IJITCE Aug 2011

UK: Managing Editor International Journal of Innovative Technology and Creative Engineering 1a park lane, Cranford London TW59WA UK E-Mail: editor@ijitce.co.uk Phone: +44-773-043-0249 USA: Editor International Journal of Innovative Technology and Creative Engineering Dr. Arumugam Department of Chemistry University of Georgia GA-30602, USA. Phone: 001-706-206-0812 Fax:001-706-542-2626 India: Editor International Journal of Innovative Technology & Creative Engineering Dr. Arthanariee. A. M Finance Tracking Center India 261 Mel quarters Labor colony, Guindy, Chennai -600032.

Mobile: 91-7598208700

www.ijitce.co.uk

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

IJITCE PUBLICATION

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING Vol.1 No.8 August 2011

www.ijitce.co.uk

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

From Editor's Desk Dear Researcher, Greetings!

The researchers in this issue put their minds together for crutch design for human walk, Bibliometrics impact factor, Milling algorithm and Model for congestion control.

Let us review world research focus for this month. NASA's Aeronautics Research Mission Directorate (ARMD) works to solve the challenges that still exist in our nation's air transportation system: air traffic congestion, safety and environmental impacts. Solutions to these problems require innovative technical concepts, and dedicated research and development. NASA's ARMD pursues the development of new flight operation concepts, and new tools and technologies that can transition smoothly to industry to become products.

Newly invented solution turns tissue transparent without distorting its shape. Developed by Atsushi Miyawaki of the RIKEN Brain Science Institute in Japan and colleagues, Scale could help researchers peer within tissues without destructive incisions.

Surprisingly, some of the worst conditions are in inland areas like Vermont and upstate New York, far from the coast and the worst of Irene's winds. Flooded roads left at least 13 towns in Vermont completely cut off from civilisation, and without power or running water. Still further research can save lives by inventing new tools.

The coolest stars in the galaxy have finally come out of hiding. Astronomers using data from NASA's Widefield Infrared Survey Explorer (WISE) have found six chilly almost-stars called Y dwarfs, which had been hunted unsuccessfully for more than a decade.

The world's largest solar-powered vessel, arrived in Hong Kong as it continued its quest to circumnavigate the globe powered solely by the sun. The massive 95-tonne yacht, named T没ranor PlanetSolar, has 537 square meters of solar panels on deck.

It has been an absolute pleasure to present you articles that you wish to read. We look forward to many more new technology-related research articles from you and your friends. We are anxiously awaiting the rich and thorough research papers that have been prepared by our authors for the next issue.

Thanks, Editorial Team IJITCE

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

Editorial Members Dr. Chee Kyun Ng Ph.D Department of Computer and Communication Systems, Faculty of Engineering, Universiti Putra Malaysia,UPM Serdang, 43400 Selangor,Malaysia.

Dr. Simon SEE Ph.D Chief Technologist and Technical Director at Oracle Corporation, Associate Professor (Adjunct) at Nanyang Technological University Professor (Adjunct) at Shangai Jiaotong University, 27 West Coast Rise #08-12,Singapore 127470 Dr. sc.agr. Horst Juergen SCHWARTZ Ph.D, Humboldt-University of Berlin, Faculty of Agriculture and Horticulture, Asternplatz 2a, D-12203 Berlin, Germany Dr. Marco L. Bianchini Ph.D Italian National Research Council; IBAF-CNR, Via Salaria km 29.300, 00015 Monterotondo Scalo (RM), Italy Dr. Nijad Kabbara Ph.D Marine Research Centre / Remote Sensing Centre/ National Council for Scientific Research, P. O. Box: 189 Jounieh, Lebanon Dr. Aaron Solomon Ph.D Department of Computer Science, National Chi Nan University, No. 303, University Road, Puli Town, Nantou County 54561, Taiwan Dr. Arthanariee. A. M M.Sc.,M.Phil.,M.S.,Ph.D Director - Bharathidasan School of Computer Applications, Ellispettai, Erode, Tamil Nadu,India Dr. Takaharu KAMEOKA, Ph.D Professor, Laboratory of Food, Environmental & Cultural Informatics Division of Sustainable Resource Sciences, Graduate School of Bioresources, Mie University, 1577 Kurimamachiya-cho, Tsu, Mie, 514-8507, Japan Mr. M. Sivakumar M.C.A.,ITIL.,PRINCE2.,ISTQB.,OCP.,ICP Project Manager - Software, Applied Materials, 1a park lane, cranford, UK Dr. Bulent Acma Ph.D Anadolu University, Department of Economics, Unit of Southeastern Anatolia Project(GAP), 26470 Eskisehir, TURKEY Dr. Selvanathan Arumugam Ph.D Research Scientist, Department of Chemistry, University of Georgia, GA-30602, USA.

