations.html, accessed on 2012-12-15. [11] Statistical Office of Montenegro –MONSTAT. Foreign vessels on cruise in Montenegro, 2011, from http:// www.monstat.org/userfiles/file/turizam/kruzna%20 putovanja/Foreign%20vessels%20on%20cruise%20 in%20Montenegro,%20Release%20PDF.pdf, accessed on 2012-12-15. [12] Tzannatos, E. (2010). Cost assessment of ship emission reduction methods at berth: the case of the Port of Piraeus, Greece. Maritime Policy & Management, vol. 37, no. 4, p. 427-445, DOI:10.1080/03088839.2010.48 6655. [13] Auditor’s Report for Port of Kotor (2011). from http:// scmn.me/fajlovi/LUKO201112R.pdf, accessed on 201211-15. [14] Garcia, J.-M., Brun, O., Gauchard, D. (2002). Transient analytical solution of M/D/1/N queues. Journal of Applied Probability, vol. 39, no. 4, p. 853-864, DOI:10.1239/jap/1037816024. [15] Brun, O., Garcia, J.M. (2000). Analytical solution of finite capacity M/D/1 queues. Journal of Applied Probability, vol. 37, no. 4, p. 1092-1098, DOI:10.1239/ jap/1014843086. [16] Tijms, H. (2006). New and old results for the M/D/c queue. International Journal of Electronics and Communications, vol. 60, no. 2, p. 125-130, DOI:10.1016/j.aeue.2005.11.008. [17] Rajabi, A., Dadlani, A., Hormozdiari, F., Khonsari, A., Kianrad, A. Razi, H.S. (2008). Analysis of the impact of wavelength converters on contention resolution in optical burst switching. 2nd Asia International Conference on Modelling & Simulation, p. 259-264. [18] Banks, J., Carson, S.J., Nelson, B.L., Nicol, D.M. (2010). Discrete-event System Simulation, 5th ed., Prentice Hall, Upper Saddle River. [19] White, L.R. (2012). A hierarchical production planning system simulator. International Journal of Simulation Modelling, vol. 11, no. 1, p. 40-57, DOI:10.2507/ IJSIMM11(1)4.199. [20] Flexsim Simulation Software. Flexsim, from http:// www.flexsim.com, accessed on 2012-12-15. [21] Desel, J. (2002). Model validation – A theoretical issue? Esparza, J. , Lakos, C. (eds.), 23rd International Conference ICATPN 2002, Lecture Notes in Computer Science, vol. 2360, p. 23-43.

Traffic Modelling and Performance Evaluation in the Kotor Cruise Port

535


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2012.931 Special Issue, Original Scientific Paper

Received for review: 2012-12-21 Received revised form: 2013-05-04 Accepted for publication: 2013-06-04

Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals Lu, B. – Park, N.K. Bo Lu1,* – Nam Kyu Park2

1 Dalian

University, Institute of Electronic Commerce and Modern Logistics, China University, School of Port and Logistics, South Korea

2 Tongmyong

With the rapid expansion of international trade, a distinctive feature of the contemporary container terminal industry is that competition has become more intense. If container terminal managers can gain a proper appreciation of their various productivity factors, they may be able to identify which factors have a more positive influence on productivity. The core of sensitivity analysis for evaluating terminal productivity is to remove input variables one by one, then re-estimate the correlation between productivity and investment. From this perspective, sensitivity analysis provides a more appropriate benchmark for identifying which factors are more critical for productivity improvement. This analysis has been variously studied utilizing either Data Envelopment Analysis or Regression Analysis. Given the strengths associated with these two analyses, this paper applies both approaches to the same set of data for 28 major East Asian container terminals and compares the results of the efficiency. The results of this study can provide a useful reference to port managers for developing improvement strategies. Keywords: data envelopment analysis, regression analysis, sensitivity analysis, container terminal productivity

0 INTRODUCTION A distinctive feature of the contemporary container port industry is that competition has become fiercer than ever [1]. Improving productivity sufficiently to accommodate a large portion of the anticipated increase in container traffic presents a particular challenge to terminal operators and port authorities. As the demand for international trade and global logistic services continues to increase, to remain competitive [2], in [3] the authors claimed that container terminals have to invest heavily in sophisticated equipment or in dredging channels to accommodate the most advanced and largest container ships. It is necessary to note that pure physical expansion is constrained by a limited supply of available land, especially for urban-centric ports, and escalating environmental concerns [4]. In addition, excessive and inappropriate investment also can induce inefficiency and the wasting of resources. In this context, expanding port capacity by improving the productivity of terminal facilities and exploring the critical factors affecting the productivity appears to be a viable solution [5]. For a container terminal, productivity performance makes a significant contribution to the terminal’s survival prospects and competitive advantage [6]. Traditionally, the performance of a container terminal has been evaluated with numerous attempts at calculating and seeking to improve or optimize the operational productivity of cargo handling at the berth and container yard [7]. A conceptual framework for analysing the outcomes of potential competitive strategies and their expected 536

payoffs for container terminal operators in the container handling industry is presented in [8]. It is based on the integration of Bowley’s linear model of aggregate demand of product differentiation with Porter’s “Diamond” model. The authors developed ten competitive strategies for container terminal operators in order to present a theoretical scenario of two competing container terminal operators to exemplify the effectiveness of these strategies in terms of the number of TEUs handled, prices charged, and profits earned. The data envelopment analysis (DEA) methodology has been applied to the evaluation of container terminal performance in the literature. For example, in [9] the first work to advocate the application of the DEA technique to the terminals’ context is presented; it remains a purely theoretical exposition, rather than a genuine application. DEA window analysis using panel data relating to the eight container ports in Japan is conducted in [10]. In [1], DEA-CCR (Charnes Cooper and Rhodes model) and DEA-Additive models are used to analyse the efficiency of four Australian and 12 other international container ports. Applying DEA to estimate the relative efficiency of a sample of Portuguese and Greek seaports is given in [11]. DEA and stochastic frontier analysis have been used to study the efficiency of the world’s largest container ports and compare the results obtained in [3]. In [12], the relevance of DEA was analysed to estimate the productive efficiency of the container port industry. Available DEA panel data approaches were applied to a sample of 25 leading container ports and evaluated in [13].

*Corr. Author’s Address: Institute of Electronic Commerce and Modern Logistics, Dalian University, Dalian, China, lubo_documents@hotmail.com


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

In [14], five models of DEA were applied to identify trends in port efficiency of major container ports in the Asia-Pacific region. The impact of different groups on the efficiency of 28 container ports from 12 countries and regions in Asia was studied in [15]. In [16], the DEA method has been proven to be a suitable tool for evaluating performance with multiple inputs and outputs in respect to 77 global container ports. It was found that the number of berths and the capital deployed are the most sensitive measures impacting the performance of most container ports. In contrast with previous investigations, this study aims to explore the relationship among productivity indicators of container terminals and determine which factors have a stronger impact on productivity. Regression analysis (RA) is primarily applied to analyse the relationship between one dependent variable and several independent variables. Here, we apply and extend some of the papers in which authors introduced a methodology that incorporates a new variable into the regression analysis that captures the unique weighting of each comparable unit, [17] to [20]. In [17], the different models (DEA and RA) were used in various combinations to determine efficiency estimation and evaluation. The relative merits of DEA and RA in assessments of the comparative efficiencies of organizational units were considered in [18]. In [19], the major differences between RA and DEA were identified, and their appropriateness as a primary means for assessing relative efficiency in the context of regulation with an application to the water industry were evaluated. A methodology that includes a new independent variable, the comparable unit’s DEA relative efficiency, into the RA is applied in [20]. From an overall perspective, in this paper, the sensitivity analysis method provides a more appropriate benchmark for identifying which factors are more responsible for the fluctuations in productivity of container terminal by removing the input variables one by one, and then re-estimating the correlation between productivity and investment. Sensitivity analysis has been studied, utilizing either DEA or RA. However, the existing literature reveals a lack of empirical evidence in relation to the comparative effectiveness of sensitivity analysis in an application to the port industry. This paper aims to fill this gap by applying the two approaches to analyse container terminal productivity. The paper is structured as follows. The descriptions of DEA-CCR and RA with research procedure are given in Section 1. Section 2 provides the data collection, efficiency analyses and standardization of variables. Empirical results

and sensitivity analyses of DEA-CCR and RA are presented in Section 3. The major differences between RA and DEA-CCR with the relative merits are considered in Section 4. Finally, conclusions are drawn in Section 5. 1 RESEARCH METHODOLOGIES 1.1 Data Envelopment Analysis (CCR Model) In order to describe the research methodologies for determining critical factors at container terminals, the first step is to explain the proposed analyses. DEA is a non-parametric method of measuring the efficiency of a decision-making unit (DMU). It was applied into operations research, where authors introduced the CCR model [21]. The evaluation of container terminal efficiency using the DEA-CCR method begins by choosing appropriate input and output variables. Like in [3] but applying the expressions to container terminal efficiency, let the inputs be xk = ( x1k , x2 k , xMk ) ∈ R+M (in this case: berth length, quay crane, yard area, terminal crane and yard tractor) to produce outputs yk = ( y1k , y2 k , yNk ) ∈ R+N denoted as throughput per berth. The row vectors xk and yk forms the kth rows of the data matrices X and Y, respectively. Let λk = (λ1 , λ2 , λk ) ∈ R+K be a non-negative vector that forms the linear combinations of the K container terminals. Finally, let e = (1, 1, ..., 1) be a suitably dimensioned vector of unity values. In this study, the output-oriented DEA-CCR model seeks to maximize the proportional increase in output variables while remaining within the production possibility set. An output-oriented efficiency measurement problem can be written as a series of K linear programming envelopment problems, as shown in Eqs. (1) to (4) [3] and [12].

max U , (1) U ,λ

subject to

U y' − y 'λ ≤ 0, (2)

X 'λ − x ' k ≤ 0, (3)

λ ≥ 0 (DEA-CCR).

(4)

The combination of equations from Eqs. (1) to (4) form the DEA-CCR model. Because the CCR model gives a value of 1 for all efficient DMUs, it is unable to establish any further distinctions among the efficient DMUs.

Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals

537


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

In this study, the DEA-CCR model conducts the efficiency value analysis first when efficiency is less than 1; this means that the efficiency of the inputs and outputs variables are not appropriate and that it is necessary to decrease inputs or increase outputs. However, when the scale efficiency is less than 1 it indicates inefficiency, meaning that the operational scale is not achieving an optimal value and that the operational scale should be enlarged or reduced (based on the return to scale). In addition, it is possible to compare the technical efficiency value with the scale efficiency value, with the smaller value of the two, indicating the major cause of inefficiency. Furthermore, the slack variable analysis handles the utilization rate of input and output variables. It does this by assessing how to improve the operational performance of inefficient DMUs by indicating how many inputs to decrease, and/or how many outputs to increase, so as to render the inefficient DMUs efficient. Finally, the sensitivity analysis removes the input variables one by one, and then re-estimates the aggregate efficiency. This facilitates an overall understanding of which input variables are more critical for efficiency improvement [16]. In summary, the flow process of multiple DEA-CCR analyses can be depicted as shown in Fig. 1.

variables are held fixed. In restricted circumstances, RA can be used to infer causal relationships between the independent and dependent variables. However, it is also widely used for prediction and forecasting, [22] and [23]. For this study, multiple linear regression analysis is adopted to test the hypothesized inter-relationship between the dependent performance variable and the independent variables that relate to the productivity of container terminals, [4] and [5]. This study tries two methods: Enter and Backward, of which one is selected as the best method for results (see Fig. 2). First, it is necessary to estimate the model with all the predictors. Second, this research method follows the inputs of all of the selected variables, then estimates all the predictors, using the enter method and backward elimination method. The backward elimination method is used with settings at 5% significance levels. If all the regression coefficients are significant, the procedure stops; otherwise, those with the smallest significance will be eliminated from the model. The procedure finally stops when all the regression coefficients are significant. Finally, analysing and explaining the regression results, some tests (R-squared and Adjusted R-squared, ANOVA and coefficient analysis) are conducted on each independent variable and F-test for the overall regression (see Fig. 2).

Fig. 1. Flow process of DEA-CCR model

Fig. 2. Regression analysis model

1.2 Regression Analysis (RA) The second research method proposed in this study is RA. It is used to determine which among the independent variables are related to the dependent variable, and to explore the forms of these relationships. More specifically, it aids in understanding how the typical value of the dependent variable changes when any one of the independent variables is varied, while the other independent 538

The regression formulation based on parameters β0, β1, ..., βr–1 with independent variables x1, ..., xr–1 for n container terminals can be set up as follows: yi = β 0 + β1 x1 + ... + β r −1 xr −1 + ε r , r = 1, ..., n. (5) 1.3 Research Procedure The research procedure of this study is summarized in Fig. 3 where the application of the proposed methods

Lu, B. – Park, N.K.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

is shown. On the basis of the literature review, survey and interviews, we eliminate the duplication factors so the initial input/independent variables can be chosen. Then, we assume a database for the 28 East Asian container terminals for which we decide the input/ independent variables and output/dependent variables. After the collection of data, we apply DEA-CCR and RA by selecting the output/dependent variable. Consequently, sensitivity analysis is performed to compare the obtained results from both methods. In such a manner, we identify the critical factors at container terminals. This paper has two objectives. In the first, we propose the standardization of variables in order to constitute the accurate platform for the second objective. Moreover, definitions of variables should highlight the processes at container terminals and the factors impacting productivity. The second objective is to apply the two research methods to estimate the critical factors. This study represents a good starting point for identifying the sources of inefficiency and proposes services for improving operational performances at container terminals.

Fig. 3. Application of research methods

2 EFFICIENCY ANALYSES AND STANDARDIZATION OF VARIABLES 2.1 Data Collection The first step towards conducting the productivity of container terminals is to define the port performance characteristics based on the combination of

productivity factors (inputs) and then to highlight the outputs. Thus, many critical factors that relate to the terminal operation are to be considered. The sample comprises 28 East Asian major container terminals in 2008. The size of the sample is determined as a function of data availability and is more suitable for comparison [4] to [7]. The productivity indicator of this paper is assumed to be the annual throughput per berth. It is noteworthy that this study defines input/independent and output/dependent variables of each container terminal at the level of the berth. Accordingly, the standardization of variables is denoted as input/ independent variables per berth and output/dependent variables per berth. The data collection of the included container terminals is summarized in Table 1, [4] to [7]. The list of standardized input/independent variables includes: size of yard area (YA), number of quay cranes (QC), terminal cranes (TC), yard tractors (YT) and berth length (BL). Throughput is denoted as a standardized output/dependent variable. 2.2 Definitions of Variables Defining the objectives of container terminals is most important when determining the variables for efficiency measurement, [3]. In our paper, the objective is to maximize the throughput per berth; employment or any information on the equipment and count it as an input/independent variable. The variables should serve to reflect the existence of critical factors at container terminals. In this paper, the main objective is assumed to be the verification and maximization of terminal productivity. According to previously investigated papers ([4] to [7] and [24]), we chose the variables listed in Table 1 for both analyses: DEA-CCR and RA. On the basis of efficient use of infrastructure and facilities, the previously mentioned independent variables seem to be the most suitable factors to incorporate into the models as input variables. There are some other factors that can be included in this analysis, such as crane operating hours, equipment age and maintenance etc., but they have not been included in this study. The variables should reflect the actual objectives and process of container terminal production as accurately as possible. The importance, difficulties and potential impact of variable definition can be found in [24]. Before defining variables at container terminals, we should incorporate them into the model as input/ independent variables: length of berth, number of quay cranes, size of yard area, number of yard cranes,

Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals

539


StrojniĹĄki vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

Table 1. Data statistics Variables Terminal HIT - Hongkong International Terminal in port of Hong Kong Shekou - Shekou Container Terminal in port of Shenzhen COSCO - COSCO-HIT Container Terminal DPI - DPI Container Terminal BICT - Busan International Container Terminal, CS-4 - Chuanshan Container Terminal Phase 4 MTL - Hongkong Modern Terminal in port of Hong Kong PCTC - Pyeongtaek Container Terminal in port of Pyeongtaek NBCT - Ningbo Container Terminal in port of Ningbo HBCT - Hutchison Busan Container Terminal KX3-1 - KOREX Phase 3-1 Container Terminal in port of Gwangyang Chiwan - Chiwan Container Terminal KBCT - Korea Express Busan Container Terminal in port of Busan ACT - Asia Container Terminal HGCT - Hutchison Gamman Container Terminal DPCT - Dongbu Pusan Container Terminal NBSCT - Ningbo Second Container Terminal in port of Ningbo Hanjin - Busan Hanjin Container Terminal Yantian - Yantian Container Terminal in port of Shenzhen ICT - Incheon Container Terminal in Incheon Nansha - Nansha container Terminal in port of Guangzhou JUCT - Jeong-il Ulsan Container Terminal in port of Ulsan UTC - Uam Container Terminal in port of Busan DBE2-1 - Gwangyang DBE Phase 2-1 KIT2-2 - Korea International Terminal Phase 2-2 in port of Gwangyang HKTL - Hutchison Kwangyang Terminal GICT1 - Gwangyang International Container Terminal Phase 1 SGCT - Sun Gwang Container Terminal in Incheon Average

and number of yard tractors per berth as summarized in Fig. 4. On the other side, throughput per berth is in function of previously mentioned variables; as observed in the literature, this is the most important and widely accepted indicator of terminal productivity. It is the primary basis on which container terminals are compared. Container throughput is the most appropriate and analytically tractable indicator of the effectiveness of productivity. From the perspective of the first method, DEACCR provides the minimization of the use of inputs and maximization of the outputs. It is possible to acquire a variety of analytical results about the productivity efficiency for 28 container terminals. This procedure first identifies efficient container terminals and ranks the sequence of them, then finds the reasons the others are inefficient. This is followed by the identification of the potential areas 540

Yard Area/ Berth [ha] 15.0 13.2 20.0 17.3 9.3 13.8 14.9 16.8 14.9 25.2 17.5 16.7 14.3 22.9 12.9 12.3 21.0 9.6 37.2 12.3 7.4 6.9 8.4 20.7 21.0 21.0 17.6 12.2 16.2

Quay Crane/ Berth (number) 4.0 4.3 3.6 4.1 4.1 4.1 4.0 4.9 3.5 3.3 4.0 4.0 4.0 2.8 2.8 2.8 3.0 2.0 3.0 3.3 2.5 3.0 3.0 2.0 2.0 2.5 2.0 1.5 3.2

Terminal Crane/ Berth (number) 16.0 15.1 11.2 10.5 11.8 12.0 13.0 14.0 16.0 10.7 14.5 8.0 10.0 9.2 6.8 10.8 8.0 6.0 8.0 10.0 6.5 6.5 7.0 5.0 6.0 8.5 3.8 3.5 9.6

Yard Tractor/ Berth (number) 37.5 27.7 20.4 23.3 23.3 23.3 23.0 30.7 17.0 16.7 27.5 50.0 30.0 14.6 12.6 14.4 12.0 10.0 15.0 16.0 11.5 7.0 7.0 15.0 15.0 11.5 4.0 4.0 18.6

Berth Length/ Berth [m] 325 301 360 288 377 306 311 275 342 340 317 287 251 321 319 280 330 311 353 293 232 325 301 360 288 377 306 311 310

Throughput/ Berth (TEU) 877000 817143 700000 700000 692083 655556 650570 640000 632997 600000 600000 589000 588000 468353 420594 409165 403603 355991 333333 284868 279569 172448 169952 166371 124590 76120 51638 14772 445490

of improvement for inefficient terminals by applying the slack variable method. Finally, by comparing the efficiency scores between the container terminals, the results can identify which input or output variables are more critical to the models.

Lu, B. â&#x20AC;&#x201C; Park, N.K.

Fig. 4. Variables at container terminal


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

However, before utilizing the RA, it is necessary to be aware that all of the collected factors cannot be handled as dependent and independent variables. The possibility for duplication exists among the initial factors and non-linear correlation between dependent and independent variables. However, considering these circumstances first is required to establish the variables pre-processing model. 3 EMPIRICAL RESULTS 3.1 DEA-CCR Model Efficiency Results Solver-EMS version 1.3 software is employed to provide the DEA-CCR model results. It is used to analyse the efficiency of the chosen container terminals. Empirical results derived from the first model are presented in Table 2. On the basis of five input variables, efficiency scores are drawn in the first column while ranking is shown in the last column.

According to this table, nine container terminals are efficient because their efficiency scores are equal to 1. Otherwise, the results in bold mean the lowest value of five variables per terminal indicate they were relatively inefficient terminals. For instance, if we consider the Hanjin container terminal, the critical input is YT with a value of 0.69 and according to efficiency score, it ranks 14th place. There are three terminals that rank first place (Shekou, COSCO and CS-4) while the last ranking is presented for SGCT container terminal in Incheon port. Next is the identification of critical factors for each terminal. It is important to note that we have included four different ports in China and five in the Republic of Korea. Each port includes one or more container terminals as follows: five container terminals (ACT, COSCO, DPI, HIT and MTL) are located in the port of Hong Kong; seven (DPCT, BICT, HBCT, HGCT, KBCT, UTC and Hanjin) in the port of Busan; three (Chiwan, Shekou and Yantian) in the

Table 2. Empirical results derived from DEA-CCR model Variables Terminal HIT Shekou COSCO DPI BICT CS-4 MTL PCTC NBCT HBCT KX3-1 Chiwan KBCT ACT HGCT

Efficiency Scores 1 1 1 1 1 1 1 1 1 0.98 0.93 0.91 0.9 0.9 0.89

Yard Area/ Berth 0.96 1 1 1 1 1 1 1 1 0.98 0.93 0.86 0.9 0.87 0.87

Quay Crane/ Berth 1 1 1 1 1 1 1 1 1 0.98 0.93 0.91 0.9 0.9 0.88

Terminal Crane/ Berth 1 1 1 0.71 1 1 1 1 1 0.9 0.9 0.88 0.9 0.73 0.88

Yard Tractor/ Berth 1 1 1 1 0.83 1 0.97 0.95 0.92 0.91 0.77 0.9 0.84 0.9 0.85

Berth Length/ Berth 1 1 1 1 1 1 1 1 1 0.98 0.93 0.91 0.9 0.9 0.89

Ranking 3 1 1 13 10 1 2 4 5 6 11 7 9 12 8

DPCT

0.8

0.79

0.8

0.79

0.68

0.8

15

NBSCT

0.78

0.78

0.76

0.78

0.73

0.75

12

Hanjin Yantian ICT Nansha JUCT UTC DBE2-1 KIT2-2 HKTL GICT1 SGCT

0.75 0.74 0.66 0.66 0.65 0.52 0.49 0.35 0.33 0.18 0.1

0.7 0.71 0.66 0.66 0.65 0.5 0.49 0.35 0.33 0.18 0.1

0.75 0.73 0.66 0.66 0.65 0.52 0.49 0.35 0.31 0.18 0.1

0.72 0.69 0.66 0.61 0.65 0.5 0.4 0.35 0.3 0.18 0.1

0.69 0.74 0.43 0.63 0.38 0.47 0.49 0.19 0.33 0.15 0.06

0.75 0.74 0.66 0.66 0.65 0.52 0.49 0.35 0.33 0.18 0.1

14 18 16 20 17 19 22 21 23 24

Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals

14

541


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

3.2 RA Models Efficiency Results Regression models are statistical models that have the advantages of precisely analysing the relationship between one dependent variable and several independent variables, and identifying which factors have a stronger impact on the productivity of container terminal. Regression models also enable

the relationships to be explored. The Solver-SPSS software is employed to provide RA model results. Independent variables (the same as input variables for DEA-CCR model) are established by extracting the whole major factors that have a relationship with productivity and the normalization process, which consists of three steps. After that, the interrelationship between independent variables and the dependent variable (throughput per berth) can be established. Table 3. Distribution of critical factors derived from DEA-CCR for each port

Input variables

Busan

Shenzhen

Ningbo

Gwangyang

Incheon

Guangzhou

Ulsan

Pyeongtaek

Sum

Port

Hong Kong

port of Shenzhen; three (CS-4, NBCT and NBSCT) in the port of Ningbo; five (DBE2-1, GICT1, HKTL, KX3-1 and KIT2-2) in the port of Gwangyang; two (SGCT and ICT) in the port of Incheon; and one each (Nansha, JUCT and PCTC) in the ports of Guangzhou, Ulsan and Pyeongtaek. In Table 3, the number of critical factors for each port is given based on the results in bold from Table 2. We have performed an analysis of all nine ports separately to determine their characteristics. For example, Hong Kong’s container terminals have the most critical factor TC while no port has indicated BL as being a critical factor at container terminals. The statistics show that YT is the most frequent critical factor in 16 container terminals, which is 57% of the 28 considered terminals. TC is the second critical factor with 25% while QC occupies third place with 11% and YA with 7%. Similarly, according to [16], we state that the number of yard tractors per berth is the major capital input in port operations. The efficiency of a port is the next factor to determine. Therefore, the highest average efficiency is provided by the container terminal in Pyeongtaek, which yields 0.99, while the lowest efficiency is reached in the Incheon container terminal at 0.353. Port efficiency statistics for each port are shown in Table 4. The main gap is between container terminals Shekou, COSCO and CS-4 with an efficiency of 1 on one side and on the other SGCT, which has one tenth the efficiency. Therefore, according to DEA-CCR analysis, we can conclude that Hong Kong, Shenzhen and Ningbo have the most efficient container terminals.

YA/Berth QC/Berth TC/Berth YT/Berth BL/Berth

1 1 2 1 0

0 0 1 6 0

1 1 1 0 0

0 1 0 2 0

0 0 2 3 0

0 0 0 2 0

0 0 1 0 0

0 0 0 1 0

0 0 0 1 0

2 3 7 16 0

With respect to obtaining a regression model summary with the backward elimination method, the R-squared value or correlation coefficient, ranges from 0.812 to 0.843, and the adjusted R-squared, which is the adjusted coefficient of determination, ranges from 0.797 to 0.807. This indicates the independent variables YA, QC, TC, YT and BL explain the regression analysis model as having a good fit. Accordingly, the RA model shows remarkable statistical significance, as can be seen in Table 5. The independent variables have been excluded from Models 1 to 4 in the following order: YA from Model 2, BL and YA from Model 3, and QC, BL and YA from Model 4, which suggests that YT and TC are the most important variables. The standard error of the estimate is also given, which indicates a measure of the accuracy of predictions. Other statistics have also been used to evaluate the critical factors at a container terminal. Using ANOVA analysis, we have also applied four models

Table 4. Port efficiency statistics derived from DEA-CCR for each port Port Statistics No. of cont. term. No. of effic. port Maximum value Minimum value Mean value Standard Deviation

542

Hong Kong

Busan

Shenzhen

Ningbo

Gwangyang

Incheon

Guangzhou

Ulsan

Pyeongtaek

5 1 1 0.71 0.9576 0.0447

7 0 1 0.47 0.8106 0.1649

3 1 1 0.69 0.8713 0.1320

3 1 1 0.73 0.9147 0.1270

5 0 0.93 0.15 0.4352 0.2869

2 0 0.66 0.06 0.353 0.3960

1 0 0.66 0.61 0.644 -

1 0 0.65 0.38 0.596 -

1 0 1 0.95 0.99 -

Lu, B. – Park, N.K.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

with the elimination method and calculated the sum of squares and mean square values. From Table 6, the significant probability is always equal to 0.00 in applying independent variables removed in an orderly manner (as in the case of Table 5). In contrast, the F-test analyses the overall regression of the model and reflects satisfied results, which are also summarized. ANOVA analysis results show that the most significant independent variables are YT and TC.