Review Board Members Dr. T. Christopher, Ph.D., Assistant Professor & Head, Department of Computer Science, Government Arts College(Autonomous), Udumalpet, India.

Dr. T. DEVI Ph.D. Engg. (Warwick, UK), Head, Department of Computer Applications, Bharathiar University, Coimbatore-641 046, India.

Dr. Giuseppe Baldacchini ENEA - Frascati Research Center, Via Enrico Fermi 45 - P.O. Box 65,00044 Frascati, Roma, ITALY. Dr. Renato J. orsato Professor at FGV-EAESP,Getulio Vargas Foundation, S찾o Paulo Business School, Rua Itapeva, 474 (8째 and ar),01332-000, S찾o Paulo (SP), Brazil Visiting Scholar at INSEAD,INSEAD Social Innovation Centre, Boulevard de Constance,77305 Fontainebleau - France Y. Benal Yurtlu Assist. Prof. Ondokuz Mayis University Dr. Paul Koltun Senior Research ScientistLCA and Industrial Ecology Group,Metallic & Ceramic Materials, CSIRO Process Science & Engineering Private Bag 33, Clayton South MDC 3169,Gate 5 Normanby Rd., Clayton Vic. 3168 Dr.Sumeer Gul Assistant Professor, Department of Library and Information Science, University of Kashmir, India

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011 Chutima Boonthum-Denecke, Ph.D Department of Computer Science, Science & Technology Bldg., Rm 120,Hampton University, Hampton, VA 23688 Dr. Renato J. Orsato Professor at FGV-EAESP,Getulio Vargas Foundation, São Paulo Business SchoolRua Itapeva, 474 (8° andar) ,01332-000, São Paulo (SP), Brazil Lucy M. Brown, Ph.D. Texas State University,601 University Drive, School of Journalism and Mass Communication,OM330B,San Marcos, TX 78666 Javad Robati Crop Production Departement,University of Maragheh,Golshahr,Maragheh,Iran Vinesh Sukumar (PhD, MBA) Product Engineering Segment Manager, Imaging Products, Aptina Imaging Inc. doc. Ing. Rostislav Choteborský, Ph.D. Katedra materiálu a strojírenské technologie Technická fakulta,Ceská zemedelská univerzita v Praze,Kamýcká 129, Praha 6, 165 21 Dr. Binod Kumar M.sc,M.C.A.,M.Phil.,ph.d, HOD & Associate Professor, Lakshmi Narayan College of Tech.(LNCT), Kolua, Bhopal (MP) , India.

Dr. Paul Koltun Senior Research ScientistLCA and Industrial Ecology Group,Metallic & Ceramic Materials,CSIRO Process Science & Engineering Private Bag 33, Clayton South MDC 3169,Gate 5 Normanby Rd., Clayton Vic. 3168 DR.Chutima Boonthum-Denecke, Ph.D Department of Computer Science,Science & Technology Bldg.,Hampton University,Hampton, VA 23688 Mr. Abhishek Taneja B.sc(Electronics),M.B.E,M.C.A.,M.Phil., Assistant Professor in the Department of Computer Science & Applications, at Dronacharya Institute of Management and Technology, Kurukshetra. (India). doc. Ing. Rostislav Chotěborský,ph.d, Katedra materiálu a strojírenské technologie, Technická fakulta,Česká zemědělská univerzita v Praze,Kamýcká 129, Praha 6, 165 21 Dr. Amala VijayaSelvi Rajan, B.sc,Ph.d, Faculty – Information Technology Dubai Women’s College – Higher Colleges of Technology,P.O. Box – 16062, Dubai, UAE Naik Nitin Ashokrao B.sc,M.Sc Lecturer in Yeshwant Mahavidyalaya Nanded University Dr.A.Kathirvell, B.E, M.E, Ph.D,MISTE, MIACSIT, MENGG Professor - Department of Computer Science and Engineering,Tagore Engineering College, Chennai Dr. H. S. Fadewar B.sc,M.sc,M.Phil.,ph.d,PGDBM,B.Ed. Associate Professor - Sinhgad Institute of Management & Computer Application, Mumbai-Banglore Westernly Express Way Narhe, Pune - 41 Dr. David Batten Leader, Algal Pre-Feasibility Study,Transport Technologies and Sustainable Fuels,CSIRO Energy Transformed Flagship Private Bag 1,Aspendale, Vic. 3195,AUSTRALIA Dr R C Panda (MTech & PhD(IITM);Ex-Faculty (Curtin Univ Tech, Perth, Australia))Scientist CLRI (CSIR), Adyar, Chennai - 600 020,India Miss Jing He PH.D. Candidate of Georgia State University,1450 Willow Lake Dr. NE,Atlanta, GA, 30329 Jeremiah Neubert Assistant Professor,Mechanical Engineering,University of North Dakota Hui Shen Mechanical Engineering Dept,Ohio Northern Univ.

Xiangfa Wu, Ph.D. Assistant Professor / Mechanical Engineering,NORTH DAKOTA STATE UNIVERSITY Seraphin Chally Abou Professor,Mechanical & Industrial Engineering Depart,MEHS Program 235 Voss-Kovach Hall,1305 Ordean Court,Duluth, Minnesota 55812-3042

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011 Qiang Cheng, Ph.D. Assistant Professor, Computer Science Department Southern Illinois University Carbondale Faner Hall, Room 2140-Mail Code 45111000 Faner Drive, Carbondale, IL 62901 Dr. Carlos Barrios, PhD Assistant Professor of Architecture,School of Architecture and Planning, The Catholic University of America Y. Benal Yurtlu Assist. Prof. Ondokuz Mayis University Lucy M. Brown, Ph.D. Texas State University,601 University Drive, School of Journalism and Mass Communication,OM330B,San Marcos, TX 78666 Dr. Paul Koltun Senior Research ScientistLCA and Industrial Ecology Group, Metallic & Ceramic Materials CSIRO Process Science & Engineering Dr.Sumeer Gul Assistant Professor, Department of Library and Information Science, University of Kashmir,India Chutima Boonthum-Denecke, Ph.D Department of Computer Science, Science & Technology Bldg., Rm 120,Hampton University, Hampton, VA 23688 Dr. Renato J. Orsato Professor at FGV-EAESP, Getulio Vargas Foundation, S찾o Paulo Business School, Rua Itapeva, 474 (8째 and ar)01332-000, S찾o Paulo (SP), Brazil Dr. Wael M. G. Ibrahim Department Head-Electronics Engineering Technology Dept. School of Engineering Technology ECPI College of Technology 5501 Greenwich Road - Suite 100,Virginia Beach, VA 23462 Dr. Messaoud Jake Bahoura Associate Professor-Engineering Department and Center for Materials Research Norfolk State University,700 Park avenue, Norfolk, VA 23504 Dr. V. P. Eswaramurthy M.C.A., M.Phil., Ph.D., Assistant Professor of Computer Science, Government Arts College(Autonomous), Salem-636 007, India.

Dr. P. Kamakkannan,M.C.A., Ph.D ., Assistant Professor of Computer Science, Government Arts College(Autonomous), Salem-636 007, India.