R

R-Squared

1 2 3 4

0.918(a) 0.917(b) 0.910(c) 0.901(d)

0.843 0.840 0.828 0.812

Table 7. Coefficients analysis results Model

1

Table 5. Regression model summary Model

which suggest that variables YT and TC have the main impact on reaching throughput per berth. The results are summarized in Table 7.

Adjusted R-Squared 0.807 0.812 0.806 0.797

Std. Error of the Estimate 108471.3905 107046.0609 108838.2806 111440.0392

Note: (a) Predictors: (Constant), BL, YT, YA, TC, QC; (b) Predictors: (Constant), BL, YT, TC, QC; (c) Predictors: (Constant), YT, TC, QC; (d) Predictors: (Constant), YT, TC

2

3

Table 6. ANOVA analysis results Model Regression 1 Residual Total Regression 2 Residual Total Regression 3 Residual Total Regression 4 Residual Total

Sum of Squares 1.390E12 2.589E11 1.649E12 1.386E12 2.636E11 1.649E12 1.365E12 2.843E11 1.649E12 1.339E12 3.105E11 1.649E12

Mean Square 2.780E11 1.177E10

F-test

Sig.

23.6

0.000a

3.464E11 1.146E10

30.2

0.000b

4.549E11 1.185E10

38.4

0.000c

6.693E11 1.242E10

53.9

4

(Constant) YA QC TC YT BL (Constant) QC TC YT BL (Constant) QC TC YT (Constant) TC YT

Ustd. Coefficients -75592.627 0.233 92825.355 34056.108 5679.013 -765.770 -65849.041 85529.610 34277.077 6010.075 -632.240 -232134.707 79042.594 32715.465 5905.614 -133732.596 43582.420 8693.668

Beta value 0.059 0.325 0.492 0.238 -0.150 0.300 0.496 0.252 -0.120 0.277 0.473 0.248 0.630 0.365

T-test

Sig.

-0.490 0.632 1.705 3.260 1.774 -1.470 -0.434 1.628 3.327 1.929 -1.350 -2.590 1.486 3.143 1.864 -2.180 5.745 3.327

0.629 0.534 0.102 0.004 0.090 0.156 0.668 0.117 0.003 0.066 0.192 0.016 0.150 0.004 0.075 0.039 0.000 0.003

Finally, results from regression show that, in the scope of the independent variables, the numbers of YT and TC have a strong impact on berth productivity. The regression formulation, which is composed of the number of yard tractors and terminal cranes is: Y = −133732.596 + 43582.420 X 1 + 8693.668 X 2 , (6)

0.000d

Note: (a) Predictors: (Constant), BL, YT, YA, TC, QC; (b) Predictors: (Constant), BL, YT, TC, QC; (c) Predictors: (Constant), YT, TC, QC; (d) Predictors: (Constant), YT, TC

The same four models have been analysed using coefficient analysis. Applying that and the T-test, which tests the single variable significance of the model, also reflect satisfied results. It shows that the BL variable has a negative impact on the dependent variable (throughput per berth). Similarly, the standardized coefficients’ Beta value appears minus the value for the variable BL. It means that, as throughput per berth increasing, the independent variable berth length would not be appropriate to handle this increased throughput. This implies that in this case BL is a critical factor for container terminals. However, the significant probability is decreasing,

where X1 is the number of terminal cranes (TC) while X2 represents the number of yard tractors (YT). 4 IMPLICATIONS AND COMPARISON The empirical results from DEA-CCR and RA analyses show that the input/independent variables for determining the critical factors at container terminals have a great contribution to terminal productivity. It is clear that the significance of critical factors for productivity is similar in respect to the chosen container terminals. The results indicate the first implication that the most important of factors are TC and YT. Table 8 shows the order of significance for the critical factors derived from both analyses. Obviously, both models affirm the importance of a number of terminal cranes and yard tractors. Another implication of the results achieved in the analyses is that the facilities of a container terminal,

Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals

543


StrojniĹĄki vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

such as YA and BL, appear to exhibit lower levels of significance than the equipment. Although it is shown that huge YA or long BL do not always imply the higher productivity of terminals, it indicates that if the number of handling equipment like TC and YT were increased, the container terminal could improve its productivity. Table 8. Order of significance for critical factors derived from DEACCR and RA analyses Models Input/ independent variables YA QC TC YT BL

DEA-CCR order of sign. 3 4 1 2 5

RA order of sign. 5 3 1 2 4

Output/ dependent variable

Throughput per berth

In relation to the DEA-CCR model, the YA variable is in third place, with QC following. BL represents the input variable with the lowest level of significance. According to the coefficient results of RA, in the scope of the independent variables, the number of YT, TC and QC have a stronger impact on the throughput per berth. The YA also has a positive correlation, but in contrast, berth length has a negative correlation.

0.7368, 0.7075 and 07686 for TC, YT and all input/ independent variables, respectively. Finally, the Beta value gives the lowest results with 0.6300, 0.3650 and 0.1928 for the same variables, respectively. According to the results of both analyses, the following conclusions can be drawn. The higher the numbers of TC and YT, the higher the berth productivity will be. With respect to BL which has a negative correlation to throughput per berth, the increase of BL leads to a decrease in productivity. In contrast, YA differs from one case to another. For example, the yard area of the COSCO container terminal in Hong Kong is 20 ha and the throughput per berth is 700,000 TEU while the yard area of the HBCT container terminal in Busan is 25.2 ha, and the throughput per berth is 600,000 TEU. In contrast with the container terminals of the Republic of Korea, the yard area of Chinese ports is much larger. However, the reason the YA is not selected as an effective independent variable is that most container terminals use an off-dock container YA to improve yard utilization and rapid response to customers. The same assumptions are drawn with the average number of QC. Again, if comparing COSCO and HBCT container terminals, the average number of QC that are used at terminals for the first one is 3.6 while for the second is 3.3. COSCO reached 700,000 TEU per berth while HBCT reached 600,000 TEU per berth in 2008. It is a massive difference in total throughput even though the average number of QC at both container terminals is similar. This means that the number of QC is not related to the throughput directly. 5 CONCLUSIONS

Fig. 5. Comparison of coefficients of efficiency for input/ independent variables (two different models)

Next, we compared the coefficients of efficiency for TC, YT and all input/independent variables in respect to different models (DEA-CCR, R-Squared, Adjusted R-Squared and Beta value as models of RA). The trend lines are shown in Fig. 5. The results for R-Squared imply the highest efficiency of container terminals with 0.8120, 0.8120 and 0.8430 for TC, YT and all input/independent variables, respectively. Next are the Adjusted R-Squared results of efficiency with 0.7970, 0.7970 and 0.8070 for the same variables, respectively. DEA-CCR proposes the efficiencies of 544

This study proposes DEA-CCR and RA models to assess the sensitivity analysis for identifying critical factors of productivity for container terminals. The sensitivity analysis shows that the input/independent variables for determining the critical factors at container terminals make a great contribution to terminal productivity. The sensitivity of an input or output measure is defined as the range of changes of the input or output measure to improve the efficiency of the frontier. The most sensitive factors are denoted as critical for considered container terminals. The empirical results indicated that the most important of them are TC and YT in the container terminalsâ&#x20AC;&#x2122; efficiency evaluation of the 28 chosen terminals in nine ports. These results have provided useful information indicating how relatively inefficient container terminals can improve their efficiency.

Lu, B. â&#x20AC;&#x201C; Park, N.K.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

As a benchmarking analysis, this study provides the two different approaches to efficiency measurement of container terminals, DEA-CCR and RA and compares their efficiency. These methods are applied to the same data set, and sensitivity analysis is conducted to compare the results of critical factors at container terminals. The productivity of a container terminal is influenced by a range of factors, which were removed one by one, and the correlation between productivity and investment then re-estimated. Furthermore, another major objective of this study was to compare the results obtained from applying DEA-CCR and RA. It is important to note some advantages of DEA-CCR over RA being as follows: DEA-CCR measures performance against efficient rather than average performance, DEA-CCR offers more accurate estimates of relative efficiency because it is a boundary method; DEA-CCR normally yields more accurate targets because it is a boundary method, and so on. However, some advantages of RA over DEA-CCR may be summarised as: RA offers a better predictor of future performance at the collective DMU level if it is assumed inefficiencies cannot be eliminated, RA offers the facility to estimate confidence intervals for point estimates, and RA could yield better estimates of individual maximum (minimum) levels where outputs (inputs) can vary independently of one another, among others [17] to [20]. Moreover, there are also some limitations of the research. Several factors of terminal productivity are not included in the variables, such as the number of vessel arrivals, manpower, the service time of vessels, and so on. However, the acquisition of data is quite difficult, and the combinations of independent/input variables and dependent/output variables that are utilized in this study also have to be adjusted. In further research, this study will enlarge the number of terminals and variables. The individual terminal simulation model will represent the direction for future investigations. 6 ACKNOWLEDGEMENTS The study is supported by Liaoning Provincial Education Department (LPED) and a Tongmyong University Research Grant. The study is also financially supported by the LPED Project (W2012246), LPE and Science “Twelfth Five” Planning 2012 Project (JG12DB345), MOMAF (Ministry of Maritime and Fishery Affairs) Republic of Korea, through the “Evaluation of Port Capacity” project of the University.

7 REFERENCES [1] Tongzon, J., Heng, W. (2005). Port privatization, efficiency and competitiveness: some empirical evidence from container ports. Journal of Transportation Research A: Policy and Practice, vol. 39, no. 5, p. 405-424, DOI:10.1016/j.tra.2005.02.001. [2] Le-Griffin, H.D., Murphy, M. (2006). Container port productivity: experiences at the ports of Los Angeles and Long Beach. METRANS 2006 Conference Proceedings, p. 1-21. [3] Cullinane, K., Wang, T.F., Song, D.W. (2006). The technical efficiency of container ports: Comparing data envelopment analysis and stochastic frontier analysis. Journal of Transportation Research Part A, vol. 40, p. 354-374, DOI:10.1016/j.tra.2005.07.003. [4] Park, N.K., Kim, J.Y., Lu, B. (2008). Critical factors for container port productivity. Proceedings of International Association of Maritime Economists, p. 262-272. [5] Park, N.K., Lu, B. (2010). A study on productivity factors of Chinese container ports. Journal of Korean Navigation and Port Research, vol. 34, no. 7. p. 559566, DOI:10.5394/KINPR.2010.34.7.559. [6] Lu, B., Park, N.K. (2010). Operational performance evaluation of Korean major container terminals. Journal of Korean Navigation and Port Research, vol. 34, no. 9, p. 719-726, DOI:10.5394/ KINPR.2010.34.9.719. [7] Lu, B., Huo, Y.F., Park, N.K. (2012). A study on logistics requirements potential model based on spatial economics. Proceedings of XX. International Conference on Material Handling, Constructions and Logistics, p. 351-356. [8] Yap, W.Y., Lam, J.S.L., Cullinane, K. (2011). A theoretical framework for the evaluation of competition between container terminal operators. Singapore Economic Review, vol. 56, no. 4, p. 535559, DOI:10.1142/S0217590811004456. [9] Roll, Y., Hayuth, Y. (1993). Port performance comparison applying data envelopment analysis (DEA). Maritime Policy and Management, vol. 20, no. 2, p. 153-161, DOI:10.1080/03088839300000025. [10] Itoh, H. (2002). Efficiency changes at major container ports in Japan: A window application of data envelopment analysis. Review of Urban and Regional Development Studies, vol. 14, no. 2, p. 133-152, DOI:10.1111/1467-940X.00052. [11] Barros, C.P., Athanassiou, M. (2004). Efficiency in European seaports with DEA: Evidence from Greece and Portugal. Maritime Economics and Logistics, vol. 6, no. 2, p. 122-140, DOI:10.1057/palgrave. mel.9100099. [12] Cullinane, K.P.B., Wang, T.F. (2007). Data envelopment analysis (DEA) and improving container port efficiency. Brooks, M., Cullinane, K.P.B. (eds.) Devolution, port governance and port performance,

Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals

545


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 536-546

Research in Transportation Economics, vol. 17, p. 517-566, Elsevier, Amsterdam. [13] Cullinane, K., Wang T.F. (2010). The efficiency analysis of container port production using DEA panel data approaches. OR Spectrum, vol. 32, no. 3, p. 717738, DOI:10.1007/s00291-010-0202-7. [14] Lin, L.C., Tseng, C.C. (2007). Operational performance evaluation of major container ports in the Asia-Pacific region. Journal of Maritime Policy and Management, vol. 34, no. 6, p. 535-551, DOI:10.1080/03088830701695248. [15] Wu, J., Yan, H., Liu, J. (2009). Groups in DEA based cross-evaluation: An application to Asian container ports. Maritime Policy & Management, vol. 36, no. 6, p. 545-558, DOI:10.1080/03088830903346095. [16] Wu, J., Yan, H., Liu, J. (2010). DEA models for identifying sensitive performance measures in container port evaluation. Maritime Economics & Logistics, vol. 12, no. 3, p. 215-236, DOI:10.1057/ mel.2010.6. [17] Bowlin, W.F., Charnes, A., Cooper, W.W. (1985). Data Envelopment Analysis and regression approaches to efficiency estimation and evaluation. Annals of Operations Research, vol. 2, p. 113-138, DOI:10.1007/BF01874735. [18] Thanassoulis, E. (1993). A comparison of Regression Analysis and Data Envelopment Analysis as alternative methods for performance assessments. Journal of Operational Research Society, vol. 44, no. 11, p. 1129-1144, DOI:10.1057/jors.1993.185. [19] Cubbin, J., Tzandikas, G. (1998). Regression versus data envelopment analysis for efficiency

546

measurement: an application to the England and Wales regulated water industry. Utilities Policy, vol. 7, no. 2, p. 75-85, DOI:10.1016/S0957-1787(98)00007-1. [20] Klimberg, R.K., Lawrence, K.D. Yermish, I., Lal, T., Mrazik, D. (2009). Using regression and Data Envelopment Analysis (DEA) to forecast bank performance over time. In Lawrence, K.D., Kleinman, G. (eds.) Financial Modeling Applications and Data Envelopment Applications (Applications of Management Science, vol. 13, Emerald Group Publishing Limited, p. 133-142. [21] Charnes, A., Cooper, W.W., Rhodes, E. (1978). Measuring the efficiency of decision making units, European Journal of Operational Research. vol. 2, no. 6, p. 429-444, DOI:10.1016/0377-2217(78)901388. [22] Kleinbaum, D.G., Kupper, L.L., Muller, K.E, Nizam, A. (2007). Applied regression analysis and other multivariable methods. Thomson Higher Education, Belmont. [23] Cicek, A., Kivak, T., Samtas, G., Cay, Y. (2012). Modeling of thrust forces in drilling of AISI 316 stainless steel using artificial neural network and multiple regression analysis. Strojniški vestnik – Journal of Mechanical Engineering, vol. 9, no. 7-8. p. 492-498, DOI:10.5545/sv-jme.2011.297. [24] Lu, B., Wang, X.L. (2012). Comparative studies on efficiency evaluation of Chinese and Korean major container terminals. Journal of Advances in Information Sciences and Service Sciences, vol. 56, p. 645-654.

Lu, B. – Park, N.K.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2012.933 Special Issue, Original Scientific Paper

Received for review: 2012-12-27 Received revised form: 2013-05-14 Accepted for publication: 2013-05-22

Ports Sustainability: A life cycle assessment of Zero Emission Cargo Handling Equipment Vujičić, A. – Zrnić, N. – Jerman, B. Andrija Vujičić1* – Nenad Zrnić2 – Boris Jerman3

1 Dunav Insurance Company, Serbia of Belgrade, Faculty of Mechanical Engineering, Serbia 3 University of Ljubljana, Faculty of Mechanical Engineering, Slovenia 2 University

The goal of this paper is to present and evaluate the latest trends in the cargo handling equipment (CHE) industry, aimed at mitigating the environmental impact of container terminal operations and contributing to the sustainability of ports. The most common machines for handling containers are described and dealt with separately, with a focus on electric CHE, usually referred to as ‘zero emission’ CHE. In a separate chapter, recommendations on methodologies suitable for investigation of the environmental footprint of CHE without on-site measuring are reviewed. The life cycle assessment (LCA) methodology as a tool for comparison of conventional and ‘zero emission’ technology is emphasised with examples. The conventional diesel rubber-tired gantry (RTG) crane and utility tractor rig (UTR) are compared with an electric RTG and UTR using an LCA approach. Keywords: port container terminals, cargo handling equipment, zero emission, rubber-tired gantry, utility tractor rig, life cycle assessment

0 INTRODUCTION With more than 90% of global freight moving by containers, container transport industries have an immense influence and role in the global economy. Ports are a core component in the international supply chain and play an enormous role in regional economies; regional development is directly related to the ability of ports to adapt to emerging challenges. The economic strength behind ports and container terminals unfortunately comes with a heavy environmental burden. The growing port activities and the densely populated cities where most ports are located, combined with already pollution-saturated air and water, are imposing threats to public health and environment in general. Many ports today are considered to be the largest sources of air pollution in coastal cities and awareness of the necessary action for the reduction of pollution has become the matter of public concern. Related research regarding this topic can be found in the work of Cannon [1] and [2]. It is also necessary to note that the environmental footprint of ports is rising to the top of the port authorities’ agenda at the same moment an economic downturn has already exerted stress on container operations struggling to remain cost effective. Emissions in ports come from several different sources. The main pollutants at ports are vessels, harbour craft, cargo-handling equipment (CHE) and trains and heavy vehicles within or near the port. Each source of port emissions is treated using different solutions but with the same goal of reduced environmental impact. From recommendations to

solve the problem of cold ironing of vessels by Nikitakos [3] to the ‘green’ intralogistics investigated by Kartnig et al. [4], the ideal is the zero emission port. The objective of this paper is to investigate the background of the ‘zero emission’ concept. The tool used for investigation is the life cycle assessment (LCA) methodology. The two most common machines found at the container terminals, the rubber tired gantry (RTG) crane and the utility tractor rig (UTR), have been selected as an example for a comparison of conventional and ‘zero emission’ technologies. 1 CARGO HANDLING EQUIPMENT Although the greenhouse gas (GHG) emissions from ports that have adopted strategies to mitigate environmental impact have some effect, CO2 emissions from CHE operations are still rising, according to latest reports from Starcrest [5]. The reasons could be found in the increased rate of container handlings, the modest application of ‘clean’ technologies, and often weak or absent emission regulation policies for nonroad diesel engines found in CHE. In port container terminals, containers are ferried around using specially designed cranes and special forklifts, tractors or trucks. The containers are lifted from a marine vessel by a crane and later moved or picked by other crane, handler or forklift. The container can also be transported around the terminal using UTR, which is also known as a ‘yard truck’. Most common in CHE fleets are UTRs, followed by forklifts, handlers and gantry cranes. CHE is

*Corr. Author’s Address: Dunav Insurance Company, Blagoja Parovica 19, Beograd, Serbia, a.vujicic@dunavauto.co.rs

547


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

conventionally powered by internal combustion engines that are (according to the reports of Starcrest [6]) powered by diesel engines with non-road emission standards in 95% of cases. Due to the fact that handling loaded containers is an energy-intensive function, CHE is often considered to be one of the most significant sources of air pollution caused by terminal operations. In this paper, the focus is on the RTG cranes and UTRs, since they are the most common pieces of CHE, accounting for over 55% of all CHE in container terminals [6]. RTGs are dependent on the support of UTRs for quick container transport across the terminal; a combined evaluation of their environmental impact and operating costs is a common approach in terminal planning, also recommended by Böse [7].

Today E-RTG’s disadvantages have been overcome with the cable reel and most recently with a drive-in conductor bar solution with a collector trolley that automatically engages and disengages. Although the main disadvantage remains, i.e. the need to adapt the terminal for electrification, the fact is that with the latest solutions, the environmental advantages of E-RTG are again relevant. For 90% of operating time, E-RTG cranes use only electricity and use their diesel engines for the remaining 10% of time, during block changes and maintenance. The manufacturers of E-RTG promise a massive potential for CO2 reduction.

1.1 Rubber Tired Gantry Crane In port container terminals, RTG cranes are used for the movement of shipping containers, once they are placed on to the distribution channels from a vessel. The cranes are powered by a diesel generator set, consisting of a diesel engine coupled with an alternator. An RTG crane is capable of moving containers weighing up to 50 tonnes at a rate of 20 moves per hour. Since it is one of the largest machines on tires in the world, powered by a large non-road diesel engine, turning it into an eco-friendly machine is a challenging task. A conventional diesel generator set provides electrical power for the hoist, trolley, and gantry electric motors, as well as for the routine demands of the crane. Utilizing this type of power system on a RTG allows the crane to move independently throughout the container terminal as required by daily port operation. The freedom of movement and the high peak power demand for the hoist motor consume a large amount of fuel and emit significant amounts of GHG. Today, a variety of technologies and systems are available to reduce fuel consumption and emissions, and improve overall RTG efficiency. They include technologies such as variable-speed generators, flywheel energy storage, hybrid RTGs with regenerative breaking and super or ultra-capacitor technology and electrified ‘zero emission’ cranes. Most of them are available as retrofits for conventional cranes, but also as new manufactured RTG options. In the past, electrified RTG (E-RTG) cranes were often avoided due to complicated electric power cable arrays, reduced movability and limited flexibility. 548

Fig. 1. Rubber Tired Gantry crane

1.2 Utility Tractor Rigs Utility tractor rigs (UTR; often also called terminal tractors, yard trucks or hostlers) are heavy-duty offroad single cab trucks designed for moving cargo containers. They are by far the most common type of CHE used at container terminals [6]. The UTR is connected to a trailer, which it uses to transport containers. In a typical operation, a container is loaded onto a trailer by a piece of CHE, such as a crane, handler or RTG. The UTR then tows the trailer with the container to a destination within the terminal where the container is unloaded by another piece of CHE. In each use of a crane or handler, the container is ferried around the yard using a UTR. Therefore, UTRs are the most influential category of equipment in terms of fuel consumption and air emissions [6]. In order to mitigate the environmental footprint of UTRs, a wide range of technologies is available with high possibilities of mainstream application. The most common solutions tested at container terminals make use of alternative fuel options (LPG, CNG and biodiesel), hybrid (diesel-electric and dieselhydraulic) and electric drives, often regarded as ‘zero emission’ UTR.

Vujičić, A. – Zrnić, N. – Jerman, B.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

Fig. 2. Utility tractor rig

The long idling periods and stop-and-go movements with bursts of accelerations of UTRs are significant emission sources followed by noise pollution adjacent to affected neighbourhoods. Fortunately, the UTRs’ duty cycles are highly predictable and suitable for the application of emission reduction technologies. This generally due to the high levels of idle with up to 50% of total operating time. The major issue of idling is efficiently resolved with hybrid and electric UTRs, where no energy is used during stops, reducing both exhaust emissions and noise pollution. One of the latest trends in the CHE industry is the fully electric UTR, which is also referred as ‘zero emission’ equipment. As with most electric on-road vehicles, the success of this concept is to a significant extent dependant on the efficiency of its batteries. Electric UTRs are available with lead, nickel, and lithium-ion battery packs; . Depending on battery pack sizes, which for UTRs range from 150 to 300 kWh, autonomy could be sufficient for two shift operations. 2 EVALUATION OF CARGO HANDLING EQUIPMENT FOOTPRINT In order to understand the environmental impact of container terminals, various models and tools that quantify the relevant emissions can be used or developed. Each model can vary significantly in terms of both complexity and accuracy, as well as time and resources on the other. Furthermore, some methodologies and practices are more suitable for vessels or rail machines than CHE. The non-modelling approach to create an emission inventory of CHE is to directly measure emissions or energy consumption. Although it could be considered the most accurate way, it is also the most expensive and time demanding, and can only be performed after the infrastructure is in place. Direct

emissions measurement, thus negates early stage planning process, but is more suitable for establishing the baseline inventory. Therefore, the modelling is more appropriate as a preventive approach, and as support for decision making. The complexity of modelling methodologies can also vary depending on intended use and users, and can also be time and resource consuming if a detailed and validated model is wanted, according to Liu et al. [8]. A simple model for predictions of emissions and energy consumption at terminals can be found in the research of Geerling and Duin [9]. Regardless of which modelling approach is chosen, it enables the prediction of the emissions of any source at port without actually ever visiting facility. This can also be used for the comparison of different types of CHE. The emissions are estimated with developed models and/or inventories for representing CHE technical data, such as rated power, model year, fuel type, annual hours of operation, load data, etc. The average emissions per engine for each equipment type are then multiplied by the number of operating shifts or hours, further by the total number of engines in the CHE group and annual emissions for certain equipment. The downside of these approaches is that any uncertainty in the baseline parameters can eventually lead to significant uncertainty in the final results of estimated CHE emissions. This is of considerable importance, especially when comparison of any type of CHE is made, since even the slightest aberration in early modelling can result in favouring one piece of equipment over another. 2.1 The Life Cycle Assessment as Comparison Tool Certain pieces of handling equipment have advantages over others and certain terminal configurations (size, average number of handling operations, etc.) are more suitable for certain types of CHE; nevertheless, regarding overall port sustainability, it is necessary to determine which solution or equipment has the smaller environmental footprint [10]. With regards to tools for quantifying the environmental impact of products or process, certain methodologies have advantages over others, but for the sustainability assessment, one of the most appropriate ways is to carry out an LCA. Investigations regarding sustainability assessment via LCA can be found in the research of Pušavec and Kopač [11]. The LCA methodology outlined in ISO 14040 [13] is a quantitative tool for the assessment of the environmental impacts of products and services.

Ports Sustainability: A life cycle assessment of Zero Emission Cargo Handling Equipment

549


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

It is a systematic approach for analysing the entire life cycle stages from material extraction through manufacturing, use and eventually disposal or (preferably) recycling. Therefore, it is often called a ‘cradle to grave’ analysis. The application of LCA studies in the field of material handling has caught the attention of the engineering researchers in recent years, but unfortunately little literature on the subject of assessment of CHE is available [14] and [15]. Despite the modest number of published works on assessments of CHE via LCA, there are several facts indicating that this methodology could be used for this purpose. First, it offers a consistent comparison of the production phase of a machine, via a ‘cradle to gate’ assessment, the propulsion system and even an alternative fuel solution with a ‘well-to-tank’ assessment, since an LCA can address the evaluation of the environmental burden of fossil fuels from extraction, refining, transportation, distribution, as well as electricity production. Chester and Horvath [14] often used an LCA for comparison of well-totank emissions related to use of petrol or diesel and electric energy in transport. Secondly, an LCA offers evaluations based on inventory data without the need for on-site measurements. This makes it a useful tool for both designers and industry, as well as for stakeholders and policy makers. 3 EXAMPLES OF LIFE CYCLE ASSESSMENT OF CARGO HANDLING EQUIPMENT For the purpose of this paper, a comparative LCA of conventional over electric RTG and UTR is conducted. This approach is appropriate when the goal of study is to identify significant issues in each phase of the life cycle of products or production systems that have substantial similarities. Since the chosen models are identical in function and the main difference is the change from diesel to electric technology, only the LCA methodology can respond to the sustainability sensitivities. 3.1 Life Cycle Assessment Assumptions In order to simplify the LCA comparison and avoid possible data uncertainties, certain assumptions and limitations have been made. This is in accordance with the LCA standards and practice, since the intended purpose of the study is solely scientific research and not a commercial one. 550

The assumptions are made with respect to the operational life of CHE and port terminal experience in order to avoid significant virtualization in the modelling approach. The conventional diesel model of RTG crane and UTR are set as basic models, while other two electric models share over 95% of the same structure and components (gantry, chassis, wheels, cabin, etc). With the adoption of this modelling principle, most of inventory base of conventional diesel RTG and UTR can be used for evaluation of electric models. The assessment is divided into life-cycle three phases. First is the production or ‘upstream’ phase, called ‘cradle to gate’, which addresses all processes from material extraction and depletion through parts production and model assembly and finally distribution to port terminal. The second is the ‘use’ phase, referred as ‘gate to grave’ is the phase with the longest and largest environmental impact. This phase is also addressed with respect to the ‘well-totank’ principle. The final phase is the scrapping and disposal or recycling of RTG and UTR, called ‘end of life’. The ‘cut-off’ principle is applied in the manufacturing phase, enabling authors to leave out parts, components and processes that weigh less than 5% of the total mass of chosen model, or have an insignificant contribution to overall environmental impact. The Functional Unit (FU) is also defined according to the LCA practice. The FU is set as one operating hour of models at a container terminal, separately described for RTG crane and UTR. The container terminal location and electric power grid mix are chosen to be an EU port and use an EU-25 power grid mix. This selection is made with respect to software used for LCA study. The LCA is conducted using holistic balancing software (‘GaBi’, from the German Ganzheitlichen Bilanzierung) developed by PE International, which is (according to a survey by Cooper and Fava [16]) the most common LCA tool used by over 58% of surveyed practitioners. In order to provide the most accurate data to a study, the GaBi software inventory data on ‘end of life’ is compared with results from Zackrisson et al. [17] and Majeau-Bettez et al. [18] who researched lithium-ion battery disposal and recycling.

Vujičić, A. – Zrnić, N. – Jerman, B.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

3.2 Life Cycle Assessment Comparison of Rubber Tired Gantries The main features of a conventional RTG crane model for an LCA study are the following: total weight of a crane is 115 tonnes; lifting capacity is 40 tonnes; diesel engine displacement is 12 litres; peak power is 300 kW with 600 kVA AC/DC generator. Once the results of ‘upstream’ phase of conventional RTG crane model are determined, they are used for electric RTG. The RTG electric solutions are available as retrofits and brand new cranes. The diesel engine is not removed since it is necessary for movement of crane outside of the conductor bar yard. Therefore, the ‘upstream’ phase of an E-RTG crane differs from a conventional one, only with regard to the impact results of add-on kit parts. The environmental impact of the conversion of container terminal in order to serve the needs of E-RTG is not taken into account, although the contribution to the overall score of electrification is not in doubt. The FU for the RTG is defined as one operating hour at a container terminal, where 50% of time is spent for crane movement, 30% for lifting operations and 20% for spreader movement. The annual operation time for RTG crane is 5,000 working hours. The life cycle is 15 years. 3.3 Life Cycle Assessment Comparison of Utility Tractor Rigs The main features of the adopted UTR models for LCA study are the following: net and gross weight is 8 and 70 tonnes, respectively; diesel engine displacement is 8 litres with 200 kW peak power. The main features of an electric UTR are an electric motor with 240 kW of power and 140 kWh with 320 V lithium-ion battery pack.

The base model of UTR used for the ‘upstream’ phase is a conventional diesel UTR. The electric UTR model uses the entire inventory of conventional UTR but, without data for a diesel engine and related components, but with data added for an electric motor and lithium ion battery pack. The FU for the UTRs is defined as one operating hour at a container terminal (yard work), where 40% of a time is spent in idling, 35% of time is related to lower load and 25% with high load. The annual operation time for UTR is 3,500 working hours. The life cycle is 10 years. 4 RESULTS The obtained emissions inventory and mandatory elements of conventional and electric RTG cranes and UTRs are included in the selection of impact categories in accordance with ISO 14040-44 [13]. Environmental impacts of life cycles are classified and characterised via a problem-oriented approach (midpoint), and the LCIA (life cycle impact assessment) stage is addressed with most common LCIA methods used in Europe and the US. The first method, the Centre for Environmental Sciences Leiden (CML) has been developed by the Institute of Environmental Sciences from Leiden and is, according to Azapagic [19], the most widelyused and is considered to be the most complete methodology. It uses primarily European data to derive its impact factors. The impact categories for the global warming potential (GWP) and ozone layer depletion are based on IPCC (Intergovernmental Panel on Climate Change) factors. In this paper, the latest version of the CML 2001 is used. The second impact assessment methodology TRACI (Tool for the Reduction and Assessment of Chemical and Other Environmental Impacts) is developed by the US EPA (United States

Table 1. Life Cycle Impact Assessment of Rubber Tired Gantries - CML method CML 2001

Impact / [unit]

Phase

RTG type diesel electric diesel electric diesel electric diesel electric

Cradle to Gate Gate to Grave End of Life Cradle to Grave

Global Warming Potential [kg CO2 eq.] 335,830.77 344,437.66 6,017,597.26 1,462,726.14 -15,843.96 -13,625.81 6,337,584.07 1,793,537.99

Acidification Potential [kg SO2 eq.] 1,766.85 1,812.51 78,635.93 12,520.48 -200.51 -172.43 80,202.27 14,160.56

Eutrophication Potential [kg Phos. Eq.] 96.93 98.99 13,547.53 1,583.79 0.00 0.00 13,644.46 1,682.78

Ozone Depletion Potential [kg R11 eq.] 0.06 0.06 0.00 0.21 0.00 0.00 0.06 0.26

Ports Sustainability: A life cycle assessment of Zero Emission Cargo Handling Equipment

Photochemical Ozone Creation [kg Ethene eq.] 100.11 102.78 8,382.59 1,117.36 -21.15 -18.19 8,461.55 1,201.95

Radioactive waste [kg] 745.19 773.13 1.86 2,757.85 -1.86 -1.60 745.19 3,529.38

551


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

Table 2. Life Cycle Impact Assessment of Rubber Tired Gantries - TRACI method TRACI

Impact/[unit]

Acidification Air

Eutrophication Water

Global Warming Air

Human Health Criteria Air Point Source

Phase

RTG type

[mol H+ eq.]

[kg N eq.]

[kg CO2 eq.]

[kg PM2.5 eq.]

diesel electric diesel electric diesel electric diesel electric

933.58 934.65 4,450,249.43 698,156.07 -1,335.00 -376.00 4,449,848.01 698,714.72

Cradle to Gate Gate to Grave End of Life Cradle to Grave

24.00 24.00 4,616.07 530.17 -1.87 2.67 4,638.20 556.84

335,818.04 343,206.03 6,169,662.32 1,449,098.83 -16,128.50 -14,135.00 6,489,351.86 1,778,169.87

199.23 199.23 40,011.83 21,003.59 -2.02 0.01 40,209.04 21,202.82

Table 3. Life Cycle Impact Assessment of Utility Tractor Rigs - CML method CML 2001 Phase Cradle to Gate Gate to Grave End of Life Cradle to Grave

Impact / [unit] UTR type diesel electric diesel electric diesel electric diesel electric

Global Warming Potential [kg CO2 eq.] 36,174.60 68,024.60 746,884.86 377,300.00 246.67 4,950.20 783,306.12 450,274.80

Acidification Potential [kg SO2 eq.] 191.95 360.95 9,760.85 2,002.00 1.31 26.27 9,954.11 2,389.21

Eutrophi-cation Potential [kg Phos. Eq.] 8.66 25.00 1,681.68 90.30 0.06 4.17 1,690.40 119.47

Ozone Depletion Potential [kg R11 eq.] 0.01 0.02 0.00 0.09 0.00 0.15 0.01 0.26

Photochem-ical Ozone Creation [kg Ethene eq.] 11.21 21.08 1,040.53 116.90 499.65 2.47 1,551.39 140.45

Radioactive waste [kg] 117.45 220.86 0.00 1,225.00 0.80 1.14 118.25 1,447.00

Table 4. Life Cycle Impact Assessment of Utility Tractor Rigs - TRACI method TRACI

Impact/[unit]

Acidification Air

Eutrophication Water

Global Warming Air

Phase

UTR type diesel electric diesel electric diesel electric diesel electric

[mol H+ eq.] 101.48 188.06 553,808.82 288,893.49 166.14 3,790.28 554,076.44 292,871.83

[kg N eq.] 2.14 29.00 574.44 299.66 1.31 26.27 577.90 354.92

[kg CO2 eq.] 36,501.96 73,022.56 767,780.20 387,926.71 230.33 5,089.60 804,512.49 466,038.87

Cradle to Gate Gate to Grave End of Life Cradle to Grave

Environmental Protection Agency). The detail that is developed by that US EPA has influenced the wide use of this methodology in North America. It differs from the CML methodology because the data comes primarily from US sources. In this way, the results of the paper are equally applicable to EU and US ports, since RTG cranes and UTRs can be found in operation in the ports of both areas. Other abbreviations used further in the paper are: acidification potential (AP), eutrophication potential (EP), ozone depletion potential (ODP), photochemical ozone creation potential (POCP) and human health criteria air-point source (HHCAPS). 552

Human Health Criteria Air Point Source [kg PM2.5 eq.] 28.46 75.00 4,979.25 2,566.62 0.25 109.59 5,007.96 2,751.22

4.1 Life Cycle Assessment results of Rubber Tired Gantries The results show the modest contribution of the ‘cradle-to-gate’ phase to the overall impact, which is extremely common for products with long life cycles, especially large fossil fuel-powered machines. The raw material depletion and energy consumption for manufacturing process of approx. 1.8×106 MJ are the most significant issues of the ‘cradle to gate’ phase. The differences between the environmental impacts of the production phase of conventional and electric RTG is approximately 2.5%. This ratio is less than the ‘cut off’ criteria assumption of 5%, underlining that

Vujičić, A. – Zrnić, N. – Jerman, B.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

difference in production of E-RTG over conventional power has little additional environmental consequence. The ‘gate to grave’ or ‘use’ phase is predominantly shaped by the energy intensity of FU and the long life cycle. The differences between conventional and electric RTG in terms of GHG emissions and LCIA are significant. The results show an over 70% of CO2 eq. reduction for electric RTG, which could be attributed to the use of the EU-25 power grid mix from which 539 grams of CO2 eq. per kWh is emitted. The result would vary considerably in cases of different footprints for other power plants (i.e. coal versus wind energy).

recyclable, no negative effects can be associated with this phase. 4.2 Life Cycle Assessment Results of Utility Tractor Rigs The ‘cradle to gate’ phase of diesel and electric UTRs show substantial differences in the enginemanufacturing process. The environmental impact of lithium-ion battery pack production is much greater than of conventional diesel engines. The GWP for a 140 kWh battery pack of the chosen electric UTR model is approx. 37,000 kg of CO2 eq., while for an 8-litre 6-cylinder diesel engine it is less than 3,000 kg of CO2 eq. This proportion is a significant issue that has to be taken in to consideration.

Fig. 3. Life cycle impact of Rubber Tired Gantries – CML method Fig. 5. Life cycle impact of Utility Tractor Rigs – CML method

Fig. 4. Life cycle impact of Rubber Tired Gantries – TRACI method

The ‘end of life’ phase shows that the potential LCA environmental impact could be positive (negative values in Tabs 1 and 2). This is the case with recycling, when energy savings are greater than disposal of leftovers. Since the steel used for gantry and spreader weights almost 100 tonnes and is highly

The results of the ‘gate to grave’ phase show a similar trend of lesser environmental impact of electric UTR, as with RTG cranes. In case of electric UTR, the GWP is up to 50% reduced over its entire life cycle. Again, this is with the EU-25 power grid mix and the results could vary depending on power plant footprint. An additional downside that can slightly diminish the efficiency of electric UTR may be the life cycle of lithium-ion battery packs. They lose some capacity potential after a certain number of recharges, i.e. between 1,500 and 2,500 working hours. Since the life cycle of UTRs is usually longer (3,500 hours in this study), one replacement in order to maintain UTR’s power capacity at maximum level can be expected. The additional case of their replacement during a 10year life cycle of UTR is not part of the introduced LCA, but this scenario is presented within a ‘what if’ analysis. The ‘end of life’ phase is also influenced with lithium ion batteries disposal. The GWP impact of the disposal of a 140 kWh lithium ion battery pack is

Ports Sustainability: A life cycle assessment of Zero Emission Cargo Handling Equipment

553


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

close to 5,000 kg of CO2 eq., while the diesel engine is 300 kg of CO2 eq. It is necessary to underline that life cycle of diesel engine with environmentally undemanding overhauling can easily be extended by an additional 5 years.

electric UTR. If the additional environmental impact of battery replacement is taken into account, as with case of presented electric UTR no. 4, the overall results are actually worse than conventional dieselburning UTR.

Fig. 6. Life cycle impact of Utility Tractor Rigs – TRACI method

Fig. 7. Life cycle impact of Utility Tractor Rigs depending on power grid mix

4.3 What-If Analysis Because the selection of the power grid mix can, due to nature of LCA, provide entirely different results, a short ‘what if’ analysis is conducted. In order not to overextend the comparison of data, only GWP is taken into account. The calculation for the replacement of a lithium ion battery pack of an electric UTR after 5 years is also shown in Fig. 7. The assumptions made for comparison of environmental impacts of UTRs presented in Fig. 3 as follows: • Electric 1; refers to results from LCA, without a battery change and EU-25 power grid mix with GWP of 0.539 kg of CO2 eq. per kWh; • Electric 2; the same as above with a battery replacement (disposal of old batteries and entire life cycle of new batteries); • Electric 3; for this UTR, the power grid mix is adapted to be the world average with GWP of 0.749 kg of CO2 eq. per kWh; The battery replacement is also taken into account; • Electric 4; The power grid mix is an average GWP for coal power plants approximated to 1 kg of CO2 eq. per KWh; • Diesel; results from LCA study. The analysis shows that electric UTR has lower GWP only up to the level of 0.9 kg of CO2 eq. per kWh of electric energy. In the case of coal-burning power plants, the emissions of a conventional diesel UTR are only replaced with a similar level of GWP of 554

5 CONCLUSIONS The efforts of the CHE industry in providing ports and container terminals with environmentally more efficient technologies are becoming increasingly visible. Almost every piece of CHE today is offered with some solution for the reduction of emissions and energy consumption, from alternative fuels and hybrid technology, to promising ‘zero emission’ concepts. The ‘zero emission’ concept applied to the RTG crane and UTR as the core of CHE is investigated using LCA methodology. The entire life cycles of a conventional diesel RTG crane and UTR are compared with electric ones in order to reveal any sustainability sensitivities that are common with energy source transitions. In this respect, the results of LCA present the electrification of CHE as a feasible and sustainable solution aimed mitigating the environmental impact of ports. The transition from diesel to electric handling equipment is a step forward, although from the LCA perspective, ‘zero emission’ operations are impossible to achieve. The ‘gate-to-grave’ phase of electric CHE shows significant reductions in GWP, AP and EP with differences between RTG and UTR. Based on the comparative LCA, it has been proved that electrification of CHE has a greater potential in reducing the emissions and energy consumption of larger and heavier equipment.

Vujičić, A. – Zrnić, N. – Jerman, B.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 547-555

The energy-consuming work operations of RTG cranes are more suitable for electrification despite the remaining ‘tailpipe’ emissions during block changes. The obtained results show that due to the long life span and the large number of annual operating hours GWP, savings cannot be derived from any power grid mix scenario. However, electric UTRs have no ‘tailpipe’ emissions but are noticeably influenced by the environmental burden of lithium ion batteries. The life span of UTRs can be twice longer than the life span of batteries necessary to provide stored energy, and their replacement along with different power grid mix scenario than the one used in the LCA can jeopardize the emission reduction potential of electric UTRs. The LCA has proved itself to be a valuable tool for the comparison of such complex products as the CHE. It offers a systematic approach for sustainability evaluations of life cycles of conventional and new technologies. However, in order to avoid any serious data inventory uncertainties (often pointed out by LCA critics), a random comparison of on-site measurements results with functional unit assumptions is recommended. 6 ACKNOWLEDGMENT A part of this research is supported by the bilateral project No. 651-03-1251/2012-09/50 for scientific cooperation between Serbia and Slovenia. 7 REFERENCES [1] Cannon, J.S. (2008). U.S. Container Ports and Air Pollution: A Perfect Storm. Energy Futures Inc., Boulder. [2] Cannon, J.S. (2009). Container Ports and Air Pollution. Energy Futures Inc., Boulder. [3] Nikitakos, N. (2012). Green Logistics - The concept of Zero Emissions Port. FME Transactions, vol. 40, no. 4, p. 201-206. [4] Kartnig, G., Grösel, B., Zrnić, N. (2012). Past, state of the art and future of intralogistics. FME Transactions, vol. 40, no. 4, p. 193-200. [5] Starcrest (2012). The Port of Los Angeles: Inventory of Air Emissions for Calendar Year 2011. Starcrest Consulting Group, Long Beach. [6] Starcrest (2008). San Pedro Bay Ports: Emissions Forecasting Methodology and Results. Starcrest Consulting Group, Poulsbo. [7] Böse, J.W. (2011). Handbook of Terminal Planning, Springer, New York, DOI:10.1007/978-1-4419-8408-1. [8] Liu, C.I., Jula, H., Ioannou, P.A. (2002). Design, simulation, evaluation of automated container terminals. IEEE Transactions on Intelligent

Transportation Systems, vol. 3, no. 1, p. 12-16, DOI:10.1109/6979.994792. [9] Geerlings, H., Duin, R. (2011). A new method for assessing CO2-emissions from container terminals: a promising approach applied in Rotterdam. Journal of Cleaner Production, vol. 19, no. 6-7, p. 657-666, DOI:10.1016/j.jclepro.2010.10.012. [10] Zrnić, N., Vujičić, A. (2012). Evaluation of environmental benefits of CHE emerging technologies by using LCA. 12th International Material Handling Research Colloquium, Conference Proceedings, p. 1628. [11] Pušavec, F., Kopač, J. (2011). Sustainability assessment: cryogenic machining of Inconel 718. Strojniški vestnik – Journal of Mechanical Engineering, vol. 57, no. 9, p. 637-647, DOI:10.5545/sv-jme.2010.249. [12] Ostad, H., Collado-Ruiz D. (2011). Tool for the environmental assessment of cranes based on parameterization. International Journal of Life Cycle Assessment, vol. 16, no. 5, p. 392-400, DOI:10.1007/ s11367-011-0280-z. [13] ISO 14040, (2006). ISO Report 14040 Environmental Management - Life Cycle Assessment, Principles and Framework. International Organization for Standardization, Geneva. [14] Kim, J., Rahimi, M., Newell, J. (2012). Life-cycle emissions from port electrification: A case study of cargo handling tractors at the port of Los Angeles. International Journal of Sustainable Transportation, vol. 6 no. 6, p. 321-337, DOI:10.1080/15568318.2011 .606353. [15] Chester, M.V., Horvath, A. (2009). Environmental assessment of passenger transportation should include infrastructure and supply chains. Environmental Research Letters, vol. 4. no. 9, p. 1-8, DOI:10.1088/1748-9326/4/2/024008. [16] Cooper, J.S., Fava, J.A. (2006). Life cycle assessment practitioner survey – Summary of results. Journal of Industrial Ecology, vol. 10, no. 4, p. 12-14, DOI:10.1162/jiec.2006.10.4.12. [17] Zackrisson, M., Avellána, L., Orlenius, J. (2010). Life cycle assessment of lithium-ion batteries for plug-in hybrid electric vehicles - Critical issues. Journal of Cleaner Production, vol. 18, no. 15, p. 1519-1529, DOI:10.1016/j.jclepro.2010.06.004. [18] Majeau-Bettez, G., Hawkins, T., Strømman, A. (2011). Life cycle environmental assessment of lithiumion and nickel metal hydride batteries for plug-in hybrid and battery electric vehicles. Environmental Science & Technology, vol. 45, no. 10, p. 4548-4554, DOI:10.1021/es103607c. [19] Azapagic, A, (2010). Life cycle assessment as a tool for sustainable management of ecosystem services, in Harrison, R.M., Hester, R.E. (eds.), Ecosystem services. RSC Publishing, Cambridge, p. 140-168, DOI:10.1039/9781849731058-00140.

Ports Sustainability: A life cycle assessment of Zero Emission Cargo Handling Equipment

555


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2012.940 Special Issue, Original Scientific Paper

Received for review: 2012-12-29 Received revised form: 2013-03-21 Accepted for publication: 2013-04-10

Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure Rusiński, E. – Moczko, P. – Pietrusiak, D. – Przybyłek, G. Eugeniusz Rusiński – Przemysław Moczko* – Damian Pietrusiak – Grzegorz Przybyłek Wroclaw University of Technology, Poland In this paper, studies into the causes of fatigue cracks of a jaw crusher supporting structure are presented. The problems appeared after short time of operation at a new crushing facility. A large number of fatigue cracks with a high growth rate and bolt failures were observed in the crusher supporting structure. Considering the high dynamic forces that occur during the operation of such equipment, an investigation into the problem was undertaken in order to prevent catastrophic failure of the crushing station. A specially developed, combined numerical and experimental method was used to determine the reasons for the problems and to solve them. Keywords: jaw crushers, fatigue resistance, resonance, numerical simulations, experimental methods

0 INTRODUCTION High dynamic forces are present during the operation of crushing stations. The entire load is transferred to the supporting structures and foundations during operation. Such loads have to be considered at the design stage and special care must be paid to modal parameters of the structure in order to prevent possible resonance problems. The present state of the art concerning static load is well developed and provides full resistance in terms of the ultimate strength range of the structures [1] and [2]. Design with a fatigue resistance approach is still problematic due to the complexity of the fatigue phenomenon. Additionally, vibration and, in some cases, resonance problems intensify the occurrence of the fatigue fractures. Although this problem has been known for many years, its complete treatment has become possible by the common use of sophisticated experimental and numerical techniques [3] and [4]. There are a lot of examples of failures caused by unpredictable dynamic loads [5] and [6]. Reliable assessment and a proper solution are now possible due to the complex numerical-experimental approach, which is presented in Fig. 1. This procedure can be used for various cases of failures. The modified procedure can be adapted for use in the design stage of new structures as well. Application to a real structure failure was carried out at one of the Polish mines. The crushing station consists of typical equipment such as a chute, conveyor system, hammer and jaw crusher, which is supported by a steel frame anchored to the concrete foundations. A view of the crushing station is shown in Fig. 2. Within a short time (few weeks) of starting operations at the new facility, a number of fatigue cracks and bolt damage to the crusher supporting 556

structure were observed. A view of the cracks in the side vertical bracing is shown as an example in Fig. 3.

Fig. 1. Investigation procedure for determining the reasons for failure

Fig. 2. Jaw crusher layout (supporting structure highlighted in red, marked with arrow)

The growth rate of fatigue cracks was high. The maximum length of cracks was found to be 160 mm. The significant level of damage meant that immediate action would have to be taken in order to prevent

*Corr. Author’s Address: Wroclaw University of Technology, Lukasiewicza 7/9, 50-371 Wroclaw, Poland, przemyslaw.moczko@pwr.wroc.pl


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563

irreversible damage to the whole structure. But first the cause of the damage had to be identified.

Fig. 4. Locations and directions of the measuring points Fig. 3. Fatigue crack in the side bracing

A combined numerical and experimental method was used to determine the reasons for the poor durability of the structure and to resolve them. In the next sections, the various investigations are presented. 1 EXPERIMENTAL AND NUMERICAL STUDIES The first part of study covers measurements of dynamic parameters of the structure and an estimation of the dynamic loads causing the fractures. Dynamic loads acting on the crusher frame have a stochastic character, while the material provided to the crusher varies in size and mechanical properties in a random way. This situation required a determination of the maximum possible load acting on the structure in experimental terms. The measured loads were then used for the numerical strength calculations of the crusher’s supporting structure. In the following subsections, the results of both approaches are presented.

Fig. 5. Scanning head of the POLYTEC PSV-400 vibrometer

1.1 Experimental Tests The first part of investigations was aimed at measuring the dynamic behaviour of the structure and estimating the dynamic loads causing the fractures. Measurements of the displacement vibrations of the supporting frame were carried out using the following equipment: • vibrometer - COMMTEST vb7, • scanning vibrometer – POLYTEC PSV-400. Tests were performed at two points (A, B, Fig. 4), where displacements of the structure were measured in two horizontal directions, x longitudinal and y transverse. The testing equipment used during the tests is shown in Figs. 5 and 6.

Fig. 6. COMMTEST vb7 vibrometer

As a result of the measurements, a vibration spectrum and time traces were obtained. An example of a graph of the measurements results is shown in Fig. 7. The transverse vibrations of the structure are caused by resonance effects, which can be seen in the time traces of the vibrations where a characteristic beat effect is present. Analysis of the measurement results led to the conclusion that the structure operates at a resonance condition close to the excitation frequency fex = 3.33 Hz.

Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure

557


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563

Fig. 7. Time and spectrum trace with a visible resonance effect - transverse displacement

The maximum amplitudes of the displacements measured during the tests are presented in the Table 1. Table 1. Maximum amplitudes of the displacements Description Displacement measurements

Displacement [mm] x (longitudinal) y (transverse) 1.12 1.92

The displacements were used to calibrate the FEM model, which was created to evaluate the dynamic behaviour and strength of the structure in its existing and modified form.

of a linear shell, beam, as well as rigid and spring elements. Mass elements (squares) simulate the selfweight of the equipment placed on the supporting structure, such as the crusher unit, electrical motor, etc. There are 58000 elements and a similar number of nodes generated in total in the model of the structure. The boundary conditions consist of restraints and loads. Supporting points of the entire structure were restrained to simulate the behaviour of the real structure. The loads applied in the simulations are based on the results of the measurements and are described in the section below.

1.2 FEM Calculations In order to estimate the influence of the measured vibrations and to calculate the dynamic loads causing such effects, the FEM [7] model of the supporting structure was created. The model was used as a calibration tool for the dynamic forces caused by crushing cycles. Based on the results of the calibration calculations such forces were estimated. As a first step, a numerical model of the crusher’s supporting structure was created. Shell modelling was used to evaluate the exact stress effort and deformation of the structure. Such an approach enables the calculation of stress level in the notches, which is impossible with the use of beam modelling. Fig. 8 presents a model of the supporting structure with the shell elements generated. The model consists 558

Fig. 8. Numerical FEM model of the supporting structure

Numerical simulations were aimed at confirming the dynamic behaviour of the structure using modal

Rusiński, E. – Moczko, P. – Pietrusiak, D. – Przybyłek, G.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563

analysis and calculating and applying dynamic loads to perform strength calculations. Numerical modal analysis of the model confirmed the presence of the resonance effect under operating conditions. Table 2 lists the frequencies of the first three modes. Fig. 9 presents the deformation plot of the first mode, which is responsible for the resonance vibrations. The direction of the vibrations is in line with those measured on site and the frequency is close to the excitation load (fex), which is fex = 3.33 Hz, f1 = 3.39 Hz, fex <~ f1 . Table 2. Natural frequencies of the crusher’s supporting structure No. of natural mode 1 2 3

Frequency [Hz] 3.39 5.96 14.50

Areas, marked with arrows on the numerical model, indicate fatigue problems.