Dr. V. Karthikeyani Ph.D., Assistant Professor of Computer Science, Government Arts College(Autonomous), Salem-636 008, India.

Dr. K. Thangadurai Ph.D., Assistant Professor, Department of Computer Science, Government Arts College ( Autonomous ), Karur - 639 005,India.

Dr. N. Maheswari Ph.D., Assistant Professor, Department of MCA, Faculty of Engineering and Technology, SRM University, Kattangulathur, Kanchipiram Dt - 603 203, India.

Mr. Md. Musfique Anwar B.Sc(Engg.)

Lecturer, Computer Science & Engineering Department, Jahangirnagar University, Savar, Dhaka, Bangladesh.

Mrs. Smitha Ramachandran M.Sc(CS)., SAP Analyst, Akzonobel, Slough, United Kingdom.

Dr. V. Vallimayil Ph.D., Director, Department of MCA, Vivekanandha Business School For Women, Elayampalayam, Tiruchengode - 637 205, India.

Mr. M. Moorthi M.C.A., M.Phil., Assistant Professor, Department of computer Applications, Kongu Arts and Science College, India Prema Selvaraj Bsc,M.C.A,M.Phil Assistant Professor,Department of Computer Science,KSR College of Arts and Science, Tiruchengode Mr. G. Rajendran M.C.A., M.Phil., N.E.T., PGDBM., PGDBF., Assistant Professor, Department of Computer Science, Government Arts College, Salem, India.

Dr. Pradeep H Pendse B.E.,M.M.S.,Ph.d Dean - IT,Welingkar Institute of Management Development and Research, Mumbai, India

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011 Muhammad Javed Centre for Next Generation Localisation, School of Computing, Dublin City University, Dublin 9, Ireland Dr. G. GOBI Assistant Professor-Department of Physics,Government Arts College,Salem - 636 007 Dr.S.Senthilkumar Post Doctoral Research Fellow, (Mathematics and Computer Science & Applications),Universiti Sains Malaysia,School of Mathematical Sciences, Pulau Pinang-11800,[PENANG],MALAYSIA. Manoj Sharma Associate Professor Deptt. of ECE, Prannath Parnami Institute of Management & Technology, Hissar, Haryana, India RAMKUMAR JAGANATHAN Asst-Professor,Dept of Computer Science, V.L.B Janakiammal college of Arts & Science, Coimbatore,Tamilnadu, India Dr. S. B. Warkad Assoc. Professor, Priyadarshini College of Engineering, Nagpur, Maharashtra State, India Dr. Saurabh Pal Associate Professor, UNS Institute of Engg. & Tech., VBS Purvanchal University, Jaunpur, India Manimala Assistant Professor, Department of Applied Electronics and Instrumentation, St Joseph’s College of Engineering & Technology, Choondacherry Post, Kottayam Dt. Kerala -686579

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

Contents 1. Developed Model to Control Congestion on Converge Network……….[1] 2. An Efficient Approach for Discovering Impact Factor of E-Books using EigenFactor and UCINET…..[6]

3. Toward a 21st Century Crutch Design for Assisting Natural Gait……[12] 4. Optimization of machining parameters and tool selection in 2.5D milling using Genetic Algorithm…….[21]

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

Developed Model to Control Congestion on Converge Network John S. N#1., Okonigene R. E#2., Akinade B.A#3., Kasali. I. #4., Adewale A. A#5.

#1,5

Department of Electrical and Information Engineering, Covenant University, Ota, Ogun State, Nigeria.

1

Johnsam8@hotmail.com wale_yink@yahoo.com

5

#2

Department of Electrical and Electronics Engineering, Ambrose Alli University, Ekpoma, Edo State, Nigeria.

2

robokonigene@yahoo.com

#3,4

Department of Electrical and Electronic Engineering, University of Lagos, Lagos State, Nigeria.