Fig. 10. Comparison of fatigue crack location in the numerical model and the existing structure

The locations of the cracks present on the crusher’s supporting structure and confirmed in the numerical calculations are presented in Fig. 11.

Fig. 11. Fatigue crack locations

Fig. 9. Natural frequency mode shape of the crusher’s supporting structure – Mode 1

The next step was to perform fatigue calculations. Considering that dynamic forces are created in the jaw area of the crusher unit, loads in both horizontal directions (x and y) were applied. Based on the displacement results, multiplication factors were calculated to obtain the same values for the displacements as those measured on site. For such conditions, strength (fatigue) calculations were performed in consideration of ISO 5049-1 standard requirements [8]. The structure of the crusher support was considered to be class C (number of fatigue cycles exceeds 6×105). The steel grade used for the structure was considered to be St3S (Fe 360). The results of fatigue calculations confirmed that the areas where cracks are present on the structure are overstressed and do not have proper fatigue resistance. There was 100% correlation between the cracks locations on the real structure and those in the numerical model (an example is shown in Fig. 10).

2 MODIFICATIONS OF THE CRUSHER SUPPORTING STRUCTURE Results of the test measurements as well as modal analysis and strength calculations confirmed that there are two main reasons for the poor operating conditions of the crusher supporting structure: 1) resonance vibrations in the transverse direction caused by not enough stiffness in the structure in this direction, 2) weak design of the joints, which have low resistance to loads due to operating conditions with movement in both horizontal directions (an example of such a design is shown in Fig. 12). The design of the joints would be sufficiently strong in the case of a static load, however the geometry of the gusset plates does not assure proper load transfer in the case of a dynamic load. Lateral forces acting on the faceplate and the bolted connection cause high effort in those elements. Moreover, the joints do not assure smooth stiffness changes. Therefore, strong bending and pulsations are present and cracks occur as a result of rapid fatigue.

Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure

559


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563

of the side bracing has a positive influence on the structure dynamics.

Fig. 12. Example of the weak design of the joints

The above reasons, identified by the numericalexperimental procedure, indicate the improper design of the structure, which has to be changed in order to prevent critical damage to the structure. 2.1 Solving the Resonance Problem As measured on site and calculated with use of the FEM model, the structure suffers from transverse resonance vibration caused by overlapping of the excitation (crusher rpm) frequency and the first natural vibration mode. The natural frequencies must be shifted away from the excitation frequency in order to enable safe operation of the crushing station. The basic equation for natural frequency calculations ω0 [Hz] is presented below:

ω0 =

k , (1) m

where k is stiffness [N/m], and m is mass [kg]. Taking into account the above equation, either the crusher support can be stiffened to increase the natural frequencies or additional mass can be placed on the frame to decrease the vibration frequency. The first solution was chosen because the cross section of the crusher supporting frame was shown not to be stiff enough in transverse direction. The easiest way to increase stiffness would be to incorporate diagonal bracing inside the frame. However, the conveyor line located inside the frame prevents this. Therefore another stiffening solution was designed. Two of three cross-sections of the frame were reinforced with side bracing as shown in Fig. 13. Thus, there are four bracings in total anchored to the concrete foundation and welded to the existing structure. This change was incorporated into the FEM model to confirm its influence on the natural frequencies of the modified supporting structure. The position of the side bracing was mainly determined by the ease of assembly on the already existing structures. The results of the modal analysis of the modified crusher supporting structure are shown in subsection 2.3 of the paper. The results confirm that the position 560

Fig. 13. Side bracing design

2.2 Weak Joints Modifications According to the FEM calculations on the existing structure, the fatigue resistance of critical joints (see Fig. 11) does not fulfil ISO-5049 standards [8]. This was confirmed by the presence of the many cracks observed in these areas (an example of such cracking is shown in Fig. 3). Due to the condition of the existing joints, which exhibit multiple cracks and thus multiple repairs, it was decided to cut out all weak joints and incorporate new redesigned ones. Special care was paid to the joint’s design to assure proper fatigue resistance. This new solution required the replacement of existing longitudinal bracings with new ones. Fig. 14 shows the new design of the longitudinal bracings incorporated into a geometrical model of the crusher supporting structure. In order to enable easy manufacturing and built process into the existing structure, welding connections were considered on the designing stage.

Fig. 14. New design of side and longitudinal bracings

Rusiński, E. – Moczko, P. – Pietrusiak, D. – Przybyłek, G.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563

The supporting structure with redesigned longitudinal and side bracings was analysed with the use of FEM models to check and confirm its resistance to dynamic loads (resonance) and fatigue. The results of these analyses are presented below. 2.3 Numerical Check Calculations of the Modified Structure In the first step, a modified model of the crusher supporting structure was checked for resonance issues. Modal analysis of the model confirmed that the natural frequencies of the structure were shifted significantly up. Table 3 lists the frequencies of the first three modes. Fig. 15 presents a deformation plot of the first mode, which was responsible for the resonance vibrations in the existing structure. The first natural frequency is shifted from the excitation frequency and equals: fex = 3.33 Hz, f1 = 8.58 Hz, fex < f1. There are no natural frequencies close to the first harmonic of the excitation frequency, which equals: fex,1h = 6.66 Hz. The above results confirm that there is no risk of resonance during operation of the modified supporting structure of the crusher. Table 3. Natural frequencies of the modified crusher supporting structure No. of natural mode 1 2 3

Frequency [Hz] 8.58 8.72 16.85

whether the new design of the joints shows proper fatigue resistance. Calculations were carried out using the FEM model in accordance with ISO-5049 standard requirements. As a result of such calculations, the stress amplitude σa at each point of the structure was obtained. The maximum value calculated in the structure was found in the gusset plate of the longitudinal bracing: σamax = 42 MPa. Fig. 16 shows a view on this area with the stress amplitude plot. The maximum stress peak is located in the welded connection of the gusset plate and pipe section. In this area, a K3 notch was designed where the permissible stress amplitude equals: σaperm = 60 MPa. A comparison of both values confirms that the modified structure of the crusher supporting structure is safe from the fatigue resistance point of view.

Fig. 16. Stress amplitude σa in the modified crusher supporting structure

As a comparison of the existing and new design, Fig. 17 presents the same area of the gusset plate in the old design. The maximum stress amplitude in this area equals σamax = 117 MPa. This value exceeds the permissible values of the stress amplitude, which was indicated by the fatigue cracks in the maximum stress area.

Fig. 15. Natural frequency mode of the modified crusher supporting structure – Mode 1

The next step was to perform fatigue check calculations of the modified structure to verify

Fig. 17. Stress amplitude σa in the existing crusher supporting structure [MPa]

Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure

561


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563

3 MODIFICATIONS OF THE STRUCTURE, TESTS MEASUREMENTS In the previous section, the new designs of the modifications to the crusher supporting structure were presented. The results of the dynamic (modal) and fatigue calculations confirmed that the modified structure should be safe under normal operating conditions. The new design of the side and longitudinal bracings were incorporated into the structure prior to removal of existing weak joints and bracings. Fig. 18 shows an example of the redesigned joint before and after modification.

Fig. 18. Weak joint; a) after, and b) before modification

The final step of the investigation was taking measurements of the dynamic behaviour of the

modified structure in order to confirm that resonance was no longer present during operation. The tests were performed with the use of a COMMTEST vb7 vibrometer at the same locations as the tests on the original structure. The vibration spectrum and time traces obtained are shown in Fig. 19. There are no resonance vibrations present in the time trace of the measured signals. The RMS value of the vibrations is 400% lower compared to the vibrations of the original structure. 4 CONCLUSIONS In this paper, an experimental and numerical approach to investigating the reasons for the serious fatigue failures of the jaw crusher supporting structure is presented. The investigation method dedicated to the high performance machines was used for this purpose. Experimental vibrations measurements and FEM calculations allowed us to obtain the dynamic loads caused by the crushing process, to evaluate the influence of such loads on the fatigue resistance of the structure, and to confirm the operating conditions with the resonance effect present. Fatigue calculations confirmed that the structure shows fatigue weakness in particular joints. It must be emphasized that there is 100% correlation between the crack locations on the real structure and those in the numerical model.

Fig. 19. Time and spectrum displacement signal of transverse vibrations of the modified structure

562

Rusiński, E. – Moczko, P. – Pietrusiak, D. – Przybyłek, G.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 556-563

In order to solve the resonance problems and to avoid fatigue cracks in the structure, additional side bracings were incorporated and existing longitudinal bracings were replaced with redesigned ones. The new design was checked for dynamic and fatigue behaviour using the modified FEM model. In the final stage, the modified structure was tested using a vibration measuring device. The test measurements confirmed that there was no resonance and that the structure exhibited proper behaviour during normal operation. The combined, experimental-numerical method can be applied to various operational problems of industrial equipment. This method is one way to improve design quality and can be used on new or existing structures. This method also allows us to identify reasons for failures, to measure operational parameters, to calibrate numerical models, and to investigate various phenomena causing potential problems. The verified numerical models can be used for various analyses in order to solve existing problems or to pinpoint possible future failures. This combined, experimental-numerical method has been used and verified in many studies performed by the authors of this paper [9] and [10]. 5 REFERENCES [1] Ismar, H., Burzic, Z., Kapor, N., Kokelj, T. (2012). Experimental investigation of high-strength structural steel welds, Strojniški vestnik - Journal of Mechanical Engineering, vol. 58, no. 6, p. 422-428, DOI: DOI:10.5545/sv-jme.2011.281. [2] Saoudi, A., Bouazara, M., Marceau, D. (2011). Fatigue failure study of the lower suspension vehicle arm using a multiaxial criterion of the strain energy

density. Strojniški vestnik - Journal of Mechanical Engineering, vol. 57, no. 4, p. 345-356, DOI:10.5545/ sv-jme.2009.074. [3] Arsic, M., Bosnjak, S., Zrnic, N., Sedmak, A., Gnjatovic, N. (2011). Bucket wheel failure caused by residual stresses in welded joints. Engineering Failure Analysis, vol. 18, no. 2, p. 700-712, DOI:10.1016/j. engfailanal.2010.11.009. [4] Bosnjak, S., Zrnic, N., Simonovic, A., Momcilovic, D. (2009). Failure analysis of the end eye connection of the bucket wheel excavator portal tie-rod support. Engineering Failure Analysis, vol. 16, no. 3, p. 740750, DOI:10.1016/j.engfailanal.2008.06.006. [5] Rusinski, E, Czmochowski, J, Iluk, A, Kowalczyk, M. (2010). An analysis of the causes of a BWE counterweight boom support fracture. Engineering Failure Analysis, vol. 17, no. 1, p. 179-191, DOI:10.1016/j.engfailanal.2009.06.001. [6] Rusinski, E., Dragan, S., Moczko, P., Pietrusiak, D. (2012). Implementation of experimental method of determining modal characteristics of surface mining machinery in the modernization of the excavating unit. Archives of Civil and Mechanical Engineering, vol. 12, no. 4, p. 471-476, DOI:10.1016/j.acme.2012.07.002. [7] Rusiński, E., Czmochowski, J., Smolnicki, T. (2000). Advanced Finite Element Method. Wroclaw University of Technology, Wroclaw. (In Polish) [8] ISO 5049-1:1994 (1994). Mobile equipment for continuous handling of bulk materials - Part 1: Rules for the design of steel structures. International Organisation for Standardization, Geneva. [9] Rusiński, E., Moczko, P., Czmochowski, J. (2008). Numerical and experimental analysis of a mine’s loader boom crack. Automation in Construction, vol. 17, no. 3, p. 271-277, DOI:10.1016/j.autcon.2007.05.010. [10] Rusiński, E., Moczko, P. (2010). A combined numerical-experimental method for determining the spatial distribution of a residual stress in a notch. Materials Science -Poland, vol. 28, no. 1, 393-399.

Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure

563


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572 © 2013 Journal of Mechanical Engineering. All rights reserved. DOI:10.5545/sv-jme.2010.268 Special Issue, Original Scientific Paper

Received for review: 2010-12-28 Received revised form: 2011-03-03 *Accepted for publication: 2011-09-05

Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals Marković, N. – Schonfeld, P. Nikola Marković – Paul Schonfeld*

Department of Civil and Environmental Engineering, University of Maryland, College Park, USA This paper studies the recovery of an intermodal freight system from a major disruption and develops a model for optimising vehicle schedules under disrupted conditions. The proposed model optimises the recovery of a single-terminal system with relatively short feeder routes on which vehicle roundtrip times are exponentially distributed and arrivals at the terminal are Poisson-distributed. Mathematical expectations are used to formulate the deterministic equivalent for the scheduling problem and a genetic algorithm is applied to optimise the schedules on main routes. The model developed in this paper can be applied to single-terminal transfer systems with any combination of transportation modes using discrete vehicles, as long as the feeder arrivals do not deviate much from the assumed Poisson distributions. Since its computational time is relatively insensitive to the numbers of vehicles on feeder routes, this model can be used to efficiently optimise intermodal systems with numerous vehicle arrivals. Keywords: scheduling, disruption, intermodal, Poisson, genetic algorithm, transfer

0 INTRODUCTION Efficient transfer coordination in an intermodal transportation network can reduce the dwell times of cargos at the transfer terminals where various routes interconnect, thereby also increasing the vehicle utilisation rates, reducing the need for direct routes to connect many origins and destinations, reducing storage requirements at transfer terminals, and improving total system efficiency. In this paper we analyse an intermodal freight system with a single transfer hub and develop a model that optimises the schedule of vehicles on main routes while assuming Poisson arrivals on feeder routes. This model determines the departure times on main routes that minimize the supplier’s overall system cost, including storage, vehicle, in-terminal operation and late delivery penalty costs. The optimisation problem addressed in this paper is related to some classical problems of operations research, such as machine scheduling, lot sizing, and supply chains. Somewhat related machine scheduling problems can be found in [1] to [3]. [4] to [7] address the scheduling problem in transfer systems, but under different conditions from those considered here. For example, [4] and [5] analyse different transfer coordination policies and determine the thresholds in the intermodal systems with complex multistop routes and lower variance in travel durations, both typical for normal operations. In this paper, we analyse the case with high variances in travel times, which are typical of disrupted operations and model the arrivals as a Poisson process. [6] and [7] deal with scheduling takeoff times, a problem that will be studied in this paper. Both papers optimise 564

departures on a single airline route and under different demand assumptions from those considered here (i.e. [6] assumes uniform demand, whereas [7] adopts time dependent demand). This paper is based on the same framework as [8] and develops a model which, unlike [8], is suitable for intermodal systems with numerous arrivals of vehicles on feeder routes. In [8], Marković and Schonfeld develop a scheduling model which assumes generally distributed vehicle roundtrip durations and vehicles operating on multiple feeder routes. Low computational efficiency of the stochastic program used in [8] enabled only the optimisation of schedules in systems with relatively few arrivals on feeder routes. In this paper we provide a computationally less demanding model by assuming exponentially distributed vehicle roundtrips and fixed fleet size on feeder routes. These assumptions allow us to model the arrivals as a stationary Poisson process and derive the expectations needed to formulate a scheduling problem that is optimised much more efficiently than the stochastic program in [8]. Thus, the model developed here can efficiently optimise large intermodal systems with numerous arrivals on feeder routes. In this paper we analyse the recovery of a system from a major disruption during which large amounts of freight have accumulated along the feeder routes, which are assumed here to be served by trucks. To dissipate the backlogs we let the trucks on feeder routes operate nonstop and deliver cargo to the terminal where the freight is transferred to main routes, which are assumed here to be aircraft routes. Thus our transfer terminal represents an airport hub. We use pre-determined fleet sizes on feeder routes and seek to optimise the number of departures and specific

*Corr. Author’s Address: Department of Civil & Environmental Engineering, University of Maryland, College Park, MD 20742, USA, pschon@umd.edu


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

departure schedules on main (air) routes. We consider one-directional flow going from origins along the feeders’ routes towards destinations at the main routes, as might be expected in emergency evacuations or recoveries from major disruptions. In Section 1 we describe the operations within the observed intermodal system and explain the tradeoff between different types of costs. The anticipated types of costs, which are included in the objective function, are formulated in Section 2. Section 3 explains the constraints, while Section 4 provides the model formulation that is further tested on numerical examples designed in Section 5. Finally we draw conclusions and suggest possible extensions of this work.

the earlier the takeoff is scheduled, the greater are the chances that an airplane’s capacity will be underused due to insufficient level of stock. Operating less than full airplanes may require running additional flights, thereby increasing the airline service cost.

1 PROBLEM We consider an intermodal system with relatively short truck routes that feed cargo to major airplane routes (Fig. 1), which has suffered a major disruption. In order to reduce the backlogs accumulated along the feeder routes while the system is inoperative, each truck operates nonstop and fully loaded between an origin and the hub, without pausing between such round trips while backlogs persist. The trucks collect freight from multiple origins along their feeder routes and deliver it at the airport hub. When the takeoff on route is scheduled at time , the airplane is filled to capacity with freight, as long as freight backlogs persist. If the airplane cannot carry all the freight waiting at the airport, the remaining freight has to wait for the next flight with available capacity. On the other hand, if prior to the takeoff, there is little freight in the terminal’s storage connecting to route l, the airplane’s capacity is underused and an additional flight may be needed later. For simplicity, we assume that all trucks are similar and all operate at equal maximum capacity. Moreover, we assume that airplanes have similar capacities. Finally, we assume that the expected amount of cargo waiting for connections can never exceed a preset multiple (e.g. 0.8) of the terminal’s storage capacity. Our objective is to find the optimal (i) number of takeoffs on each air route and (ii) corresponding schedule, for the given probabilistic durations of roundtrips on truck routes. In computing total cost we consider the storage cost, in-terminal operation cost, penalty for late delivery, and airline service cost. A tradeoff exists between the aforementioned types of costs. The earlier one schedules the takeoff, the lower are the storage and penalty costs associated with the freight that successfully connects. However,

Fig. 1. Intermodal freight system

2 COSTS In this section we introduce the notation used and explain how various types of costs are computed. We begin with the assumptions that allow us to model the arrivals on feeder routes as a Poisson process. We then compute the arrival intensities, which are further used in the development of storage, in-terminal operation, penalty, and airline cost. Suppose that a single truck operates on a relatively short feeder route i whose starting and end point is the terminal where the truckload connects to the airplane route l. Let’s assume that the duration of the truck’s roundtrip is exponentially distributed with a mean denoted as 1/ λil . Moreover, it is reasonable to assume that the observed transportation process has the following three properties: 1. The probability that a truck will accomplish more than one roundtrip within an infinitesimal time interval is negligible. 2. The duration of a roundtrip does not depend on the duration of the previously completed roundtrip. 3. The probability that a roundtrip will end within the time interval t depends on the interval’s length, rather than on the time period in which t was observed. Having adopted the above assumptions, we can model the truck arrivals as a Poisson process with the mean arrival rate λil according to [9] and [10]. If we assign to feeder route i more than one truck, the

Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals

565


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

arrival rate on route i is given in Eq. (1), in which ni represents the number of trucks assigned to feeder route i. ri = ni λil . (1)

Furthermore, if we define with Il the set of feeder routes connecting to main route l, the arrival rate of truckloads connecting to route l is:

λ l = ∑ ni λil . (2)

i∈I l

If we denote the jth takeoff time on route l as t lj , the expected number of truckloads connecting to route l and arriving to terminal between two consecutive flights is: E[TR l (t lj − t lj −1 )] = λ l (t lj − t lj −1 ). (3)

1 l ( S j −1 + S lj )(t lj − t lj −1 )CDT . (6) 2

We can further compute the total storage cost for freight connecting to route l by summing Eq. (6) over all the flights in Jl. 1 SC l = ∑ ( S lj + S lj −1 )(t lj − t lj −1 )CDT . (7) 2 j∈J l Finally, we can compute the total storage cost by summing Eq. (7) over the set L, which denotes main routes.

SC =

1 ∑ ∑ (S lj + S lj −1 )(t lj − t lj −1 )CDT . (8) 2 l∈L j∈J l

2.2 In-terminal Operation Cost 2.1 Storage Cost To compute the storage cost, we need to keep track of the inventory level. Moreover, since multiple feeder routes connect to multiple main routes, we need to know the stock for each main route. Therefore we define the variable S lj , which defines the inventory level of freight connecting to main route l, after the jth takeoff. We also define Alj representing the amount of freight transported in the jth flight on route l. Considering the inflow and outflow of freight into the terminal storage, Eq. (4) has to hold for all flights on all routes. Please note that S0l , t0l and A0l all equal 0. Moreover, we assume that all the freight arriving at the terminal before the last scheduled takeoff has to be flown. Thus we also set S nll to equal 0.

S lj −1 + λ l (t lj − t lj −1 ) = Alj + S lj ∀j ∈ J l ∀l ∈ L. (4)

Moreover, since we do not know in advance if there will be enough freight in the terminal’s storage to fill the airplane, we specify in Eq. (5) that the airplane will be loaded with all the connecting freight found in terminal that can fit within the airplane’s capacity, denoted Ac.

{

}

Alj = max Ac , S lj −1 + λ l (t lj − t lj −1 ) . (5)

Based on the previous derivations, in Eq. (6) we can compute the storage cost between two consecutive flights for freight connecting to route l. Please note that CDT denotes the storage cost per truckload-hour.

566

Here we analyse the loading and unloading cost due to the cargo transfer from trucks to airplanes. We assume that the in-terminal operation cost is lower when a truck arrives slightly before the takeoff and takes its truckload directly to the airplane, instead of unloading it in the terminal storage. Therefore, let’s define parameter d so that a truck arriving within the (t lj − d , t lj ) interval takes its truckload directly to the airplane. Now we can compute the expected number of truckloads that will be loaded directly on the airplane: bd = ∑ ∑ λ l d . (9)

l∈L j∈J l

Here we assume that d is smaller than the interval between two consecutive flights on the same route. Thus, the expected number of truckloads being loaded directly onto the airplane depends on the number of takeoffs rather than on their schedule. If we denote Ctd to be the unit in-terminal operation cost for the case of direct transfer to the airplane, Cti to be the unit cost for the case of indirect transfer to the airplane, and G to be the total number of truckloads, then the total in-terminal operation cost is:

IC = bd Ctd + (G − bd )Cti . (10)

2.3 Penalty Cost A penalty is imposed for late delivery, reflecting the lower value of freight that is delivered later. Here we assume that the time of the takeoff is relevant

Marković, N. – Schonfeld, P.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

for computing the penalty cost. We define a penalty function fp as the piecewise linear function of takeoff time starting from the beginning of the observed time period (the moment the system starts recovering from a disruption), as shown in Fig. 2.

Fig. 2. Penalty function fp

Now we can compute the penalty cost by summing the penalty for all the flights on all the air routes, as shown in Eq. (11). Please note that we again use Alj as defined in Eq. (5), which denotes the number of truckloads carried on the jth takeoff on route l. PC = ∑ ∑ A f p (t ). (11)

l∈L j∈J l

l j

l j

2.4 Airline Cost

The last constraint we consider is the terminal’s storage capacity. We assume that the expected amount of freight at the terminal should never exceed the multiple ms of the storage capacity Sc. Since we previously explained how the expected inventory level for freight connecting to route l at takeoff time t lj is computed, we must now ensure that the total expected inventory never exceeds the storage capacity msSc. Thus we define parameter pt and control the total inventory level at time pt. In order to do so we first need to find the inventory level of freight connecting to route l at time pt. We introduce sl which denotes the takeoff time on route l prior to pt and define kl which equals the takeoff index j.

{

AC = ∑ nl C Al . (12)

}

s l = max t lj : t lj ≤ pt , (15)

k l = index j : (t lj ≤ pt ∧ t lj +1 ≥ pt ). (16)

Now we can compute the expected inventory of freight connecting to route l as: S kl l + λ l ( pt − s l ). (17)

The last type of cost considered is the airline service cost, which covers the use of both airplanes and airport facilities. It is proportional to the number of the airplane roundtrips (takeoffs). We denote the number of takeoffs on route l as nl. Moreover, we denote as C Al the cost of an airplane roundtrip on route l. Finally, the total airline service cost is:

t lj − t lj −1 ≥ tmin ∀j ∈ J l ∀l ∈ L. (14)

Finally we can define the storage capacity constraint by summing Eq. (17) over the set of main routes and setting the sum below the storage capacity Sc multiplied by ms (a safety factor). Please note that the constraint in Eq. (18) should hold for any real value of time parameter pt.

∑S l∈L

l kl

+ λ l ( pt − s l ) ≤ Sc ms ∀pt ∈ R. (18)

l∈L

3 CONSTRAINTS

4 MODEL

In this section we analyse several constraints needed in order for the mathematical model to fairly represent the real world. The first constraint that we consider is the time window constraint for takeoffs. Utilisation of airport facilities is often restricted to certain time slots. Thus each takeoff must be scheduled within a prespecified time interval. Therefore, the time window constraint is:

In the previous section, the types of costs and constraints considered were explained. Now we can present the mathematical formulation of the model in Eqs. (19) to (30), which represents a nonlinear program. Here we provide a compact formulation of the objective function using Eqs. (8), and (10) to (12). In Eqs. (20) to (30), we provide the constraints and other previously derived relationships.

a lj < t lj < blj ∀j ∈ J l ∀l ∈ L. (13)

Since limited airport capacity might require a minimum time interval between any two flights, we introduce the following constraint.

MinTC = SC + IC + PC + AC , (19)

subject to: S lj −1 + λ l (t lj − t lj −1 ) = Alj + S lj ∀j ∈ J l ∀l ∈ L, (20)

Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals

567


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

S lj −1 + λ l (t lj − t lj −1 ) = Alj + S lj ∀j ∈ J l ∀l ∈ L, (20)

a lj < t lj < blj ∀j ∈ J l ∀l ∈ L, (21)

t lj − t lj −1 ≥ tmin ∀j ∈ J l ∀l ∈ L, (22)

∑S l∈L

l kl

+ λ l ( pt − s l ) ≤ Sc ms ∀pt ∈ R, (23)

{

}

Alj = max Ac , S lj −1 + λ l (t lj − t lj −1 ) , (24)

s l = max t lj : t lj ≤ pt , (25)

k l = index j : (t lj ≤ pt ∧ t lj +1 ≥ pt ), (26)

bd = ∑ ∑ λ l d , (27)

{

}

each truck route. We seek to optimise the number of takeoffs and corresponding schedule assuming that all the freight arriving at the terminal before the last takeoff has to be transported. For this case, we assume the input data from Table 2. The optimisation results for 4 to 9 takeoffs are presented in Table 3. We present an optimised schedule for six different numbers of takeoffs and corresponding costs in dollars. Please note that within “other costs” we consider storage, penalty, loading and unloading costs. Moreover, by marginal savings in other costs we consider savings in storage, penalty, loading and unloading costs due to introducing an additional roundtrip flight. Table 1. Vehicle size and roundtrip duration Feeder route 1 2 3 4 5 6 7 8 9 10

l∈L j∈J l

l l l l S0= t0= A = S= 0, (28) 0 nl

t lj ∈ R+ ∀j ∈ J l ∀l ∈ L, (29)

nl ∈ Z + ∀l ∈ L. (30)

The total cost function is a function of the number of takeoffs and takeoff times, as explained in the problem statement. The nonlinear model shown in Eqs. (19) to (30) optimises the schedule while taking into consideration the capacity of airplanes, airport and terminal storage, and time windows for takeoffs. In the following section we apply a genetic algorithm (GA) to optimise the schedule in two case studies. Interested readers may refer to [11] and [12] for more information about GA’s. 5 APPLICATION In order to test our model, we designed two case studies. In the first case, the schedule in an intermodal system with a single main route is optimised. In this simplified optimisation problem we examine the anticipated tradeoff in types of costs through sensitivity analysis. In the second case we analyse a complex system with multiple main routes and time windows. 5.1 Case Study with a Single Air Route We analyse a system with ten feeder truck routes connecting to a single airplane main route. In Table 1 we provide the average roundtrip time on each feeder route, as well as the number of vehicles operating on 568

Average roundtrip duration 1 / λ1 [hr] 1.23 1.97 1.73 2.10 2.16 1.94 2.18 1.86 1.68 2.00

Number of trucks in feeder route 3 4 1 2 1 2 5 2 1 3

The results presented in Table 3 show that the minimum total cost occurs in the case with five takeoffs. Therefore we can conclude that at the cost of 7000 $/flight, one more flight than necessary to satisfy the demand should be introduced. Moreover, we can observe that storage, penalty and loading/ unloading cost decrease with the increase in the number of takeoffs. This outcome was expected and it confirmed the tradeoff between types of cost that was explained in the problem statement. We also note that the marginal savings in storage, penalty and loading/ unloading cost decreases with the number of aircraft roundtrips, which is another anticipated outcome. Based on the values for storage, penalty and loading/unloading cost we can explore how different flight costs affect the optimised number of takeoffs and thereby the schedule. In Fig. 3, we plot total cost for the case of 4, 5, 6, 7, 8 and 9 roundtrips vs. aircraft roundtrip cost. Fig. 3 also shows five threshold values for airplane roundtrip cost which determine the optimal number of takeoffs. Those values are 1357, 1871, 2771, 4277 and 7509 dollars, respectively. Clearly, for a relatively low cost per plane roundtrip, the total system cost is optimised by scheduling more takeoffs than necessary to satisfy the demand. As the

Marković, N. – Schonfeld, P.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

Table 2. Input data

Ac C1A

Airplane capacity Flight cost Terminal storage capacity Storage multiple (safety factor) Storage cost

Sc ms CDT d

Amount of time

50 truckloads

In-terminal cost

7,000 $/roundtrip

In-terminal cost

Ctd cti

25 $/truckload 45 $/truckload

1 n

15 hrs 0.8 hrs

87.5 truckloads

Time of the last takeoff

t

0.8

Minimum headway

4 $/truckload-hr

Arrival rate based on the data from Table 1

tmin λ1

15 min

Penalty function

fp (t)

12.95 veh/hr 0 if t≤2 125t-250 if 2<t≤10 1000 if t>10

Table 3. Optimised Schedules and Costs Number of flights

Airline cost [103$]

Other cost [103$]

4 5 6 7 8 9

28 35 42 49 56 63

151 143 139 136 134 133

Marginal savings in other cost NA 8 4 3 2 1

Optimised takeoff times for the given number of flights [hr]

Total cost [103$]

Fl.1

Fl.2

Fl.3

Fl.4

Fl.5

Fl.6

Fl.7

Fl.8

Fl.9

179 178 181 185 190 196

3.6 2.5 2.0 2.0 2.0 2.0

7.3 5.0 4.0 3.6 3.3 3.2

11.1 7.5 6.0 5.2 4.7 4.3

15.0 11.3 8.0 6.8 6.0 5.5

NA 15.0 11.5 8.4 7.4 6.6

NA NA 15.0 11.7 8.7 7.8

NA NA NA 15.0 11.9 8.9

NA NA NA NA 15.0 12.0

NA NA NA NA NA 15.0

Fig. 3. Sensitivity Analysis

airline cost increases, the optimal number of takeoffs decreases until it eventually drops to the minimum number needed to satisfy the demand.

5.2 Case Study with Multiple Airline Routes Here we consider the case of multiple feeder routes connecting to three air routes of similar lengths. Since

Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals

569


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

we consider the case with three dozen feeder routes, we do not present the expected roundtrip duration and the number of vehicles on each route, as we did in the previous example. Instead, we provide the computed arrival rates of vehicles connecting to three air routes. Moreover, we assume the same values from the previous numerical example, but this time we include time windows into our analysis. The aforementioned data are provided in Table 4. Please note again that, for each route, all the freight arriving at the terminal by the time of the last flight has to be loaded into airplanes and transported to its destination.

the cases when no feasible solution is found either due to overloading of the terminal storage or due to not delivering all the freight that has arrived by the time of the last takeoff on each air route. From Table 5 we conclude that at 9000 $/flight, the total cost is optimised by scheduling 4 flights on each main route (case 5), which also equals the minimum number of flights needed to provide a feasible solution. Finally, in Table 6 we provide the optimised schedule for all 8 feasible combinations of flights from Table 5. Table 6. Optimised schedule

Table 4. Input data Case

Route

Optimised Schedule on Route

Flight cost on all routes

C1A

9,000 $

Arrival rate on route 1 Arrival rate on route 2 Arrival rate on route 3

λ1 λ2 λ3

8.0 veh/hr 8.5 veh/hr 7.5 veh/hr

(a11 , b11 )

(1.5,2.5)

(a12 , b21 )

(3.1,4.5)

(a12 , b12 )

3

3.19 | 6.22 | 9.25 | 12.63 | 16.00

(1.2,3.6)

1

2.17 | 4.50 | 8.75 | 15.00

(a22 , b22 )

(4.5,6.4)

2

2.69 | 5.12 | 7.57 | 10.79 | 14.00

3

3.17 | 6.25 | 9.33 | 16.00

1

2.19 | 4.50 | 7.27 | 11.13 | 15.00

2

2.69 | 5.40 | 8.12 | 14.00

Time windows for Route 1

Time windows for Route 2

2 3

2 3

(7.0,10.0)

3 1

3 1

(a , b )

(2.3,4.2)

(a23 , b23 )

(5.2,6.5)

Minimum time interval between any two flights

tmin

0.5 hours

Last takeoff on Route 1

tn1

15 hours

Last takeoff on Route 2

tn2

14 hours

Last takeoff on Route 3

tn3

16 hours

(a , b ) Time windows for Route 3

Case 1 2 3 4 5 6 7 8 9 10 11 12

Total Cost [103$] NA NA NA NA 378 387 386 384 390 389 394 396

The optimization results for the case study with three main routes are given in Table 5. NA stands for 570

6

7

8

9

10

Table 5. Optimised cost Number of Flights on Route l l=1 l=2 l=3 3 3 3 4 3 3 3 4 3 3 3 4 4 4 4 4 4 5 4 5 4 5 4 4 5 5 4 5 4 5 4 5 5 5 5 5

5

11

12

1

2.19 | 4.50 | 8.75 | 15.00

2

2.69 | 5.51 | 8.12 | 14.00

3

3.19 | 6.26 | 9.33 | 16.00

1

2.19 | 4.50 | 8.75 | 15.00

2

2.69 | 5.40 | 8.12 | 14.00

3

3.19 | 6.26 | 9.33 | 16.00

1

2.15 | 4.43 | 6.71 | 8.73 | 15.00

2

2.65 | 5.04 | 7.44 | 9.83 | 14.00

3

3.15 | 6.21 | 9.33 | 16.00

1

2.19 | 4.37 | 6.56 | 8.75 | 15.00

2

2.69 | 5.34 | 8.12 | 14.00

3

3.19 | 5.84 | 7.62 | 11.81 | 16.00

1

2.18 | 4.50 | 8.75 | 15.00

2

2.68 | 5.04 | 7.40 | 9.75 | 14.00

3

3.18 | 6.21 | 9.25 | 12.63 | 16.00

1

2.01 | 4.27 | 6.51 | 8.75 | 15.00

2

2.51 | 4.77 | 7.01 | 8.25 | 14.00

3

3.01 | 5.37 | 7.72 | 11.86 | 16.00

6 CONCLUSIONS This paper studied the recovery of a single-terminal intermodal freight system from a disruption. A model was developed that optimises the schedule of vehicles on main routes assuming Poisson arrivals on feeder routes. A genetic algorithm was used to optimise several case studies and sensitivity analysis confirmed

Marković, N. – Schonfeld, P.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

the anticipated tradeoff in types of cost. Moreover, the model developed in this paper was applied to case studies including feeder truck routes and main airline routes. However, this model could be applied to other combinations of transportation modes without many modifications, as long as the arrivals do not deviate much from Poisson distributions. Since the arrivals were modeled as a Poisson process, the computational efficiency of the model is fairly insensitive to the number of feeder routes or operating vehicles. Therefore the proposed scheduling model can be successfully applied to optimise the performance of busy intermodal systems with numerous vehicle arrivals. Several assumptions built into this paper may be relaxed in the future in order to make the model more general. The current model could be improved to provide good robust solutions even for the case when some of the three properties of the Poisson process listed in Section 1 do not hold. Moreover, the current analysis assumes fixed numbers of vehicles operating on the feeder routes. Future work may consider variable fleet sizes on feeder routes and thereby nonstationary arrival intensities. 7 NOTATION The following symbols are used in this paper: λil parameter of the exponentially distributed duration of truck roundtrip on feeder route i connecting to main route l i index of feeder route l set of feeder routes Il set of feeder routes connecting to main route l; clearly I l ∈ I j index of takeoffs on route l Jl set of takeoffs on route l l index of main route kl index of the takeoff on route l prior to time pt L set of main routes tj time of the jth takeoff on route l nl number of takeoffs on main route l ni number of trucks on feeder route i ri arrival rate on feeder route i Ac capacity of an airplane Sc capacity of terminal’s storage ms storage multiple CDT in-terminal dwell cost C Al flight cost on route l SCl storage cost associated with freight connecting to main route l SC storage cost

d the amount of time such that a truck arriving within the (tj – d,tj) interval will take its truckload directly to the airplane S lj inventory level of freight connecting to main route l after the jth takeoff bd the expected number of truckloads that will be transferred directly from trucks to airplanes Cti cost of in-terminal operations Ctd cost of in-terminal operations when the truck takes its truckload directly to the airplane IC overall cost for in-terminal operations fp ( t lj ) penalty function per truckload loaded into airplane at moment t lj PC overall penalty cost AC overall airline cost TC total cost tmin minimum time interval between any two takeoffs a lj the lower bound for the jth takeoff on route l blj the upper bound for the jth takeoff on route l Alj amount of freight carried in the jth takeoff on route l pt control parameter used to check the inventory level R+ set of nonnegative real numbers Z+ set of nonnegative integers 8 ACKNOWLEDGMENTS This work was sponsored by the U. S. Department of Transportation (USDOT) and the Maryland State Highway Administration (MSHA). The authors thank the Center for Integrated Transportation Systems Management (CITSM) of the University of Maryland for its support. 9 REFERENCES [1] Li, W., Glazebrook, D.K. (1998). On stochastic machine scheduling with general distribution assumptions. European Journal of Operations Research, vol. 105, no. 3, p. 525-536, DOI:10.1016/S0377-2217(97)00088-X. [2] Petkov, S., Maranas, C. (1997). multiperiod planning and scheduling of multiproduct batch plants under demand uncertainty. Industrial & Engineering Chemistry Research, vol. 36, p. 4864-4881, DOI:10.1021/ie970259z. [3] Pundoor, G., Chen, Z.-L. (2009). Joint cyclic production and delivery scheduling in a two-stage supply chain. International Journal of Production Economics, vol. 119, no. 1, p. 55-74, DOI:10.1016/j.ijpe.2009.01.007. [4] Ting, C.J., Schonfeld, P. (2005). Schedule coordination in a multiple terminal transit network. Journal of Urban Planning and Development, vol. 131, no. 2, p. 112-124, DOI:10.1061/(ASCE)0733-9488(2005)131:2(112).

Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals

571


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, 564-572

[5] Chen, C.C., Schonfeld, P. (2010). Modeling and performance assessment of intermodal transfers at cargo terminals. Transportation Research Record: Journal of the Transportation Research Board, vol. 2162, no. 1, p. 53-62, DOI:10.3141/2162-07. [6] Teodorović, D. (1988). Simultaneous determining departure time and flight frequency on a route. Airline Operations Research, Gordon and Breach Science, New York, p. 131-145. [7] Chang, S.W., Schonfeld, P. (2004). Optimized schedules for airline routes. Journal of Transportation Engineering, ASCE, vol. 130, no. 4, p. 412-418, DOI:10.1061/(ASCE)0733-947X(2004)130:4(412). [8] Marković, N., Schonfeld, P. (2011). Scheduling under uncertainty in a single-hub intermodal freight system.

Transportation Research Record: Journal of the Transportation Research Board, vol. 2238, no.1, p. 2431, DOI:10.3141/2238-04. [9] Vukadinović, S. (1988). Queuing, Naučna Knjiga, Beograd. In Serbian [10] Wolff, R. (1989). Stochastic modeling and the theory of queues. Prentice-Hall, Englewood Cliffs. [11] Goldberg, D.E. (1995). Genetic algorithms in search, optimization and machine learning. Addison Wesley, Boston,. [12] Michalewicz, Z. (1996). Genetic algorithms + data structure = evolution programs, 3rd ed., SpringerVerlag, London, DOI:10.1007/978-3-662-03315-9.

*Accepted for publication: 2011-09-05, Copyright received: 2013-03-05.

572

Marković, N. – Schonfeld, P.


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9 Vsebina

Vsebina Strojniški vestnik - Journal of Mechanical Engineering letnik 59, (2013), številka 9 Ljubljana, september 2013 ISSN 0039-2480 Izhaja mesečno

Gostujoči uvodnik

SI 97

Razširjeni povzetki člankov Saut Gurning, Stephen Cahoon, Branislav Dragovic, Hong-Oanh Nguyen: Modeliranje večtirnih strategij blažitve posledic motenj v pomorskih verigah oskrbe s pšenico Dimitrios Lyridis, Panayotis Zacharioudakis, Stylianos Iordanis, Sophia Daleziou: Modeliranje časovnih vrst terminskih pogodb na prevozne stroške z uporabo umetnih nevronskih mrež Francesco Longo, Aida Huerta, Letizia Nicoletti: Analiza uspešnosti južnomediteranskega pristanišča s simulacijo diskretnih dogodkov Davorin Kofjač, Maja Škurić, Branislav Dragović, Andrej Škraba: Modeliranje in ocena učinkovitosti prometa v potniškem pristanišču Kotor Bo Lu, Nam Kyu Park: Identifikacija kritičnih dejavnikov produktivnosti kontejnerskih terminalov z analizo občutljivosti Andrija Vujičić, Nenad Zrnić, Boris Jerman: Trajnostnost pristanišč: analiza življenjskega cikla opreme za prekladanje tovora z ničelnimi emisijami Eugeniusz Rusiński, Przemysław Moczko, Damian Pietrusiak, Grzegorz Przybyłek: Eksperimentalna in numerična preiskava utrujenostnega zloma nosilne konstrukcije čeljustnega drobilnika Nikola Marković, Paul Schonfeld: Časovno razporejanje za obnovo delovanja intermodalnega sistema z enim terminalom in prihodi po Poissonovi porazdelitvi Osebne vesti Doktorske disertacije, znanstvena magistrska dela, diplomske naloge

SI 99 SI 100 SI 101 SI 102 SI 103 SI 104 SI 105 SI 106

SI 107


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9 Gostujoči uvodnik

Gostujoči uvodnik Tematska številka: Pomorska in pristaniška logistika Mednarodna konferenca MHCL 2012 (International Conference on Material Handling, Constructions and Logistics 2012) je bila svojevrstno srečanje akademikov iz različnih okolij in znanstvenih področij ter je omogočila diskusijo o aktualnih vprašanjih, povezanih s transportnimi napravami in sistemi, gradbeno in rudarsko opremo, projektiranjem in konstruiranjem, tehnično logistiko ter tudi pomorsko in pristaniško logistiko (MPL – Maritime and Port Logistics). Gostujoči uredniki tematske številke smo med konferenčnimi prispevki izbrali šest člankov, povezanih z MPL, ter en članek v zvezi z gradbeno in rudarsko opremo. Vključen je tudi članek, ki je bil za objavo v reviji sprejet že prej, po vsebini pa sodi v to tematsko številko. V naslednjih odstavkih so prikazani poudarki posameznih izbranih člankov. Prvi članek Modeliranje večtirnih strategij blažitve posledic motenj v pomorskih verigah oskrbe s pšenico, katerega avtorji so Saut Gurning in kolegi, proučuje primernost štirih glavnih strategij za ublažitev morebitnih motenj v verigi oskrbe s pšenico. Za modeliranje tveganj v oskrbovalni verigi so uporabljeni Markovski procesi, ti modeli pa so nadalje vključeni v analizo strategij obvladovanja motenj. Pričujoči članek je nov prispevek k raziskavam, ki uporabljajo stohastični pristop k obvladovanju procesov z več motnjami. Drugi članek Modeliranje časovnih vrst terminskih pogodb na prevozne stroške z uporabo umetnih nevronskih mrež, katerega avtorji so Dimitrios Lyridis in kolegi, napoveduje prihodnje cene terminskih pogodb na transportne storitve (freight derivatives) na osnovi uporabe umetnih nevronskih mrež. Za izgradnjo in učenje nevronske mreže so uporabljeni znani podatki iz obdobja med januarjem 2005 in marcem 2009. Izdelani model investitorju predlaga, kateri položaj naj zavzame na trgu s terminskimi pogodbami na transportne storitve (kratko pozicijo za prodajo pogodb ali dolgo pozicijo za njihov nakup). Tretji članek Analiza zmogljivosti južnomediteranskega pristanišča s simulacijo diskretnih dogodkov, katerega avtorji so Francesco Longo in kolegi, prikazuje simulacijski model kompleksnega srednje velikega sredozemskega pristanišča in analizira razvoj zmogljivosti takega sistema, s posebnim poudarkom na času pretovora ladij (turnaround time of the ships). S pomočjo simulacije so bili preverjeni vplivi nekaterih ključnih dejavnikov (t. j. časi med prihodi posameznih ladij, časi natovarjanja / raztovarjanja, število avtomobilov in tovornjakov, ki naj se raztovorijo/natovorijo) na delovanje pristanišča. Četrti članek v tej tematski številki je Modeliranje in ocena učinkovitosti prometa v potniškem pristanišču Kotor, katerega avtorji so Davorin Kofjač in kolegi. Za razliko od prejšnjih raziskav pristaniške logistike za velike potniške križarke ta članek predstavlja simulacijo prometa v kotorskem pristanišču za križarke. Simulacija je uporabljena za vrednotenje zmogljivosti pristanišča in optimizacijo njegovega delovanja. Nadalje je z namenom čim večjega povečanja celotnih prihodkov pristanišča uvedena kompleksna funkcija kriterija prihodkov. Izvedenih je več simulacij, ​​ kjer so ocenjeni scenariji z razširjenim glavnim pomolom in povečano intenzivnostjo prometa. Peti izbrani članek ima naslov Identifikacija kritičnih dejavnikov produktivnosti kontejnerskih terminalov z analizo občutljivosti, avtorja pa sta Bo Lu in Nam Kyu Park. Glavni namen tega članka je določiti različne dejavnike produktivnosti, da bi dosegli pozitivnejši vpliv na produktivnost kontejnerskega terminala. Analiza občutljivosti predstavlja primerno merilo ugotavljanja kritičnih dejavnikov za izboljšanje produktivnosti. Raziskava zajema 28 večjih vzhodnoazijskih kontejnerskih terminalov. Primerjava rezultatov njihove učinkovitosti je bila izvedena z analizo ovojnice podatkov (data envelopment analysis) in regresijsko analizo. Šesti članek Trajnostnost pristanišč: analiza življenjskega cikla opreme za prekladanje tovora z ničelnimi emisijami, katerega avtorji so Andrija Vujičić in kolegi, raziskuje zgodbo v ozadju koncepta ničelnih emisij. Za raziskavo sta bila izbrana dva na kontejnerskih terminalih najpogostejša stroja: portalni žerjav s pnevmatikami in kontejnerski polpriklopnik (utility tractor rig). Analizirani in primerjani sta bili njuni konvencionalni izvedbi in njuni različici z ničelno emisijo. Prehod z dizelsko gnane opreme na električno opremo načeloma lahko šteje za korak naprej v zvezi s konceptom trajnostnega razvoja. Da je bilo možno priti do končnih sklepov, je bilo vseeno nujno izključiti nekatere trajnostne vplive, povezane s prehodom na nov vir energije. SI 97


Sedmi članek te tematske številke je Eksperimentalna in numerična preiskava utrujenostnega zloma nosilne konstrukcije čeljustnega drobilnika, katerega avtorji so Eugeniusz Rusiński in kolegi. Članek eksperimentalno in numerično raziskuje razloge nevarnih utrujenostnih odpovedi nosilne konstrukcije drobilnika (jaw crusher). Nova zasnova konstrukcije je bila raziskana glede na dinamične in utrujenostne lastnosti. Analiza je potekala s pomočjo modificiranega modela po metodi končnih elementov. Spremenjena konstrukcija je bila tudi preskušena in med delovanjem so bile opravljene meritve vibracij. Rezultati meritev so potrdili ustrezno obnašanje konstrukcije brez resonančnih problemov. Zadnji članek Časovno razporejanje za obnovo delovanja intermodalnega sistema z enim terminalom in prihodi po Poissonovi porazdelitvi, katerega avtorja sta Nikola Marković in Paul Schonfeld, obravnava vzpostavitev delovanja enoterminalskega intermodalnega tovornega sistema po prekinitvi delovanja zaradi motenj. Razvit je bil model, ki optimizira razpored vozil na glavnih progah ob predpostavki Poissonove porazdelitve prihodov na posluževalne poti. Za optimizacijo številnih proučevanih primerov je bil uporabljen genetski algoritem. Analiza občutljivosti je potrdila pričakovane spremembe v vrstah stroškov. Zahvaljujemo se uredništvu Strojniškega vestnika – Journal of Mechanical Engineering in urednikom konferenčnega zbornika za njihovo trdno podporo in spodbudo pri pripravi te tematske številke. Prav tako izražamo hvaležnost vsem avtorjem in anonimnim recenzentom člankov za njihov dragoceni čas in vloženi trud. Gostujoči uredniki: prof. dr. Nam-Kyu Park prof. dr. Branislav Dragović doc. dr. Boris Jerman

SI 98


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 99 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo: 2012-12-30 Prejeto popravljeno: 2013-05-06 Odobreno za objavo: 2013-05-17

Modeliranje večtirnih strategij blažitve posledic motenj v pomorskih verigah oskrbe s pšenico Gurning, S. – Cahoon, S. – Dragovic, B. – Nguyen, H.-O. Saut Gurning1 – Stephen Cahoon2,* – Branislav Dragovic3 – Hong-Oanh Nguyen2 1 Sepuluh

Nopember tehnološki inštitut, Indonezija univerza, Avstralska visoka šola za pomorstvo, Avstralija 3 Univerza v Črni gori, Fakulteta za pomorstvo, Črna gora

2 Tasmanska

Z globalizacijo verig oskrbe s pšenico (WSC) je povečano tveganje motenj v pomorskem delu verig postalo glavni omejitveni dejavnik za učinkovit transport pšenice od proizvajalcev (kmetov) do globalnih končnih uporabnikov. Motnje v pomorskem delu vključujejo tveganja, povezana z varnostjo in varovanjem, okoljem, infrastrukturo, trgi, organizacijami in vodstvenimi dejavniki. Deležniki v dobavnih verigah pa kratkoročnih in dolgoročnih vplivov teh motenj v veliki meri ne razumejo ali pa se jim sploh ne posvečajo. Namen tega članka je premostiti vrzel v raziskavah z modelom markovskega procesa. Z raziskovanjem in razumevanjem vzrokov in posledic motenj v pomorskem delu se lahko deležniki v dobavni verigi bolje pripravijo za upravljanje z izzivi, ki jih predstavljajo motnje v pomorskem delu, ter prepoznajo prednosti razvoja strategij za ublažitev posledic teh motenj. Članek obravnava možne strategije za ublažitev posledic motenj, do katerih prihaja pri kontejnerskem transportu pšenice v morskem delu dobavne verige. Obravnava štiri glavne strategije za blažitev posledic (zaloge in dobavni viri, spreminjanje poti v primeru nenačrtovanih okoliščin, načrtovanje obnovitve in načrtovanje neprekinjenega poslovanja) ter njihovo primernost za upravljanje z morebitnimi motnjami v verigi dobave pšenice. Za modeliranje tveganj v dobavni verigi so bili uporabljeni markovski procesi, ki so nato bili združeni v okvir za strategije v primeru motenj v dobavni verigi. Ta okvir je nato bil uporabljen za analizo in vrednotenje strategij blažitve posledic motenj v verigi oskrbe s pšenico. Rezultati uporabe markovskih verig v zveznem časovnem obdobju kažejo, da strategije blažitve, optimizirane za obdobje enega leta, vključujejo merjenje in napovedovanje stroškovnih in časovnih funkcij WSC. Poslovni subjekti v pomorskem delu verige lahko z uporabo večtirnih strateških scenarijev v splošnem zmanjšajo stopnjo tveganja na 3 do 5%, kmetje in končni porabniki pa so obremenjeni s pričakovano stopnjo tveganja v pomorskem delu 21 do 22%. Stopnja tveganja motenj v pomorskem delu za trgovce in predelovalce je 10 do 11%. Scenariji za upravljanje z motnjami v tem članku so bili ustvarjeni v načinu naknadne obdelave in uveljavljeni v enofrekvenčnem omrežju za preučitev učinkovitosti predlagane metodologije obdelave podatkov. V primeru dolgoročne uveljavitve takšnega pomorskega nadzornega sistema na osnovi blažitve posledic pa deležniki v trgovini s pšenico pričakujejo rešitev, ki bo delovala pravzaprav v realnem času. Rezultate upravljanja s pomorskimi motnjami je nato mogoče dosledno analizirati ter poiskati različne kazalnike in predstaviti rezultate na lahko razumljiv način, na primer z vrednotenjem prenašanja tveganj. Poleg tega niso bili zabeleženi procesi sprejemanja odločitev na področju sodelovanja in koordinacije ter stroški pridobivanja informacij, kar je morda vplivalo na interpretacijo statusa pomorskih motenj. V prihodnjih raziskavah bodo lahko vključena druga merila vedenj na ravni omrežja ali skupnosti, kjer prihaja do motenj v pomorskem delu, kakor tudi eksperimentalne študije odnosov med nameni in vedenji za krizne situacije. Uporaba večtirnih scenarijev upravljanja z motnjami v dobavni verigi je prispevek k obstoječi literaturi o stohastičnem pristopu v markovskih procesih odločanja. Čeprav članek obravnava dobavno verigo pšenice, je metodologija torej uporabna tudi za najrazličnejše probleme v drugih dobavnih verigah, zlasti pri žitu in kmetijskih pridelkih, ki se transportirajo kot razsuti tovor ali v kontejnerjih. Proces omogoča tudi preučitev nabora domnevnih notranjih in zunanjih dejavnikov, ki povzročajo motnje v verigi oskrbe s pšenico. Domneve, povezane s temi tveganji motenj, vključujejo tudi izvore tveganja motenj ter njihovo verjetnost in posledice. Ključne besede: kontejnerske verige oskrbe s pšenico, večtirne strategije blažitve posledic, motnje v pomorskem delu