3

bayonleakinade@yahoo.com 4 ibikunlekasali@yahoo.com

Abstract— Congestion control techniques like Active Queue Management (AQM), Carrier Sense Multiple Access/Carrier Detection (CSMA/CD) have not proven to be very efficient in the presence of overwhelming complex converge network. Thus, vast packets in a complex converged network leads to collisions, network degradation and high degree of packet loss. Bandwidth utilization factor has a high effect on the network such that controlling the level of utilization via the management of the number of users and the amount of packets on the network rendered the latency very insignificant. As a consequence of this, high throughput and very minimal packet loss was achieved in the experiment. This was confirmed analytically by varying the utilization factor between 40% and 90% while keeping other parameters in the experiment constant.

complex response functions that are required to model TCP performance in more congested environments with retransmit timeouts [3].

We considered and modeled an existing complex corporate converge network in order to evaluate and profound realistic solution to controlling congestion on the network. To derive the mathematical model the following postulations were considered: • The delay along the network path was summarized into three classes: propagation, serialization and queue delay.

• The packet sizes from the nodes were of the same length and at the router the service time Ts, for the packets was constant.

• During network congestion, packet loss indications were exclusively “through triplicate acknowledgement”, no timeout was considered.

Figure 1 shows the sources of delay on a converged network between the sources Host A to Host N on the network, with model processing delay, packet queuing and propagation delays along the link. Based on these assumptions, the net delay on the network was given as follows: Net delay = Propagation Delay (Tp) + Serialization Delay (Ts) + Queue Delay (Tq).

Keywords: Congestion, Active Queue Management, Converged Network, Bandwidth.

I. INTRODUCTION The study on congestion control was pioneered

by Van Jacobson and this was called the TCP/IP congestion control [1]. The High – Speed TCP [HSTCP] for Large Congestion windows was introduced

by Sally Floyd as a modification of TCP’s congestion control mechanism for use with TCP connections with large congestion windows [2]. It overcomes Standard TCP’s difficulty of achieving a large congestion window in environments with very low packet drop rates. High-speed TCP proposes a small modification to TCP’s increase and decrease parameters. It is designed to have a different response in environments of very low congestion event rate, and to have the standard TCP response in environments with packet loss rates of at most 10−2. In environments with low packet loss rates (typically lower than 10−3, it is possible to ignore the more

The propagation delay is described as the time it takes a signal to physically traverse the network path from source to destination [4]. This is a function of the distance of separation between the sender and the receiver and also the speed of light. It is assumed that any signal passing through a Fiber or wire does so with two-third of the speed of light ‘c’ [5]. Taken x as the distance across the network path from the source to the destination, propagation delay is given by:

1

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

x Propagation delay, T p = (sec), 0.667c

(1)

Figure I Sources of delay on a converged network. (Hosts A to Host N: packet flow sources)

For a sender to transmit a bit of signal into the outgoing queue, the serialization delay was given as follows [10 - 12]: Serialization Delay, as

=

A = (1 – ) B

Ts =

In most cases the Maximum Transmission Unit (MTU) was less than or equal to the Maximum Segment Size (MSS) (MTU ≤ MSS), Therefore,

N N = A (1 − ρ ) B

Throughput max =

and

Nµ (µ − λ )B

MTU (sec),

(5)

(2)

where N - number of packets size sent by the sender A - Available bandwidth on the channel, - Bandwidth utilization factor B - Bandwidth capacity of the link. λ - Inter-arrival rate of the packets, µ - mean service time at the node.

Queue delay, Tq =

λ2 l = (sec), B (1 − ρ ) Bµ

l=

λ2 x Nµ + + 0.667c ( µ − λ ) B (1 − ρ ) Bµ

II. Analysis of Results and discussion The net delay was obtained by considering the total delay experienced by the complex network i.e. propagation delay, serialization delay, and queue delay. To control the level of bandwidth utilization factor via the management of the number of users and the amount of packets on the network, data were generated from the mathematical model. The results were analyzed as follows: Firstly, we considered the following input parameters: Distance = 100m, Packet Size (Kbyte) = 200 – 400, MTU = 1500, Bandwidth Capacity = 100Mbps and Bandwidth utilization factor ρ = 0.4. The results of input parameters are shown in table I.

also,

(3)

where l is the length of queue.

When two nodes are connected together, the queues experienced by each packet traveling through the medium are independent of any successful arrival of the packet [6]. Latency is the summation of delays inherent on the network [7]. Thus, the net delay was deduced as follows:

Table I Generated Results given ρ = 0.4

λ2 x Nµ T p + Ts + Tq = + + 0.667c ( µ − λ ) B (1 − ρ ) Bµ (4)

2

Packet Size, (Kbyte)

Latency, (ms)

Throughput, (Mbps)

200 250 300 350 400

0.006 0.0068 0.0076 0.0084 0.0093

93.75 88.9 65.2 60.1 57.4

Packet Loss Rate, % 0.0025 0.0027 0.0028 0.0032 0.0038

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

(b)

(a)

(c)

Figure 2 (a) Response of latency to packet size, (b) Graph of response of packet size to throughput, (c) Graph of packet size against packet loss.

Figure 2(a) shows the response of latency to packet size. Increased in the packet sizes resulted in an increased in the latency. It was observed that the increment was on a gradual note, which suggested that the network had not reached a congested state.

Figure 2(b) represented the relationship between packet size and throughput. An increase in the size of the packets led to the reduction in throughput. It was observed from the graph that packet size was inversely proportional to throughput and the higher the packet size, the lower the throughput. A sharp change was observed in the throughput when the packet size was increased to 300Kbytes, with a reduction in value from 89Mbps to 65Mbps, this further point to a gradual packet build-up on the network. At the utilization value of ρ = 0.4, the packet loss underwent a steady increment, though mostly negligible, however it was of considerable value when compared in respect of the packet size between 200Kbytes and 400Kbytes with a difference of 13packets, Figure 2(c). This proved that the network performed efficiently when the network dropping of packets has a very negligible value.

Using the same mathematical model, however with bandwidth utilization factor of ρ = 0.9 considered, the average throughput reduced drastically as compared to the same packet sizes with lower bandwidth utilization factor of 0.4. The highest throughput obtained was 37.5Mbps as compared to 93.7Mbps when the utilization factor was 0.4. The latency also increased significantly from 0.042ms at 200kbyte to 0.079ms at 400kbyte. The packet loss rate was also significant, as the number of packet dropped in the network was higher (37 packets) as compared to when the utilization was 0.4. This indicates a better performance at lower bandwidth utilization factor of ρ = 0.4 compared to the utilization of ρ = 0.9.

Secondly, we considered the following input parameters: Distance=100m, Packet Size (Kbyte) =200 – 400, MTU=1500, Bandwidth Capacity=150Mbps and Bandwidth utilization factor ρ =0.4. The result shows in Table 2.

3

NTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711)

VOL.1 NO.8 AUGUST 2011

(b) (a)

(c)

4

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 Figure 3 (a) Graph of packet size against latency, (b) Graph of throughput against packet size and (c) Graph of packet size against packet loss with utilization factor of 0.9 with 150Mbps.

assumptions with regards to distance, constant range of packets, network bandwidth, and the variation of the bandwidth utilization factor between 40% and 90%. It showed some encouraging results when the bandwidth utilization was modified at 40%, but not when the utilization rises to 90%. At 90% the network began to show very feasible signs of congestion with low network performance.

Table II Generated results for Bandwidth capacity = 150Mbps & ρ = 0.4

Packet Size (Kbyte)

Latency (ms)

Throughput (Mbps)

Packet Loss Rate

200

0.0035

149

0.0005

250

0.0036

105

0.0008

REFERENCES [1]

300

0.0037

94.2

0.001

350

0.0038

88.6

0.0016

400

0.0042

85.4

0.0019

Jacobson V. “Congestion Avoidance and Control” ACM, Proceedings of SIGCOMM, Stanford, CA, 1988.