*Naslov avtorja za dopisovanje: Tasmanska univerza, Avstralska visoka šola za pomorstvo, Locked Bag 1397, Launceston 7250, Avstralija, S.Cahoon@amc.edu.au

SI 99


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 100 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo 2013-01-02 Prejeto popravljeno: 2013-06-06 Odobreno za objavo: 2013-06-12

Modeliranje časovnih vrst terminskih pogodb na prevozne stroške z uporabo umetnih nevronskih mrež Dimitrios Lyridis1 – Panayotis Zacharioudakis1 – Stylianos Iordanis1 – Sophia Daleziou2 1 Nacionalna 2 Nacionalna

tehniška univerza v Atenah, Laboratorij za pomorski transport, Grčija tehniška univerza v Atenah, Šola za aplikativne matematične in fizikalne vede, Grčija

Trg izvedenih finančnih instrumentov na področju ladijskega prevoza se je v zadnjih 30 letih močno razširil. Izvedeni finančni instrumenti so pogodbe, ki vsem udeležencem na trgu ladijskega prevoza omogočajo zmanjšanje izpostavitve do nihanj stroškov prevoza, cen goriva, obrestnih mer, menjalnih razmerij in cen ladij. V članku so uporabljene umetne nevronske mreže (ANN) za napovedovanje prihodnjih cen izvedenih finančnih instrumentov na prevozne stroške. Na osnovi zgodovinskih podatkov iz obdobja med januarjem 2005 in marcem 2009 je bila zgrajena in izučena ANN, ki daje dva različna rezultata. Investitor si lahko z pomaga z modelom pri odločanju o tem, katero pozicijo naj zavzame na trgu izvedenih finančnih instrumentov (kratko za prodajo pogodb ali dolgo za nakup pogodb). Eden glavnih namenov članka je preučiti, ali so v podatkih preteklih časovnih vrst cene trimesečnih pogodb transportne poti za suhi razsuti tovor, ki jih izdaja IMAREX po indeksih Baltic Exhange, tudi informacije, ki bi bile uporabne za napovedovanje, ter, če je res tako, v kakšni meri bi bile te informacije uporabne za uveljavitev uspešne strategije trgovanja z najzmogljivejšimi kvantitavnimi metodami, t. j. z umetnimi nevronskimi mrežami. Naša metodologija analize podatkov in naknadnega preverjanja je pokazala, da so lahko nevronske mreže nepogrešljivo orodje za napovedovanje gibanja izvedenih finančnih instrumentov na stroške prevoza in s tem za uspešne naložbe, kot je znano tudi iz širše dostopne literature. Prvič smo zbrali te matematične tehnike, da bi ustvarili praktično strategijo za trgovanje z izvedenimi finančnimi instrumenti na prevozne stroške za poti suhega razsutega tovora z ladjami velikosti capesize, panamax, handymax in supermax. V model smo vključili eksogene spremenljivke (povezane z ekonomiko ladijskega prevoza), kot so naročila ladij za razsuti tovor capesize, razvoj flote ladij za razsuti tovor capesize, vrednost ladij capesize za razrez, enoletni stroški prevoza z ladjo za razsuti tovor s 170.000 dwt, število pogodb za ladje capesize, prodaja ladij capesize, število dobav ladij capesize in cena novih ladij velikosti capesize. Napovedi pa morajo vedno služiti le kot pripomoček in pri njihovi uporabi je potrebna posebna skrb. Ne smemo pozabiti, da modeli nastanejo s samoučenjem ter da na osnovi zgodovinskih podatkov napovedujejo prihodnost, ki pa lahko prinaša tudi presenečenja. Vedno je priporočljiva uporaba modelov v kombinaciji z empiričnim znanjem in človeško presojo, zlasti v času fundamentalnih strukturnih preobratov. Iz kvantitativnih rezultatov razvitega modela je razvidno, da se uporabljeni model povezav dobro ujema z dinamiko časovnih vrst in da daje zadovoljivo natančnost z visoko stopnjo uspeha v kontekstu trgovalnih strategij. Metodologija je enostavno razširljiva tudi na multivariatne modele, ki napovedujejo vektorske časovne vrste, torej jo je mogoče posplošiti za modeliranje več sistemskih spremenljivk na trgu ladijskega prevoza ter prilagoditi tudi za napovedovanje na trgih drugih finančnih instrumentov. Arhitektura MPL je vsekakor primerna za takšno nalogo, če je realizirana z ustrezno kompleksnostjo slojev, z zadostno računsko močjo, ter je model oplemeniten z več eksogenimi pojasnjevalnimi spremenljivkami in odvisnimi spremenljivkami. Arhitektura modela bo v prihodnjih raziskavah preizkušena tudi na kompozitnih indeksih, sestavljenih iz enakomerno uteženih časovnih vrst dobička več ladjarskih podjetij. Praktična uporaba modela bi bila možna pri svetovalcih za terminske pogodbe (CTA), trgovcih in ladjarjih, kakor tudi pri vseh tistih, ki želijo dosegati boljše rezultate kot večina ostalih udeležencev na trgu s klasičnimi hevrističnimi tehničnimi metodami. Kvantitativna orodja, ki jih uporabljajo agenti, namreč postajajo vse bolj sofisticirana. Ključne besede: stroški prevoza, strategija trgovanja, umetne nevronske mreže, modeliranje trga ladijskega prevoza, napovedovanje stroškov prevoza

SI 100

*Naslov avtorja za dopisovanje: Nacionalna tehniška univerza v Atenah, Laboratorij za pomorski transport, Grčija, dsvlr@mail.ntua,gr


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 101 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo: 2013-01-09 Prejeto popravljeno: 2013-06-05 Odobreno za objavo: 2013-07-05

Analiza uspešnosti južnomediteranskega pristanišča s simulacijo diskretnih dogodkov Francesco Longo1,* Aida Huerta2, Letizia Nicoletti1 1 Univerza

v Kalabriji, Oddelek za strojništvo, energetiko in management, Italija nacionalna univerza, Oddelek za sistemski inženiring, Mehika

2 Mehiška

Modeliranje in simulacije (M & S) so dokazano nepogrešljivo orodje za snovanje, upravljanje in nadzor kompleksnih sistemov. Cilj predlagane raziskovalne študije je razvoj simulacijskega modela, ki povzema kompleksnost srednje velikega sredozemskega pristanišča, ter analiza zmogljivosti sistema, pri čemer je v središču pozornosti čas pretovora ladij. Po analizi vhodnih podatkov, razvoju, verifikaciji in validaciji simulacijskega modela je bila opravljena zasnova eksperimenta (24-faktorska zasnova) za ovrednotenje vpliva nekaterih kritičnih dejavnikov (časi med prihodi posameznih ladij, časi natovarjanja/raztovarjanja, število avtomobilov in tovornjakov, ki jih je treba raztovoriti/natovoriti) na zmogljivost pristanišča. V ta namen je bila opravljena analiza variance ter je bil določen analitični model vhodov in izhodov, ki je uporaben za vrednotenje zmogljivosti sistema. Prvi korak je bil razvoj konceptualnega modela pristanišča. Predmet raziskave je bilo srednje veliko sredozemsko pristanišče (v Salernu) s strateško logistično lego sredi Sredozemlja, ki ima ključno vlogo pri trgovinski izmenjavi Južne Italije. V konceptualnem modelu pristanišča so bili zajeti tako tisti vidiki realnega sistema, ki jih je treba vključiti v računalniški model (simulacijski model), kakor tudi tisti, ki jih je iz njega treba izpustiti. V naslednjem koraku so bili zbrani podatki in analizirani vhodni podatki. Za namene te študije so bili v simulacijskem modelu zbrani in uporabljeni podatki o dejanskih prihodih ladij od 1. januarja 2010 do 31. decembra 2011 ter od 1. januarja 2012 do 14. maja 2012. Konceptualni model je bil preveden v simulacijski model z rešitvijo Anylogic® (različica 6.4), ki so jo razvili pri XJ Technologies. Simulacijski model vključuje diagram poteka, animacijo, grafični uporabniški vmesnik in izhod simulacije. Simulacijski model je bil po verifikaciji in validaciji uporabljen za preučitev vpliva sprememb časa med prihodi ladij, časa razkladanja/nakladanja ladij Ro-Ro/Pax, števila avtomobilov in števila tovornjakov na čas pretovora ladij. Rezultati simulacije (analizirani po metodi analize variance) so jasno pokazali, da so čas med prihodi ladij, čas razkladanja/nakladanja Ro-Ro/Pax ladij, število avtomobilov in število tovornjakov ključni dejavniki zmogljivosti pristanišč z ozirom na čas pretovora. Ovrednoten je bil tudi analitični metamodel, ki povezuje čas pretovora z vhodnimi dejavniki. Metamodel je dodatno orodje, ki se lahko uporablja poleg simulacijskega modela (ali skupaj z njim) za preučevanje vpliva vhodnih dejavnikov na vedenje pristanišč. Ključne besede: logistika, pomorska pristanišča, dobavna veriga, modeliranje in simulacija

*Naslov avtorja za dopisovanje: Univerza v Kalabiji, Ulica P. Bucci, Cube 44C, 87036 Rende (CS), Italija, f.longo@unical.it

SI 101


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 102 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo: 2012-12-31 Prejeto popravljeno: 2013-05-24 Odobreno za objavo: 2013-06-24

Modeliranje in ocena učinkovitosti prometa v potniškem pristanišču Kotor Kofjač, D. – Škurić, M. – Dragović, B. – Škraba, A. Davorin Kofjač1,* – Maja Škurić2 – Branislav Dragović2 – Andrej Škraba1 1 Univerza

v Mariboru, Fakulteta za organizacijske vede, Slovenija

2 Univerza

v Črni Gori, Pomorska fakulteta, Črna gora

Prispevek obravnava modeliranje in simulacijo prometa v potniškem pristanišču Kotor z namenom oceniti in optimirati operativno politiko. Poleti se v primerjavi z ostalim delom leta poveča intenzivnost prihodov ladij. Povečana intenzivnost prometa povzroča zastoje na sidriščih, kar lahko privede do znatnega nezadovoljstva med lastniki ladij in potniki. Cilj prispevka je upravičiti investicijo v podaljšanje glavnega priveza s stalnim sidrom ali mostiščem, za skrajšanje vrst na sidriščih in povečanje prihodkov pristanišča. V raziskavi smo predvideli podaljšanje glavnega priveza iz 380 na 500 m (celotna dolžina priveza), da bi omogočili hkratno strežbo dveh večjih potniških ladij. Trenutno je izvedljivo le predlagano podaljšanje na 500 m, ker je največja globina priveza na desni strani le 5 m, medtem ko je na levi strani zaradi varnega obračanja ladij v zalivu na voljo le 120 m. Podaljšanje priveza bi še posebej v visoki sezoni, ko je frekvenca prihodov teh ladij višja, omogočilo fleksibilnejše razporejanje ladij na privezu. Za razliko od prejšnjih raziskav na področju logistike potniških pristanišč smo v prispevku predstavili dogodkovni simulacijski model prometa potniških ladij. Promet je definiran tudi z analitičnega vidika z uporabo teorije čakalnih vrst. Vpeljali smo tudi kompleksno kriterijsko funkcijo prihodkov v potniškem pristanišču z namenom maksimirati skupne prihodke. Prihodi in strežba v pristanišču potekajo po strežnem sistemu M/D/1/m. Najprej predpostavimo, da je privez zaseden in da morajo vse potniške ladje v čakalno vrsto, dokler se glavni privez ne sprosti. Naslednja predpostavka je, da ladje v čakalni vrsti na sidrišču ne priplujejo nujno na glavni privez. Ladje so določen čas strežene na sidriščih. Obstaja pa redka možnost, da se glavni privez sprosti in takrat lahko ladja iz sidrišča preide na glavni privez, vendar pa se skupni čas strežbe te ladje ne spremeni. Ko je ladja strežena, izračunamo prihodke za pristanišče na račun te ladje. Ladja ob izteku časa strežbe zapusti pristanišče. Poudarimo še, da se je v analizi dejanskih podatkov izkazala možnost uporabe determinističnega časa strežbe – deset ur za leto 2011 in devet ur za leto 2012 – saj potniške ladje v skoraj vseh primerih priplujejo zgodaj zjutraj in nato odplujejo pozno popoldan oz. zvečer. Simulacijski model je bil razvit s programsko opremo za dogodkovno simulacijo Flexsim. Vhodni podatki za simulacijski model se nanašajo na obdobje osmih mesecev v letu 2011 in obdobje sedmih mesecev v letu 2012. Model vključuje ladje z različnimi dolžinami, bruto tonažo ter vzorci prihodov in strežbe, kakor tudi podatke o stroških, ki jih imajo le-te v pristanišču. Model je bil verificiran in validiran glede na dejanske podatke. Izvedli smo simulacijske scenarije, kjer smo predvideli podaljšan privez in povečano intenzivnost prometa. Na podlagi simulacijskih rezultatov lahko trdimo sledeče. Prvič, scenarij s podaljšanim glavnim privezom omogoča večje prihodke za pristanišče kot scenarij z obstoječim privezom. Drugič, večji delež ladij na glavnem privezu bistveno zmanjša število ladij na sidriščih, ker zmanjšuje možnost nesreč ob uporabi pomožnih čolnov za prevoz potnikov na kopno. To bi lahko pomembno vplivalo tudi na prihodke od turizma v mestu Kotor. Tretjič, s predlaganim podaljšanim privezom bi v pristanišču lahko sprejeli tudi do 40% več ladij. Razviti simulacijski model lahko posplošimo in ga uporabimo tudi za simulacijo učinkovitosti ostalih potniških pristanišč, na primer s prilagoditvijo kapacitete čakalne vrste ter parametrov kriterijske funkcije prihodkov. Model lahko tudi nadgradimo za raziskavo ostalih segmentov pristanišča, kot so na primer vpliv povečanega števila izkrcanih turistov na glavni privez za lokalno gospodarstvo, izgradnja potniškega terminala, vpliv emisij plinov, upravljanje vodnega balasta itd. Ključne besede: modeliranje prometa, simulacije, validacija, prihodki pristanišča, operativna politika, potniško pristanišče Kotor

SI 102

*Naslov avtorja za dopisovanje: Univerza v Mariboru, Fakulteta za organizacijske vede, Kidričeva cesta 55a, 4000 Kranj, Slovenija, davorin.kofjac@fov.uni-mb.si


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 103 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo: 2012-12-21 Prejeto popravljeno: 2013-05-04 Odobreno za objavo: 2013-06-04

Identifikacija kritičnih dejavnikov produktivnosti kontejnerskih terminalov z analizo občutljivosti Lu, B. – Park, N.K. Bo Lu1,* – Nam Kyu Park2

1 Univerza

v Dalianu, Institut za elektronsko trgovino in sodobno logistiko, Kitajska Tongmyong, Šola za pomorski promet in logistiko, Južna Koreja

2 Univerza

Veliki kontejnerski terminali imajo velik vpliv na globalni kontejnerski promet. Njihovi upravitelji skušajo opredeliti različne dejavnike produktivnosti in med njimi poiskati tiste, ki pozitivno vplivajo na produktivnost kontejnerskih terminalov. Povečanje kontejnerskega prometa je izziv za operaterje terminalov, ki morajo predvideti zmogljivost privezov ter izvajati naložbe v sofisticirano opremo, če želijo sprejemati najnaprednejše in največje kontejnerske ladje. Glavni namen članka je opredelitev različnih dejavnikov za doseganje večje produktivnosti kontejnerskih terminalov. Analiza občutljivosti je primerna metoda za ugotavljanje dejavnikov, ki so kritičnega pomena za izboljšanje produktivnosti. V raziskavo je bilo vključenih 28 velikih vzhodnoazijskih kontejnerskih terminalov. Podatki o njihovi učinkovitosti so bili primerjani z analizo ovojnice podatkov (DEA) in z regresijsko analizo (RA). Pri obeh analizah se ena za drugo odstranjujejo vhodne spremenljivke in se ocenjuje korelacija med produktivnostjo in naložbami. Uporabljen je bil model CCR DEA kot neparametrična metoda za merjenje učinkovitosti enot sprejemanja odločitev (DMU). Metoda DEA-CCR se začne z izbiro ustreznih vhodnih in izhodnih spremenljivk. RA po drugi strani omogoča vpogled v to, kako se spreminja tipična vrednost odvisne spremenljivke ob spreminjanju katerekoli od neodvisnih spremenljivk, medtem ko ostale neodvisne spremenljivke ostajajo fiksne. Uporabljena je bila multipla linearna regresijska analiza. Model DEA-CCR skuša doseči maksimum proporcionalnega povečanja izhodnih spremenljivk znotraj območja možnega. Učinkovitost se v poteku dela obravnava, ko je njena vrednost manjša od 1 (neučinkovitost obsega), oz. učinkovitost vhodnih in izhodnih spremenljivk ni primerna ter je treba zmanjšati vhode ali povečati izhode. Obseg operacij v tem primeru ne dosega optimalne vrednosti in ga je treba povečati ali zmanjšati (odvisno od donosa na obseg). Pri regresijski metodi sta bili preskušeni metodi Enter in Backward, med njima pa je bila izbrana boljša. Postopek se ustavi, ko so signifikantni vsi regresijski koeficienti, sicer pa se tisti z najmanjšo stopnjo značilnosti odstranijo iz modela. Postopek je dokončan, ko so signifikantni vsi regresijski koeficienti. Empirični rezultati analiz DEA-CCR in RA kažejo, da vhodne/neodvisne spremenljivke za določanje kritičnih dejavnikov na kontejnerskih terminalih veliko prispevajo k produktivnosti terminalov. Prvi zaključek je, da so najpomembnejši dejavnik portalni prenosniki in terminalski vlačilci. Površina terminala in dolžina priveza sta manj pomembni kot oprema. Iz analize torej sledi, da velika površina terminala in dolg privez ne pomenita nujno večje produktivnosti terminala. Ponuja se zaključek, da lahko povečanje količine opreme za manipulacijo kot so portalni prenosniki in vlačilci izboljša produktivnost kontejnerskega terminala. Kar se tiče dolžine priveza (ta je v negativni korelaciji s kapaciteto priveza), se njegovo podaljšanje odrazi v zmanjšanju produktivnosti. Pomen velikosti terminala je od primera do primera različen. Ugotovljeno je bilo tudi to, da obalna kontejnerska dvigala niso neposredno povezana s kapaciteto, ker privezi z določenim povprečnim številom dvigal dosegajo različne kapacitete. Metodi DEA-CCR in RA sta priljubljeni sredstvi za primerjalne analize in ponujata dva različna pristopa k merjenju in primerjavi učinkovitosti kontejnerskih terminalov. Prednost metode DEA-CCR pred RA je v tem, da omogoča primerjavo z učinkovito namesto s povprečno zmogljivostjo, da kot mejna metoda daje natančnejše ocene relativne učinkovitosti, in da običajno daje natančnejše ciljne rezultate. RA sicer omogoča tudi boljše napovedovanje prihodnje zmogljivosti na zbirni ravni DMU ter ocenjevanje intervalov zaupanja za točkovne ocene. Metoda daje tudi boljše ocene posameznih maksimalnih (minimalnih) ravni, pri čemer so lahko izhodi (vhodi) neodvisni drug od drugega. Raziskava pa ima tudi svoje omejitve, saj več dejavnikov produktivnosti terminalov ni bilo vključenih med spremenljivke. Prihodnje raziskave bi se morale osredotočiti na povečanje števila terminalov in spremenljivk ter na vključitev modela za simulacijo posameznih terminalov. Ključne besede: analiza ovojnice podatkov, regresijska analiza, analiza občutljivosti, produktivnost kontejnerskega terminala, kritični dejavniki, empirični rezultati *Naslov avtorja za dopisovanje: Univerza v Dalianu, Institut za elektronsko trgovino in sodobno logistiko, Dalian, Kitajska, lubo_documents@hotmail.com

SI 103


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 104 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo: 2012-12-27 Prejeto popravljeno: 2013-05-14 Odobreno za objavo: 2013-05-22

Trajnostnost pristanišč: analiza življenjskega cikla opreme za prekladanje tovora z ničelnimi emisijami Vujičić, A. – Zrnić, N. – Jerman, B. Andrija Vujičić1* – Nenad Zrnić2 – Boris Jerman3

1 Donava zavarovanje, Srbija v Beogradu, Fakulteta za strojništvo, Srbija 3 Univerza v Ljubljani, Fakulteta za strojništvo, Slovenija 2 Univerza

Pristanišča in kontejnerski transport so hrbtenica mednarodnih dobavnih verig in imajo v globalnem gospodarstvu pomembno vlogo. Ekonomska sila pristanišč in kontejnerskih terminalov pa žal prinaša tudi velike obremenitve za okolje. Pristanišča so največji vir onesnaževanja obmorskih mest, zato je zavedanje o potrebi zmanjšanja onesnaževanja postalo stvar javnega interesa. Industrija je kot odgovor na strožje predpise o emisijah v pristaniščih ponudila vrsto rešitev za omejevanje emisij in vpliva pristanišč na okolje. Emisije v pristaniščih so posledica različnih operacij, zato se mora vsaka dejavnost z njimi spopasti na svoj način. Ponudniki opreme za prekladanje tovora (CHE) bolj kot kdajkoli prej stavijo na učinkovite tehnologije za pristanišča in kontejnerske terminale. Danes so za skoraj vsak kos opreme CHE na voljo rešitve za zmanjšanje emisij in porabe energije, od alternativnih goriv in hibridnih tehnologij do obetavnega koncepta ničelnih emisij. Namen članka je osvetlitev ozadja koncepta ničelnih emisij. Za raziskavo sta bila izbrana dva najpogostejša stroja na kontejnerskih terminalih: portalni žerjav z gumijastimi kolesi (RTG) in tovornjak za prestavljanje kontejnerjev (UTR), v konvencionalni različici ter v različici brez emisij. Čeprav je prehod z opreme za prekladanje tovora z dizelskimi motorji na električno opremo korak naprej, pa je treba upoštevati tudi nekatere trajnostne vidike pretvorbe energij. Za analizo okoljske učinkovitosti tehnologije ‘ničelnih emisij’ pri CHE je bila uporabljena metodologija ocene življenjskega cikla (LCA). LCA nudi sistematičen pristop za analizo celotnega življenjskega cikla od pridobivanja materialov prek proizvodnje, eksploatacije do odstranjevanja in recikliranja, zato se pogosto imenuje tudi analiza 'od zibelke do groba'. Analiza LCA je uporabna tudi za vrednotenje okoljskega vpliva različnih energijskih virov od vrtine do rezervoarja, ki vključuje črpanje fosilnih goriv, rafiniranje, transport in distribucijo do proizvodnje električne energije. Opravljena analiza LCA je bila za enostavnejše modeliranje razdeljena v tri faze ter uporablja večino podatkov kovencionalne opreme CHE pri vrednotenju CHE z ničelnimi emisijami. Prva faza vključuje vse procese od pridobivanja in izkoriščanja materialov prek proizvodnje delov in sestave modelov do distribucije na luški terminal, običajno pa se imenuje tudi faza ‘od zibelke do vrat’. Druga faza je eksploatacija 'od vrat do groba', zadnja faza pa je odstranjevanje ob koncu življenjskega cikla. Rezultati analize LCA kažejo, da je elektrifikacija CHE primerna in trajnostna rešitev za omejevanje okoljskega vpliva pristanišč. Faza 'od zibelke do vrat' daje manjši prispevek k celotnemu vplivu, kar je običajno za izdelke z dolgo dobo uporabnosti. Faza 'od vrat do groba' električne opreme CHE ima občutno manjši prispevek k globalnemu segrevanju, acidifikaciji in evtrofikaciji, pri čemer obstaja značilna razlika med RTG in UTR. Primerjalna analiza LCA je pokazala, da ima elektrifikacija CHE večji potencial za zmanjšanje emisij in rabe energije pri večji in težji opremi, kot so npr. žerjavi RTG. Energetsko potratne delovne operacije žerjava RTG so primernejše za elektrifikacijo kljub preostalim emisijam skozi izpušno cev pri menjavi blokov. Rezultati kažejo, da so žerjavi RTG zaradi dolge dobe uporabnosti in velikega števila delovnih ur zelo primerni za elektrifikacijo. Električni tovornjaki UTR po drugi strani nimajo emisij skozi izpušno cev, povzročajo pa velik vpliv na okolje zaradi litij-ionskih akumulatorjev. Doba uporabnosti UTR je lahko dvakrat daljša od življenjske dobe akumulatorskih baterij, potreba po menjavanju akumulatorskih baterij pa lahko skupaj z različnimi izvedbami električnega omrežja okrni potencial za zmanjšanje emisij električnih UTR. Zanimanje za metodologijo LCA se močno povečuje, le malo pa je raziskav o možnostih uporabe te metodologije pri CHE, zlasti pri žerjavih RTG. Prispevek tega članka je zato dvojen. Članek podaja informacije o okoljskem vplivu konvencionalnih in električnih izvedb dveh najpogostejših kosov opreme CHE na kontejnerskih terminalih skozi celoten življenjski cikel. Promovira pa tudi uporabo metodologije LCA kot primerjalnega orodja, ki odpira možnosti za prihodnje raziskave na tem področju ter zagotavlja vezni člen med pristopom LCA in strategijami za uvajanje CHE. Ključne besede: kontejnerski terminal, oprema za prekladanje tovora, ničelne emisije, portalni žerjav z gumijastimi kolesi, tovornjak za prestavljanje kontejnerjev, ocena življenjskega cikla SI 104

*Naslov avtorja za dopisovanje: Donava zavarovanje, Blagoja Parovica 19, Beograd, Srbija, a.vujicic@dunavauto.co.rs


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 105 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo: 2012-12-29 Prejeto popravljeno: 2013-03-21 Odobreno za objavo: 2013-04-10

Eksperimentalna in numerična preiskava utrujenostnega zloma nosilne konstrukcije čeljustnega drobilnika

Rusiński, E. – Moczko, P. – Pietrusiak, D. – Przybyłek, G. Eugeniusz Rusiński – Przemysław Moczko* – Damian Pietrusiak – Grzegorz Przybyłek Tehniška univerza v Vroclavu, Poljska

Pri obratovanju drobilne postaje v površinskem kamnolomu so se pojavile velike težave z vzdržljivostjo nosilnega ogrodja. Prvi simptomi so se pojavili že po krajšem času obratovanja novega drobilnega postrojenja. Na nosilni konstrukciji drobilnika se je pojavilo veliko utrujenostnih razpok, ki so se hitro širile, odpovedalo pa je tudi več vijačnih zvez. Zaradi velikih dinamičnih sil, ki se pojavljajo pri obratovanju takšne opreme, je bilo treba ugotoviti vzroke težav in tako preprečiti popolno uničenje drobilne postaje. Pri ugotavljanju vzrokov in odpravljanju težav je bil uporabljen kombiniran numerični in eksperimentalni postopek. Prvi del raziskave pokriva meritve dinamičnih parametrov konstrukcije ter ocenitev dinamičnih obremenitev, ki povzročajo zlome. Dinamične sile na ogrodje drobilnika so stohastičnega značaja, ker so velikost in mehanske lastnosti materiala, ki se dovaja v drobilnik, naključne spremenljivke. Zato je bilo treba eksperimentalno ugotoviti maksimalne sile, ki delujejo na konstrukcijo. Izmerjene sile so bile bistveno večje od sil, uporabljenih pri trdnostnih preračunih v fazi projektiranja. Da bi ocenili vpliv izmerjenih sil in preverili obremenitev konstrukcije v takšnih pogojih, smo ustvarili model nosilne konstrukcije po metodi končnih elementov. Rezultati preračunov utrujanja so potrdili, da so območja konstrukcije z razpokami preobremenjena in da nimajo ustrezne dinamične trdnosti. Omeniti velja, da izračuni 100-odstotno potrjujejo mesta razpok na realni konstrukciji. Meritve, modalna analiza in trdnostni izračuni so razkrili dva glavna razloga za slabo obratovalno stanje nosilne konstrukcije drobilnika: - resonančne vibracije v transverzalni smeri zaradi nezadostne togosti konstrukcije v tej smeri, - šibka zasnova spojev, ki ne nosijo delovnih obremenitev v obeh horizontalnih smereh. Za povečanje togosti sta bila dva od treh prerezov ogrodja ojačena s stranskimi oporniki. Spoji, ki so utrpeli razpoke, so bili zamenjani z novimi spoji, obstojnimi proti utrujanju. Dinamične in utrujenostne lastnosti novih rešitev so bile preverjene s posodobljenim modelom MKE. Spremenjena konstrukcija je bila končno preskušena z napravo za merjenje vibracij. Potrjeno je bilo, da je resonanca odpravljena, vedenje konstrukcije med normalnim obratovanjem pa je primerno. Problemi z vzdržljivostjo so tako odpravljeni, dodatne raziskave in spremembe pa niso več potrebne. Članek predstavlja praktične vidike analize odpovedi, spremembe konstrukcije in uveljavljanja nove rešitve. Metoda je bila spremenjena in prilagojena za realne industrijske probleme, pri katerih se raziskovalci soočajo z drugačnimi izzivi kot pri delu v laboratorijskem okolju. Predmet raziskave je bil visokozmogljiv stroj, ki obratuje v težavnih razmerah, zato je bila realizacija še toliko zahtevnejša. Ključne besede: čeljustni drobilnik, dinamična trdnost, resonanca, numerične simulacije, eksperimentalne metode

*Naslov avtorja za dopisovanje: Tehniška univerza v Vroclavu, Lukasiewicza 7/9, 50-371 Wroclaw, Poljska, przemyslaw.moczko@pwr.wroc.pl

SI 105


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 106 © 2013 Strojniški vestnik. Vse pravice pridržane. Tematska številka

Prejeto v recenzijo: 2010-12-28 Prejeto popravljeno: 2011-03-03 *Odobreno za objavo: 2011-05-03

Časovno razporejanje za obnovo delovanja intermodalnega sistema z enim terminalom in prihodi po Poissonovi porazdelitvi Marković, N. – Schonfeld, P. Nikola Marković – Paul Schonfeld*

Univerza v Marylandu, Oddelek za gradbeništvo in okoljski inženiring, ZDA

Članek obravnava obnovo delovanja intermodalnega transportnega sistema po večji prekinitvi in predstavlja model za optimizacijo časovnega razporejanja vozil v motenem stanju. Predlagani model optimizira obnovo delovanja sistema z enim terminalom z razmeroma kratkimi dovoznimi potmi, kjer je čas krožne poti vozil porazdeljen eksponencialno, prihodi na terminal pa sledijo Poissonovi porazdelitvi. Model, ki je razvit v tem članku, je mogoče uporabiti za transportne sisteme z enim terminalom in vsako kombinacijo načinov prevoza z diskretnimi vozili, če se le časi dostave ne razlikujejo preveč od privzete Poissonove porazdelitve. Namen članka je predstavitev predloga optimizacijskega modela, ki določa čase odhodov za glavne poti tako, da bodo skupni stroški za dobavitelja minimalni, ob upoštevanju različnih omejitev realnih transportnih sistemov. Problem načrtovanja je modeliran kot nelinearen matematični program in je optimiziran z genetskim algoritmom. Genetski algoritem je metahevrističen in ne zagotavlja optimalnih rešitev, z razmeroma kratkim računskim časom pa daje dobre in uporabne rešitve. Predlagani model je bil uporabljen na več primerih, kjer se blago dostavlja s tovornjaki, odvaža pa z letali. Analiza občutljivosti je potrdila pričakovane kompromise pri vrstah stroškov. Model je tudi uspešno optimiziral obnovo delovanja intermodalnega sistema, za kar je porabil le nekaj sekund računskega časa. Več predpostavk, ki so bile upoštevane v tem članku, bi bilo v prihodnje mogoče nekoliko zrahljati, s čimer bi model postal bolj splošen. Trenutni model bi bilo mogoče izboljšati za dobre in robustne rešitve tudi v primeru, da ni izpolnjena katera od treh zahtev za veljavnost Poissonovega procesa. V analizi je bilo razen tega privzeto tudi nespremenljivo število vozil na dovoznih poteh. Prihodnje raziskave bi bile lahko usmerjene v flote vozil spremenljive velikosti na dovoznih poteh ter na nestacionarno intenziteto prihodov. Članek preučuje obnovo delovanja intermodalnega transportnega sistema po prekinitvi. Razvit je bil nov model, ki optimizira časovno razporejanje vozil na glavnih poteh ob predpostavki Poissonove porazdelitve prihodov po dovoznih poteh. Predstavljeni model je bil uporabljen na več primerih, kjer se blago dostavlja s tovornjaki, glavne poti pa so letalske. Model je brez večjih sprememb uporaben tudi za druge kombinacije vrst transporta, pogoj je le, da prihodi ne odstopajo preveč od Poissonove porazdelitve. Ključne besede: časovno razporejanje, prekinitev, intermodalni, Poisson, genetski algoritem, transport

*Odobreno za objavo: 2011-09-05, Soglasje o prenosu pravic: 2013-03-05.

SI 106

*Naslov avtorja za dopisovanje: Univerza v Marylandu, Oddelek za gradbeništvo in okoljski inženiring, College Park, MD 20742, ZDA, pschon@umd.edu


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 107-112 Osebne objave

Doktorske disertacije, znanstvena magistrska dela, diplomske naloge

DOKTORSKE DISERTACIJE Na Fakulteti za strojništvo Univerze v Ljubljani so obranili svojo doktorsko disertacijo: ●    dne 4. julija 2013 Andrej LJUBENKO z naslovom: »Termoekonomsko vrednotenje energijskih produktov v daljinski energetiki« (mentor: prof. dr. Alojz Poredoš);

V preteklosti so sistemi daljinskega ogrevanja predstavljali pomembno možnost za znižanje stroškov in izboljšanje učinkovitosti rabe energije na gosteje poseljenih območjih. Ob zahtevah po izboljšanju trajnosti zadostitve energetskih potreb, zniževanju rabe energije za ogrevanje bivalnih prostorov ter dvigovanju cen energentov se pojavljajo potrebe po preverbi in kritični oceni njihovih pravih termodinamičnih in stroškovnih prednosti. Te so v disertaciji vrednotene po principih eksergoekonomike, za kar je bil razvit splošen računalniški model. Opravljena je analiza vpliva pomembnejših parametrov zasnove in obratovanja distribucijske mreže na eksergoekonomske kriterije vrednotenja dobavljene toplote. Model je uporabljen na obstoječem sistemu daljinskega ogrevanja, kjer je stanje distribucijske mreže eksperimentalno določeno na podlagi prilagojene integralne merilne metode. Obravnavana je tudi proizvodnja hladu v absorpcijskih hladilnikih z daljinsko toploto. Ugotovljeno je bilo, da sistemi daljinskega ogrevanja ob integraciji energetskih sistemov, zadostni gostoti potreb po toploti, ustrezni zasnovi in optimizaciji obratovalnih parametrov glede na trenutne razmere, omogočajo dobro izkoriščenost potencialov virov energije in nizke stroške ter posledično cene energijskih produktov. Kadar ti pogoji niso izpolnjeni se njihove prednosti hitro zmanjšajo;

●    dne 9. julija 2013 Rok FINK z naslovom: »Vpliv mikroklimatskih razmer in kakovosti zraka v mestnem okolju na zdravje in počutje ljudi s srčno-žilnimi boleznimi« (mentor: prof.dr. Sašo Medved, somentor: izr. prof. dr. Ivan Eržen);

Intenzivna urbanizacija okolja s svojimi toplotnimi in tokovnimi značilnostmi vpliva na odziv človeškega organizma, ki je lahko še posebej obremenilen za tvegane skupine ljudi. V toplotno obremenilnem obdobju smo spremljali mikroklimatske pogoje in kakovost zraka v notranjem okolju ter hkrati merili fiziološki odziv pri starostnikih, ki imajo srčno-žilne bolezni. Rezultati kažejo, da na srčni utrip in krvni tlak pomembno vplivata indeks Humideks in vsebnost CO2 v zraku (p < 0,05) kot indikator kakovosti zraka. Tako smo izdelali modele odvisnosti fiziološkega odziva od toplotne obremenitve in kakovosti zraka ter jih uporabili za vrednotenje ukrepov

blaženja vpliva toplotnih uličnih kanjonov. Z uporabo numerične analize smo vrednotili toplotne in tokovne značilnosti ozelenelih površin, drevoredov in prometa. Z metodo združevanja toplotnega odziva urbanega okolja in toplotnega odziva stavbe smo dokazali, da z oblikovanjem in prostorsko razporeditvijo gradnikov lahko v stavbi znižamo temperaturo zraka tudi za 2 °C, oziroma prepolovimo koncentracijo onesnažil glede na najslabše stanje. Z vrednotenjem ukrepov blaženja izpostavljenosti na osnovi tveganja za umrljivost smo ugotovili, da lahko tveganje zmanjšamo za tretjino glede na najslabše stanje;

●    dne 16. julija 2013 Dunja RAVNIKAR z naslovom: »Lasersko oplastenje aluminijevih zlitin s keramičnimi sestavinami« (mentor: prof. dr. Janez Grum);

V doktorskem delu je predstavljeno lasersko oplastenje aluminijeve zlitine EN AW-6082-T651 s prednanosom keramičnih sestavin TiB2-TiC z dodatkom aluminija v prašnati obliki. Optimalne pogoje laserskega oplastenja smo določili iz izmerjenih značilnosti nastale obloge in spoja med oblogo in substratom ter na osnovi rezultatov podali oceno uspešnosti procesa. Prva skupina značilnosti je bila določena s srednjo aritmetično hrapavostjo iz izmerjenih profilov oplastene površine v različnih smereh, merjenjem debelin oplastenega sloja in poroznosti. Druga skupina značilnosti vključuje kakovost oplastenega sloja in spoja med oblogo ter substratom z upoštevanjem nastale mikrostrukture podprte z mikrokemičnimi spremembami. Tretja skupina značilnosti pa podaja podatke o mehanskih lastnostih tankega oplastenega sloja, kot je mikrotrdota preko oplastenega sloja in spoja med oblogo in substratom ter merjenje zaostalih napetosti in upogiba ter obrabne odpornosti. Zaradi omejenega števila preizkusov in velikega števila značilnosti, ki vplivajo na kakovost oplastenega sloja, smo uporabili analizo variance za določevanje signifikantnosti parametrov.

Zaradi toplotnega vpliva med oplastenjem in hitrega ohlajanja obloge ter prehodnega področja so nastale mikrostrukturne spremembe na površini substrata pod oblogo. Spremenjena mikrostruktura tik pod oblogo se odraža v manjši trdoti, kar pripišemo uporabi zlitine v izločevalno utrjenem stanju. Zato smo v drugem delu raziskali vpliv toplotne obdelave oplastenih vzorcev v peči. V tretjem delu pa smo raziskali korozijsko odpornost oplastenih vzorcev in jih primerjali z korozijsko odpornostjo neoplastene aluminijeve zlitine; ●    dne 17. julija 2013 Uroš TRDAN z naslovom: »Lasersko udarno utrjevanje aluminijevih zlitin« (mentor: prof. dr. Janez Grum); SI 107


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 107-112

V raziskavi je podrobno analiziran vpliv različnih parametrov inovativnega postopka utrjevanja z laserskimi udarnimi valovi (LSP) na izločevalno utrjeni aluminijevi zlitini AA6082-T651. V prvi fazi je analizirana integriteta površine z optično in elektronsko SEM/EDS mikroskopijo. 3D topografija zaradi učinka plazme ter udarnih valov je bila analizirana na konfokalnem mikroskopu visoke ostrine ter ovrednotena s statistično metodo odzivnih površin (RSM). Nadalje, analiza mikrostrukture na transmisijskem elektronskem mikroskopu (TEM) je potrdila visoko gostoto dislokacij ter ultra-fina in nano zrna, kar se odraža z induciranimi tlačnimi zaostale napetostmi ter višjo mikrotrdoto v neposrednem podpovršju. Dodatno smo analizirali še obrabno drsno odpornost pri različnih silah obremenitve. V drugi fazi študije smo obravnavali vpliv različnih parametrov LSP obdelave na korozijsko odpornost v 0.6 M NaCl raztopini. Potrdili smo izboljšano korozijsko odpornost po procesu LSP, z večjim pasivnim področjem, izboljšano možnostjo repasivacije ter redukcijo korozijskega toka celo za faktor 30. SEM/EDS analiza je potrdila, manj intenziven kristalografski in sferični jamičasti napad na površini LSP tretiranih vzorcev, zaradi kemično bolj stabilnega oksidnega filma Al2O3, večjih debelin. To smo potrdili s spektroskopijo Augerjevih elektronov (AES) in z foto elektronsko spektroskopijo (XPS). V zadnjem delu, »in-situ« analize akustične emisije (AE) med kratkotrajnimi potenciostatskimi testi je bila potrjena izredna korelacija med aktivnostjo AE ter velikostjo gostote toka. Nadalje, analiza izbruhov AE je potrdila 3 glavne izvore AE in sicer, porušitev filma, širjenje napada ter nastanek vodika, ki se je izkazal kot najbolj emisivni izvor AE;

●    dne 19. julija 2013 Dominik KOBOLD z naslovom: »Določitev vpliva anizotropnih lastnosti na tok materiala pri plastičnem preoblikovanju gnetne magnezijeve zlitine AZ80« (mentor: izr. prof. dr. Tomaž Pepelnjak);

V doktorski nalogi smo obravnavali masivno preoblikovanje gnetne magnezijeve zlitine AZ80, ki ima zaradi osnovne heksagonalne gosto zložene (h. g.  zl.) kristalne rešetke zelo specifične preoblikovalne lastnosti. Analizirali smo vplive različnih procesnih parametrov in anizotropije na tok materiala med preoblikovanjem. Vpliv anizotropije na tok materiala pri preoblikovanju iztisnjene palice magnezijeve zlitine AZ80 smo ovrednotili v odvisnosti od različnih procesnih parametrov, kot so različne izhodiščne temperature, hitrosti in stopnje deformacij ter smeri tlačnega obremenjevanja. Omenjeno področje je bilo do sedaj izredno slabo raziskano. V nadaljevanju smo na primeru preoblikovanja magnezijeve zlitine AZ80 preučili možnost popisa anizotropije za potrebe postavitve numeričnih analiz, ki delujejo po metodi končnih elementov (MKE). Za

SI 108

popis anizotropnih lastnosti v postavljenih numeričnih MKEmodelih smo uporabili splošni Hillov kvadratni zakon plastičnega tečenja ortotropnih materialov (1948). Na osnovi eksperimentalnih študij smo razvili inverzni postopek za določitev materialnih konstant splošnega Hillovega kvadratnega zakona (1948). S postavljenimi numeričnimi MKEmodeli smo pri različnih vhodnih procesnih parametrih simulirali aksialno nakrčevanje enostavnih osnosimetričnih valjastih testnih vzorcev na različne končne višine in v konusno obliko. Rezultate numeričnih MKEanaliz smo primerjali z rezultati eksperimentalnih študij. Ugotovili smo, da je mogoče s pomočjo osnovnega Hillovega kvadratnega zakona učinkovito napovedovati preoblikovanje magnezijeve zlitine AZ80 z osnovno h. g.  zl. kristalno rešetko in da je razviti inverzni postopek primeren za določitev Hillovih ortotropnih koeficientov v materialnem modelu. Rezultati so neposredno uporabni v industrijskem okolju;

●    dne 20. avgusta 2013 Aleš GOSAR z naslovom: »Modeliranje razvoja poškodbe s prostorskimi Prandtlovimi operatorji« (mentor: prof. dr. Marko Nagode, somentor: prof. dr. Matija Fajdiga);

Napovedovanje dobe trajanja je pomemben del procesa razvoja novega izdelka, saj lahko nenadna odpoved katere od komponent povzroči gmotno škodo ali celo terja človeška življenja. Obstaja veliko metod, s katerimi na različne načine in različno dobro izračunamo pričakovano dobo trajanja. V doktorski raziskavi je predstavljen model, ki smo ga zasnovali na energijskih metodah in v katerem je za merilo utrujenostne poškodbe izbrana sprememba sproščene energije. Na podlagi energijske bilance segmenta Prandtlovega operatorja, ki se sicer uporablja za simuliranje napetostnodeformacijskih stanj, smo določili energijski večosni model in nadgradili izhodiščno metodologijo za izračun dobe trajanja. Z numerično simulacijo smo pokazali, da je preračun sproščene energije z razvitim modelom z operatorji boljši od uveljavljene metode z integriranjem histerezne zanke, saj za izračun ne potrebujemo zaključenih ciklov, pretvorbe večosnih napetostnodeformacijskih stanj v ekvivalentna enoosna stanja, vezano in sproščeno energijo pa računamo ločeno in sproti, kar je velika prednost. S primerjavo izračunanih dob trajanja z izmerjenimi smo pokazali, da je razvita metoda primerna.

* Na Fakulteti za strojništvo Univerze v Mariboru sta obranili svojo doktorsko disertacijo: ●    dne 29. julija 2013 Zoran LESTAN z naslovom: »Razvoj inteligentnega sistema za modeliranje nanašanja materiala z uporabo laserja« (mentor: prof. dr. Miran Brezočnik);


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 107-112

Lasersko nanašanje materialov predstavlja sodobno dodajalno tehnologijo, ki ima v primerjavi s preostalimi postopki nanašanja kovinskih materialov številne prednosti. Poleg minimalnega vnosa energije, kvalitetnega spoja in majhnega toplotno vplivnega območja to tehnologijo odlikuje še dobra mehanska trdnost nanesenega materiala, ki je posledica hitrega ohlajanja. Kljub perspektivnosti tehnologije pa je ta še vedno v razvojni fazi. Vpeljujejo se novi materiali ter tehnike za določitev optimalnih procesnih parametrov. V disertaciji je prikazan postopek izdelave empiričnih modelov, s katerimi lahko na podlagi uporabljenih procesnih parametrov stroja napovemo lastnosti nanesenega materiala. Za izdelavo modelov je bilo uporabljeno genetsko programiranje ter regresijska analiza, ki za iskanje regresijskih koeficientov uporablja genetske algoritme. Na podlagi eksperimentalnih podatkov sta bila izdelana modela za napovedovanje volumna in modela za napovedovanje hrapavosti. Verifikacija dobljenih modelov je bila izvedena z uporabo testne množice eksperimentalnih podatkov. Na osnovi dobljenih modelov sistem določi optimalne parametre nanašanja materiala s stališča hitrosti nanašanja, izkoristka materiala in hrapavosti površine. Določitev optimalnih procesnih parametrov stroja je bila izvedena z uporabo nedominiranega sortiranja. Rezultati ponujajo operaterju niz optimalnih nastavitev procesnih parametrov, kar omogoča izdelavo kakovostnih izdelkov;

●    dne 29. julija 2013 Iztok NOČ z naslovom: »Možnosti prilagoditve poškodbenih kriterijev populaciji otrok z uporabo numeričnega anatomskega modela otroka izbrane starosti« (mentor: prof. dr. Jože Flašker);

ATD-ji in PK so ključni pri snovanju rešitev za preprečitev poškodb. Obstoječi PK so definirani na podlagi eksperimentalnih podatkov, ki pripadajo populaciji odraslih ljudi. Zato je primernost obstoječih PK za populacijo otrok vprašljiva. Ker je pridobivanje eksperimentalnih podatkov za populacijo otrok iz etičnih razlogov skoraj neizvedljivo, je smiselno razmišljati o ustrezni modifikaciji podatkov, ki pripadajo populaciji odraslih ljudi. Mejne vrednosti PK so že bile prilagojene populaciji otrok oz. otroškim ATD-jem. A ustreznost modificiranih vrednosti je še vedno vprašljiva, saj so otroški ATD-ji mehanske naprave, ki le grobo opisujejo anatomijo otrok. Zato je bilo raziskovalno delo usmerjeno v pregled možnosti in omejitev modifikacije obstoječih PK za populacijo otrok z uporabo numeričnega anatomskega modela otroka, ki upošteva vse ključne anatomske in antropometrične posebnosti celotnega telesa otroka izbrane starosti. V prvi fazi raziskovalnega dela je bila opravljena antropometrična in anatomska analiza populacije otrok, v kateri so bile določene starostne skupine in statistični predstavniki otrok. V drugi fazi je bil razvit numerični anatomski

model otroka, ki upošteva vse anatomske posebnosti 6 letnega otroka in z antropometričnega vidika ustreza 5 percentilnemu statističnemu predstavniku otrok v starostni skupini 4-7 let. Z namenom doseganja čim višje stopnje biološke ustreznosti modela, je bila izvedena rekonstrukcija vseh ključnih telesnih tkiv in struktur, razviti pa so bili tudi novi principi modeliranja sklepov. V zadnji fazi raziskovalnega dela so bile izvedene simulacije obremenjevanja razvitega modela (NAMO6.5). Rezultati simulacij so pokazali dober odziv in potrdili visoko biološko ustreznost ter polno funkcionalnost modela. S pomočjo modela je bila izvedena ocena možnosti in omejitev modifikacije izbranih PK (HIC, ThCC) populaciji otrok. Ker sta izbrana PK pokazale pomembne pomanjkljivosti,so bile poškodbe ocenjene še na podlagi raztezka in napetosti v telesnem tkivu. Rezultati so pokazali, da je razvoj novih PK na osnovi raztezka in napetosti bolj smiseln kot modifikacija obstoječih PK.

ZNANSTVENA MAGISTRSKA DELA Na Fakulteti za strojništvo Univerze v Ljubljani je z uspehom zagovarjal svoje magistrsko delo: ●    dne 18. julija 2013 Franci VEHAR z naslovom: »Centripetalna črpalna stopnja v večstopenjski centrifugalni črpalki« (mentor: izr. prof. dr. Mihael Sekavčnik, somentor: prof. dr. Dušan Florjančič). * Na Fakulteti za strojništvo Univerze v Mariboru je z uspehom zagovarjala svoje magistrsko delo: ●    dne 12. julija 2013 Valentina JELEN z naslovom: »Primerjava uspešnosti razbarvanja odpadne vode z oksidacijskim in redukcijskim postopkom« (mentorica: prof. dr. Alenka Majcen le Marechal). DIPLOMSKE NALOGE Na Fakulteti za strojništvo Univerze v Ljubljani so pridobili naziv univerzitetni diplomirani inženir strojništva: dne 28. avgusta 2013: Matic HERZOG z naslovom: »Konstruiranje sistema spenjalnih spojev iz aluminija za urbano opremo« (mentor: izr. prof. dr. Jože Tavčar, somentor: prof. dr. Jožef Duhovnik); Anže SELAN z naslovom: »Plinski postroj v sistsemu sočasne proizvodnje toplote in električne energije« (mentor: izr. prof. dr. Mihael Sekavčnik); dne 30. avgusta 2013: Jure BARBIČ z naslovom: »Razvoj naprave za prečni, vzdolžni in krožni pomik paketov« (mentor: prof. dr. Marko Nagode); SI 109


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 107-112

Nejc BAUMAN z naslovom: »Merilna proga za vizualizacijo delovanja ploščnega prenosnika toplote« (mentor: prof. dr. Iztok Golobič); Andrej GANTAR z naslovom: »Aerodinamska analiza pršilnika fitofarmacevtskih sredstev« (mentor: izr. prof. dr. Marko Hočevar, somentor: prof. dr. Branko Širok); Blaž MAROLT z naslovom: »Razvoj sistema za izkoriščanje toplote odpadne vode« (mentor: prof. dr. Iztok Golobič, somentor: izr. prof. dr. Janez Kušar); Jaka ŠPRINGER z naslovom: »Lokalizacija avtonomnega mobilnega robota na osnovi optičnih triangulacijskih in inkrementalnih rotacijskih zaznaval« (mentor: prof. dr. Janez Diaci). * Na Fakulteti za strojništvo Univerze v Mariboru je pridobil naziv univerzitetni diplomirani gospodarski inženir: dne 30. avgusta 2013: David STAROVASNIK z naslovom: »Projekt razvoja samonosilnega stikalnega gumba pečice« (mentor: doc. dr. Iztok Palčič, somentor: prof. dr. Anton Hauc). *** Na Fakulteti za strojništvo Univerze v Ljubljani so pridobili naziv magister inženir strojništva: dne 28. avgusta 2013: Hugo ZUPAN z naslovom: »Simulacija toka materiala v montažnem in strežnem procesu posamične proizvodnje« (mentor: izr. prof. dr. Niko Herakovič); dne 30. avgusta 2013: Zerihun MELLESE MEGEN z naslovom: »Temperaturna odvisnost strižne voljnosti polietilena (POM) različnih molekularnih tež in povezava s stopnjo kristaliničnosti« (mentor: prof. dr. Igor Emri); Aamir SHABBIR z naslovom: »Določitev ekvivalentnosti vpliva tlaka in temperature na strižne in volumetrične lastnosti termoplastičnega poliuretana« (mentor: prof. dr. Igor Emri); Blaž STARC z naslovom: »Dinamska analiza prototipa naprave za štancanje pri visokih hitrostih« (mentor: prof. dr. Miha Boltežar, somentor: Thomas Thümmel); Blaž ŽUGELJ z naslovom: »Vpliv površinskega naboja na tribološke lastnosti keramike SiC in Si3N4 pri mazanju z vodo« (mentor: prof. dr. Mitjan Kalin); Matevž ZUPANČIČ z naslovom: »Izboljšanje učinkovitosti ultrafiltracije rečne vode« (mentor: prof. dr. Janez Diaci, somentor: prof. dr. Iztok Golobič). SI 110

* Na Fakulteti za strojništvo Univerze v Mariboru je pridobil naziv magister inženir strojništva: dne 26. avgusta 2013: Damijan RUPNIK z naslovom: »Numerično in eksperimentalno preverjanje ustreznosti transportne embalaže gospodinjskih aparatov« (mentor: prof. dr. Zoran Ren, somentor: dr. Matej Borovinšek). * Na Fakulteti za strojništvo Univerze v Mariboru je pridobil naziv magister gospodarski inženir: dne 1. julija 2013: Tomaž HODNIK z naslovom: »Preverjanje dobaviteljev in kakovosti materialov v podjetju Arcont« (mentor: prof. dr. Bojan Ačko, somentor: prof. dr. Duško Uršič). *** Na Fakulteti za strojništvo Univerze v Ljubljani so pridobili naziv diplomirani inženir strojništva: dne 28. avgusta 2013: Aleš ČEBAŠEK z naslovom: »Razvoj in vrednotenje sestavljalne mize« (mentor: prof. dr. Marko Nagode); Miran ZUPAN z naslovom: »Lokalno ogrevanje na lesno gorivo« (mentor: prof. dr. Vincenc Butala); dne 29. avgusta 2013: Klemen ŽVAB z naslovom: »Strah pred letenjem« (mentor: prof. dr. Rastko Golouh, somentor: izr. prof. dr. Tadej Kosel); Ervin AVSEC z naslovom: »Ekonomičnost laserskega rezanja pri globoko vlečenih izdelkih« (mentor: doc. dr. Joško Valentinčič, somentor: izr. prof. dr. Tomaž Pepelnjak); dne 30. avgusta 2013: Janez LOČIČNIK z naslovom: »Analiza stanja in določitev možnosti širitve sistema daljinskega ogrevanja« (mentor: prof. dr. Alojz Poredoš); Marko PEGAM z naslovom: »Priprava tehnologije varjenja stebrnih letev energetskega transformatorja« (mentor: prof. dr. Janez Tušek); Rok PLESTENJAK z naslovom: »Robotsko varjenje MAG v različnih legah« (mentor: prof. dr. Janez Tušek). * Na Fakulteti za strojništvo Univerze v Mariboru so pridobili naziv diplomirani inženir strojništva: dne 1. julija 2013:


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 107-112

Matjaž CERAJ z naslovom: »Pregled metod in orodij za zagotavljanje kakovosti« (mentor: prof. dr. Bojan Ačko, somentor: doc. dr. Andrej Godina); dne 29. avgusta 2013: David KRIVEC z naslovom: »Vzdrževanje toplotne postaje« (mentor: doc. dr. Samo Ulaga, somentor: doc. dr. Darko Lovrec); Blaž POSL z naslovom: »Razvoj, načrtovanje proizvodnje in stroškov športnega rekvizita Kozmotrim« (mentor: izr. prof. dr. Borut Buchmeister). *** Na Fakulteti za strojništvo Univerze v Ljubljani so pridobili naziv diplomirani inženir strojništva (UN): dne 1. julija 2013: Sandra FRLIC, Jošt GRUM, Ivana MALNAR; dne 4. julija 2013: Jan PRAPROTNIK; dne 8. julija 2013: Maja JANČAR CAFUTA; dne 9. julija 2013: Aleks VRČEK; dne 10. julija 2013: Erik MURNIK; dne 16. julija 2013: Marko GLAVNIK; dne 25. julija 2013: Anže ALJANČIČ, Jakob KOKALJ; dne 26. julija 2013: Tomaž VRTOVEC; dne 19. avgusta 2013: Matej BREGAR, Bojan BUČAR, Lea SILIČ; dne 20. avgusta 2013: Nejc DEMŠAR, Gregor ERŽEN, Rok MARKEŽIČ, Matevž RESMAN, Aleš VLAJ, Luka VODOPIVEC; dne 21. avgusta 2013: Roki OKORN, Žiga PLUT, Janez SELIŠKAR; dne 23. avgusta 2013: Igor FAČETI, Damir GRGURAŠ; dne 27. avgusta 2013: Peter GERDIN, Domen KRIŽMAN, Andrej MATOH, Primož OGRINEC, Vid RUPERT; dne 28. avgusta 2013: Benjamin FRANČIČ, Jan GRUM, Marko KRŽAN, Simona LEVIČAR, Damjan LOLIĆ, Jure MUROVEC, Gal POTRČ PAJK, Matija POŽAR, Jan PREŠEREN, Matic SLABANJA, Jan SLOKAR, Tomaž SOKLIČ, Domen VELIKONJA, Ambrož VRTOVEC; dne 29. avgusta 2013: Anet ROŽIČ; dne 30. avgusta 2013: Rok ČEPON, Dragan GRGIĆ, Urh ŠTUPAR. * Na Fakulteti za strojništvo Univerze v Mariboru sta pridobila naziv diplomirani inženir strojništva (UN): dne 29. avgusta 2013: Gašper OSWALD z naslovom: »Zanesljivost in obvladovanje stroškov v procesu razvoja izdelka v

podjetju Noži Ravne, d.o.o.« (mentor: doc. dr. Marjan Leber); Miha VOZLIČ z naslovom: »Izdelava rezkal v podjetju EMO orodjarna« (mentor: doc. dr. Mirko Ficko). * Na Fakulteti za strojništvo Univerze v Mariboru so pridobili naziv diplomirani gospodarski inženir (UN): dne 4. julija 2013: Rok KULČAR z naslovom: »Uporaba vrednostne analize pri razvoju večfunkcionalne mize« (mentor: doc. dr. Iztok Palčič, somentor: prof. dr. Anton Hauc); Luka NAPOTNIK z naslovom: »Kriteriji za izbor dobaviteljev v podjetju BSH hišni aparati d.o.o.« (mentor: doc. dr. Iztok Palčič, somentor: izr. prof. dr. Tanja Markovič Hribernik); Rožle PENCA z naslovom: »Optimiranje velikosti ekonomske serije zalog v proizvodnji« (mentor: doc. dr. Iztok Palčič, somentor: prof. dr. Majda Bastič). *** Na Fakulteti za strojništvo Univerze v Ljubljani so pridobili naziv diplomirani inženir strojništva (VS): dne 28. avgusta 2013: Karlo FUNDAK z naslovom: »Razvojno vrednotenje obračala palet« (mentor: prof. dr. Marko Nagode); Primož KAPŠ z naslovom: »Zniževanje emisij prašnih delcev iz kotlovnice na lesno biomaso« (mentor: izr. prof. dr. Andrej Senegačnik); Rok KOŽELJ z naslovom: »Energijska in eksergijska analiza toplotnih tokov pri ogrevanju stavb« (mentor: prof. dr. Vincenc Butala, somentor: doc. dr. Uroš Stritih); Andrej VIRANT z naslovom: »Eksperimentalna določitev učinkovitosti prezračevanja z metodo zmanjševanja« (mentor: prof. dr. Vincenc Butala, somentor: doc. dr. Matjaž Prek); Nejc WEISS z naslovom: »Povrnitev investicije v toplotno izolacijo stare stanovanjske stavbe pri učinkoviti rabi energije« (mentor: prof. dr. Vincenc Butala); dne 29. avgusta 2013: Matic BAJDA z naslovom: »Načrtovanje procesnega sistema čistega prostora za montažo elektromagnetnih stikal« (mentor: izr. prof. dr. Ivan Bajsić); Valerija BOGOVIČ z naslovom: »Razvoj merilnega preizkuševališča za prikaz statistično načrtovanega preizkušanja« (mentor: izr. prof. dr. Ivan Bajsić); SI 111


Strojniški vestnik - Journal of Mechanical Engineering 59(2013)9, SI 107-112

Grega DIVJAK z naslovom: »Laboratorijski sistem za polnjenje toplotne cevi« (mentor: prof. dr. Iztok Golobič); Primož PERME z naslovom: »Ploščati uparjalnik zankaste toplotne cevi« (mentor: prof. dr. Iztok Golobič); Matevž HREŠČAK z naslovom: »Analiza toka materiala in vrednosti« (mentor: prof. dr. Marko Starbek, somentor: izr. prof. dr. Janez Kušar); Tilen CEGLAR z naslovom: »Analiza vpliva hrapavosti na aerodinamične lastnosti modelarskega propelerja« (mentor: doc. dr. Viktor Šajn); Jaka DOBERLET z naslovom: »Pomen vpeljave predpisov o uvedbi tehnologije ADS-B v komercialno in nekomercialno letalstvo« (mentor: doc. dr. Patrick Vlačič, somentor: izr. prof. dr. Tadej Kosel); Ervin KLEMENČIČ z naslovom: »Mikro brezpilotno letalo s stopničastim profilom« (mentor: izr. prof. dr. Tadej Kosel); Aljaž KODRIČ z naslovom: »Določitev vzmetne karakteristike zračne vzmeti osebnih vozil« (mentor: prof. dr. Franc Kosel, somentor: doc. dr. Viktor Šajn); Miha LAMPIČ z naslovom: »Algoritem za popravek smeri letala zaradi vpliva vetra« (mentor: doc. dr. Viktor Šajn, somentor: prof. dr. Jože Rakovec); Željko MARUŠIĆ z naslovom: »Analiza racman letala z lomljenim krilom« (mentor: izr. prof. dr. Tadej Kosel); Žan BROVČ z naslovom: »Točkovno varjenje z gnetenjem aluminijeve zlitine 5754« (mentor: prof. dr. Janez Tušek, somentor: doc. dr. Damjan Klobčar); Vid POCEK z naslovom: »CMT varjenje s stržensko žico v različnih legah« (mentor: prof. dr. Janez Tušek, somentor: doc. dr. Damjan Klobčar); Boštjan RUČMAN z naslovom: »Izdelava testnega mesta za preizkušanje upognjenih nosilcev« (mentor: izr. prof. dr. Tomaž Pepelnjak); Anže SAMEC z naslovom: »Rezanje lesa z laserjem« (mentor: prof. dr. Janez Tušek); Vid VADNU z naslovom: »Varjenje visokolegiranih orodnih jekel« (mentor: prof. dr. Janez Tušek); dne 30. avgusta 2013: Jaka DUGAR z naslovom: »3D karakterizacija obrabe rezalnih orodij pri procesu struženja« (mentor: prof. dr. Janez Kopač, somentor: doc. dr. Franci Pušavec); Rok ERJAVEC z naslovom: »Primerjava in evalvacija uporabe hladilnih emulzij in rezalnega olja pri obdelavi nerjavečega jekla« (mentor: doc. dr. Franci Pušavec, somentor: prof. dr. Janez Kopač); Jožko KOKALJ z naslovom: »Nadzor in vzdrževanje hladilno-mazalnih sredstev pri serijski obdelavi izdelkov iz sive litine« (mentor: doc. dr. Franci Pušavec, somentor: prof. dr. Janez Kopač); SI 112

Klemen KOKELJ z naslovom: »Primerjava nizkocenovnih postopkov rezanja ploščatih tesnil« (mentor: doc. dr. Davorin Kramar, somentor: prof. dr. Janez Kopač); Mitja LEKŠE z naslovom: »Mehanska tehnologija industrijske obnove gredi kondenzatne črpalke« (mentor: doc. dr. Peter Krajnik, somentor: prof. dr. Janez Kopač); Urban PETRIN z naslovom: »Nadgradnja sistema za mazanje/hlajenje odrezovalnega procesa na princip podhlajenega minimalnega mazanja (MQL - minimum quantity lubrication)« (mentor: doc. dr. Franci Pušavec, somentor: prof. dr. Janez Kopač); Tomaž HRIBAR z naslovom: »Vpliv različnih plinov na obnašanje vodnega hidravličnega akumulatorja« (mentor: doc. dr. Andrej Bombač, somentor: doc. dr. Franc Majdič); Martin KOROŠEC z naslovom: »Tovorno hidravlično dvigalo« (mentor: prof. dr. Mitjan Kalin, somentor: doc. dr. Franc Majdič); Klemen MAJCEN z naslovom: »Razvoj tlačne tehtnice za zvezno delujoči hidravlični potni ventil« (mentor: prof. dr. Mitjan Kalin, somentor: doc. dr. Franc Majdič); Rok SERŠEN z naslovom: »Izboljšanje učinkovitosti energetskega sistema v podjetju« (mentor: izr. prof. dr. Andrej Kitanovski, somentor: prof. dr. Alojz Poredoš); Matevž CESTNIK z naslovom: »Analiza vzgornikov in vzdolnikov v atmosferi« (mentor: prof. dr. Jože Rakovec, somentor: izr. prof. dr. Tadej Kosel); Kristjan MARTIĆ z naslovom: »Modeliranje naprave za merjenje dinamične volumetrične voljnosti polimernih materialov« (mentor: prof. dr. Igor Emri); Aljoša NOVAK z naslovom: »Hidravlični sistemi z zaznavalom obremenitve« (mentor: prof. dr. Mitjan Kalin, somentor: doc. dr. Franc Majdič); Jakob PINTAR z naslovom: »Določitev karakteristik konvencionalnega potnega hidravličnega ventila« (mentor: prof. dr. Mitjan Kalin, somentor: doc. dr. Franc Majdič). * Na Fakulteti za strojništvo Univerze v Mariboru sta pridobila naziv diplomirani inženir strojništva (VS): dne 29. avgusta 2013: Anže BLAŽIČ z naslovom: »Vzdrževanje sistemov za lasersko kaljenje« (mentor: izr. prof. dr. Igor Drstvenšek, somentor: dr. Matjaž Milfelner); Simon HAJNC z naslovom: »Talno, konvektorsko in stensko ogrevanje s toplotno črpalko in kotlom na biomaso« (mentor: izr. prof. dr. Jure Marn).


Strojniški vestnik – Journal of Mechanical Engineering (SV-JME) Aim and Scope The international journal publishes original and (mini)review articles covering the concepts of materials science, mechanics, kinematics, thermodynamics, energy and environment, mechatronics and robotics, fluid mechanics, tribology, cybernetics, industrial engineering and structural analysis. The journal follows new trends and progress proven practice in the mechanical engineering and also in the closely related sciences as are electrical, civil and process engineering, medicine, microbiology, ecology, agriculture, transport systems, aviation, and others, thus creating a unique forum for interdisciplinary or multidisciplinary dialogue. The international conferences selected papers are welcome for publishing as a special issue of SV-JME with invited co-editor(s). Editor in Chief Vincenc Butala University of Ljubljana Faculty of Mechanical Engineering, Slovenia Technical Editor Pika Škraba University of Ljubljana Faculty of Mechanical Engineering, Slovenia Editorial Office University of Ljubljana (UL) Faculty of Mechanical Engineering SV-JME Aškerčeva 6, SI-1000 Ljubljana, Slovenia Phone: 386-(0)1-4771 137 Fax: 386-(0)1-2518 567 E-mail: info@sv-jme.eu, http://www.sv-jme.eu Print DZS, printed in 450 copies Founders and Publishers University of Ljubljana (UL) Faculty of Mechanical Engineering, Slovenia University of Maribor (UM) Faculty of Mechanical Engineering, Slovenia Association of Mechanical Engineers of Slovenia Chamber of Commerce and Industry of Slovenia Metal Processing Industry Association Cover: The Luka Koper - Port of Koper is a multipurpose seaport in Slovenia, with core activities focused on handling and warehousing of a variety of goods. Luka Koper operates the largest container terminal in the Adriatic and is a major automotive hub in the Mediterranean, handling almost half a million cars annually. With its excellent geographical position, modern infrastructure and reliable hinterland connections Luka Koper is becoming the leading port operator serving the countries of Central and Eastern Europe. Image Courtesy: Jaka Jeraša, Luka Koper

International Editorial Board Koshi Adachi, Graduate School of Engineering,Tohoku University, Japan Bikramjit Basu, Indian Institute of Technology, Kanpur, India Anton Bergant, Litostroj Power, Slovenia Franci Čuš, UM, Faculty of Mech. Engineering, Slovenia Narendra B. Dahotre, University of Tennessee, Knoxville, USA Matija Fajdiga, UL, Faculty of Mech. Engineering, Slovenia Imre Felde, Obuda University, Faculty of Informatics, Hungary Jože Flašker, UM, Faculty of Mech. Engineering, Slovenia Bernard Franković, Faculty of Engineering Rijeka, Croatia Janez Grum, UL, Faculty of Mech. Engineering, Slovenia Imre Horvath, Delft University of Technology, Netherlands Julius Kaplunov, Brunel University, West London, UK Milan Kljajin, J.J. Strossmayer University of Osijek, Croatia Janez Kopač, UL, Faculty of Mech. Engineering, Slovenia Franc Kosel, UL, Faculty of Mech. Engineering, Slovenia Thomas Lübben, University of Bremen, Germany Janez Možina, UL, Faculty of Mech. Engineering, Slovenia Miroslav Plančak, University of Novi Sad, Serbia Brian Prasad, California Institute of Technology, Pasadena, USA Bernd Sauer, University of Kaiserlautern, Germany Brane Širok, UL, Faculty of Mech. Engineering, Slovenia Leopold Škerget, UM, Faculty of Mech. Engineering, Slovenia George E. Totten, Portland State University, USA Nikos C. Tsourveloudis, Technical University of Crete, Greece Toma Udiljak, University of Zagreb, Croatia Arkady Voloshin, Lehigh University, Bethlehem, USA President of Publishing Council Jože Duhovnik UL, Faculty of Mechanical Engineering, Slovenia General information Strojniški vestnik – Journal of Mechanical Engineering is published in 11 issues per year (July and August is a double issue). Institutional prices include print & online access: institutional subscription price and foreign subscription €100,00 (the price of a single issue is €10,00); general public subscription and student subscription €50,00 (the price of a single issue is €5,00). Prices are exclusive of tax. Delivery is included in the price. The recipient is responsible for paying any import duties or taxes. Legal title passes to the customer on dispatch by our distributor. Single issues from current and recent volumes are available at the current single-issue price. To order the journal, please complete the form on our website. For submissions, subscriptions and all other information please visit: http://en.sv-jme.eu/. You can advertise on the inner and outer side of the back cover of the magazine. The authors of the published papers are invited to send photos or pictures with short explanation for cover content. We would like to thank the reviewers who have taken part in the peerreview process.

ISSN 0039-2480 © 2013 Strojniški vestnik - Journal of Mechanical Engineering. All rights reserved. SV-JME is indexed / abstracted in: SCI-Expanded, Compendex, Inspec, ProQuest-CSA, SCOPUS, TEMA. The list of the remaining bases, in which SV-JME is indexed, is available on the website.

The journal is subsidized by Slovenian Book Agency. Strojniški vestnik - Journal of Mechanical Engineering is also available on http://www.sv-jme.eu, where you access also to papers’ supplements, such as simulations, etc.

Instructions for Authors All manuscripts must be in English. Pages should be numbered sequentially. The maximum length of contributions is 10 pages. Longer contributions will only be accepted if authors provide justification in a cover letter. Short manuscripts should be less than 4 pages. For full instructions see the Authors Guideline section on the journal’s website: http://en.sv-jme.eu/. Please note that file size limit at the journal’s website is 8Mb. Announcement: The authors are kindly invited to submitt the paper through our web site: http://ojs.sv-jme.eu. Please note that file size limit at the journal’s website is 8Mb. The Author is also able to accompany the paper with Supplementary Files in the form of Cover Letter, data sets, research instruments, source texts, etc. The Author is able to track the submission through the editorial process - as well as participate in the copyediting and proofreading of submissions accepted for publication - by logging in, and using the username and password provided. Please provide a cover letter stating the following information about the submitted paper: 1. Paper title, list of authors and affiliations. 2. The type of your paper: original scientific paper (1.01), review scientific paper (1.02) or short scientific paper (1.03). 3. A declaration that your paper is unpublished work, not considered elsewhere for publication. 4. State the value of the paper or its practical, theoretical and scientific implications. What is new in the paper with respect to the state-of-the-art in the published papers? 5. We kindly ask you to suggest at least two reviewers for your paper and give us their names and contact information (email). Every manuscript submitted to the SV-JME undergoes the course of the peer-review process. THE FORMAT OF THE MANUSCRIPT The manuscript should be written in the following format: - A Title, which adequately describes the content of the manuscript. - An Abstract should not exceed 250 words. The Abstract should state the principal objectives and the scope of the investigation, as well as the methodology employed. It should summarize the results and state the principal conclusions. - 6 significant key words should follow the abstract to aid indexing. - An Introduction, which should provide a review of recent literature and sufficient background information to allow the results of the article to be understood and evaluated. - A Theory or experimental methods used. - An Experimental section, which should provide details of the experimental set-up and the methods used for obtaining the results. - A Results section, which should clearly and concisely present the data using figures and tables where appropriate. - A Discussion section, which should describe the relationships and generalizations shown by the results and discuss the significance of the results making comparisons with previously published work. (It may be appropriate to combine the Results and Discussion sections into a single section to improve the clarity). - Conclusions, which should present one or more conclusions that have been drawn from the results and subsequent discussion and do not duplicate the Abstract. - References, which must be cited consecutively in the text using square brackets [1] and collected together in a reference list at the end of the manuscript. Units - standard SI symbols and abbreviations should be used. Symbols for physical quantities in the text should be written in italics (e.g. v, T, n, etc.). Symbols for units that consist of letters should be in plain text (e.g. ms-1, K, min, mm, etc.) Abbreviations should be spelt out in full on first appearance, e.g., variable time geometry (VTG). Meaning of symbols and units belonging to symbols should be explained in each case or quoted in a special table at the end of the manuscript before References. Figures must be cited in a consecutive numerical order in the text and referred to in both the text and the caption as Fig. 1, Fig. 2, etc. Figures should be prepared without borders and on white grounding and should be sent separately in their original formats. Pictures may be saved in resolution good enough for printing in any common format, e.g. BMP, GIF or JPG. However, graphs and line drawings should be prepared as vector images, e.g. CDR, AI. When labeling axes, physical quantities, e.g. t, v, m, etc. should be used whenever possible to minimize the need to label the axes in two languages. Multi-curve graphs should have individual curves marked with a symbol. The meaning of the symbol should be explained in the figure caption. Tables should carry separate titles and must be numbered in consecutive numerical order in the text and referred to in both the text and the caption as

Table 1, Table 2, etc. In addition to the physical quantity, e.g. t (in italics), units (normal text), should be added in square brackets. The tables should each have a heading. Tables should not duplicate data found elsewhere in the manuscript. Acknowledgement of collaboration or preparation assistance may be included before References. Please note the source of funding for the research. REFERENCES A reference list must be included using the following information as a guide. Only cited text references are included. Each reference is referred to in the text by a number enclosed in a square bracket (i.e., [3] or [2] to [6] for more references). No reference to the author is necessary. References must be numbered and ordered according to where they are first mentioned in the paper, not alphabetically. All references must be complete and accurate. All non-English or. non-German titles must be translated into English with the added note (in language) at the end of reference. Examples follow. Journal Papers: Surname 1, Initials, Surname 2, Initials (year). Title. Journal, volume, number, pages, DOI code. [1] Hackenschmidt, R., Alber-Laukant, B., Rieg, F. (2010). Simulating nonlinear materials under centrifugal forces by using intelligent crosslinked simulations. Strojniški vestnik - Journal of Mechanical Engineering, vol. 57, no. 7-8, p. 531-538, DOI:10.5545/sv-jme.2011.013. Journal titles should not be abbreviated. Note that journal title is set in italics. Please add DOI code when available and link it to the web site. Books: Surname 1, Initials, Surname 2, Initials (year). Title. Publisher, place of publication. [2] Groover, M.P. (2007). Fundamentals of Modern Manufacturing. John Wiley & Sons, Hoboken. Note that the title of the book is italicized. Chapters in Books: Surname 1, Initials, Surname 2, Initials (year). Chapter title. Editor(s) of book, book title. Publisher, place of publication, pages. [3] Carbone, G., Ceccarelli, M. (2005). Legged robotic systems. Kordić, V., Lazinica, A., Merdan, M. (Eds.), Cutting Edge Robotics. Pro literatur Verlag, Mammendorf, p. 553-576. Proceedings Papers: Surname 1, Initials, Surname 2, Initials (year). Paper title. Proceedings title, pages. [4] Štefanić, N., Martinčević-Mikić, S., Tošanović, N. (2009). Applied Lean System in Process Industry. MOTSP 2009 Conference Proceedings, p. 422-427. Standards: Standard-Code (year). Title. Organisation. Place. [5] ISO/DIS 16000-6.2:2002. Indoor Air – Part 6: Determination of Volatile Organic Compounds in Indoor and Chamber Air by Active Sampling on TENAX TA Sorbent, Thermal Desorption and Gas Chromatography using MSD/FID. International Organization for Standardization. Geneva. www pages: Surname, Initials or Company name. Title, from http://address, date of access. [6] Rockwell Automation. Arena, from http://www.arenasimulation.com, accessed on 2009-09-07. EXTENDED ABSTRACT By the time the paper is accepted for publishing, the authors are requested to send the extended abstract (approx. one A4 page or 3.500 to 4.000 characters). The instructions for writing the extended abstract are published on the web page http://www.sv-jme.eu/ information-for-authors/. COPYRIGHT Authors submitting a manuscript do so on the understanding that the work has not been published before, is not being considered for publication elsewhere and has been read and approved by all authors. The submission of the manuscript by the authors means that the authors automatically agree to transfer copyright to SV-JME and when the manuscript is accepted for publication. All accepted manuscripts must be accompanied by a Copyright Transfer Agreement, which should be sent to the editor. The work should be original by the authors and not be published elsewhere in any language without the written consent of the publisher. The proof will be sent to the author showing the final layout of the article. Proof correction must be minimal and fast. Thus it is essential that manuscripts are accurate when submitted. Authors can track the status of their accepted articles on http://en.svjme.eu/. PUBLICATION FEE For all articles authors will be asked to pay a publication fee prior to the article appearing in the journal. However, this fee only needs to be paid after the article has been accepted for publishing. The fee is 300.00 EUR (for articles with maximum of 10 pages), 20.00 EUR for each addition page. Additional costs for a color page is 90.00 EUR.


http://www.sv-jme.eu

59 (2013) 9

Strojniški vestnik Journal of Mechanical Engineering

Since 1955

Papers

499

Saut Gurning, Stephen Cahoon, Branislav Dragovic, Hong-Oanh Nguyen: Modelling of Multi-Mitigation Strategies for Maritime Disruptions in the Wheat Supply Chain

Dimitrios Lyridis, Panayotis Zacharioudakis, Stylianos Iordanis, Sophia Daleziou: Freight-Forward Agreement Time series Modelling Based on Artificial Neural Network Models

517

Francesco Longo, Aida Huerta, Letizia Nicoletti: Performance Analysis of a Southern Mediterranean Seaport via Discrete-Event Simulation

526

Davorin Kofjač, Maja Škurić, Branislav Dragović, Andrej Škraba: Traffic Modelling and Performance Evaluation in the Kotor Cruise Port

536

Bo Lu, Nam Kyu Park: Sensitivity Analysis for Identifying the Critical Productivity Factors of Container Terminals

547

Andrija Vujičić, Nenad Zrnić, Boris Jerman: Ports Sustainability: A Life Cycle Assessment of Zero Emission Cargo Handling Equipment

556

Eugeniusz Rusiński, Przemysław Moczko, Damian Pietrusiak, Grzegorz Przybyłek: Experimental and Numerical Studies of Jaw Crusher Supporting Structure Fatigue Failure

564

Nikola Marković, Paul Schonfeld: Scheduling for a Single-Terminal Intermodal System Recovery with Poisson Arrivals

511

Journal of Mechanical Engineering - Strojniški vestnik

Contents

9 year 2013 volume 59 no.

Journal of Mechanical Engineering 2013 9  

The Strojniški vestnik – Journal of Mechanical Engineering publishes theoretical and practice oriented papaers, dealing with problems of mod...

Read more
Read more
Similar to
Popular now
Just for you