[2] S. Floyd. “Highspeed TCP for large congestion window”, Internet Draft draft-floyd-tcp-highspeed-01.txt, February 2003.

[3] Eric He, Rajkumar Kettimuthu, Sanjay Hegde, Michael Welzl, Jason Leigh, Chaoyue Xiong and Pascale Vicat- Blanc Primet, “Survey of Protocols and Mechanisms for Enhanced Transport over Long Fat Pipes”, Data Transport Research Group,2003/2004.

[4] Benedict Chung Wong “Two finite queues in tandem with pure delay and overflow” Technical Report submitted to Dept. of Industrial and System Engineering, Ohio State University, Columbus, Ohio, July 1975.

[5] Michele C. Weigle, Pankaj Sharma, and Jesse R. Freeman IV, “Performance of Competing High-Speed TCP Flows” Networking 2006. Networking Technologies, Services, and Protocols; Performance of Computer and Communication Networks; Mobile and Wireless Communications Systems, pp. 476-487, 2006.

[6] William Stallings “Queuing Analysis “A practical guide to an essential tool for computer scientists” Prentice Hall 2000.

[7] S. N. John, R. E. Okonigene, A. Adelakun: Impact of Latency on Throughput of a Corporate Computer Network, Proceedings: The 2010 World Congress in Computer Science, Computer Engineering, and Applied Computing (WORLDCOMP’10), Annual Summer Conference on Modeling Simulation & Visualization Methods (MSV’10), Las Vegas, Nevada, USA, July 12-15, 2010, pp.282-287.

The results in Table II are plotted in Figure 3 (a), (b), (c). This shows an increased in the bandwidth capacity.

In this case the highest achievable throughput was 149 at 200 Kbyte. While the throughput for an average file size of 400 Kbyte gave 85.4 with a lower packet loss as against what was observed when the bandwidth was .

Also, the latency was small compared to what we had in Figure 2. This clearly indicates a better network performance when the bandwidth was increased from 100Mbps to 150Mbps keeping bandwidth utilization factor constant at 0.4.

The bandwidth capacity increment generated a direct improvement on the throughput as compared with when the bandwidth was 100Mbps, as there was better network performance with a higher throughput and a distinctive lower packet loss. The latency value is very low compared to when the bandwidth was 100Mbps.

These further points to better network performance with a higher bandwidth capacity of 150Mbps at a utilization factor of 0.4 compared to when the bandwidth was at 100Mbps.

Also, the mathematical model was used with the same input parameters as Table 2, however, with utilization factor of ρ = 0.9. The maximum throughput achieved was 53.5Mbps at 200kbyte, and the minimum 43.7Mbps at 400kbyte. Latency rose to as high as 0.038 compared to 0.0042 when the utilization was 0.4. This is an important model to ascertain the behavior of complex networks, considering the effect of utilization.

III. CONCLUSION In this paper, a model for controlling the impact of link congestion on a converged network was presented. The model expresses throughput, latency and packet loss rate as a function of TCP and network parameters. The model was used to generate results with some

5

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

An Efficient Approach for Discovering Impact Factor of E-Books using Eigen Factor and UCINET Ms. D. Saraswathi# Dr. A. Vijaya Kathiravan* Mr.G.Sivakumar# Ms.R.Kavitha+ #Lecturer in Computer Science, K.S.Rangasamy College of Arts & Science (Autonomous), Tiruchengode-637215, Namakkal, TN, INDIA. * Assistant Professor in Computer Science, Government Arts College, Salem-07, TN, INDIA. +Assistant Professor, Dept. of MCA, AIMIT, St. Aloysius College(Autonomous), Mangalore, Karnataka, India saraswathi.ds@gmail.com, vijayakathiravan@gmail.com,sivakumarsen@gmail.com, kavitha.rajamanii@gmail.com

Abstract-An Impact Factor is one measure of the relative importance of a journal, individual article or scientist to science and social science literature and research. Each index or database used to create an impact factor uses a different methodology and produces slightly different results, revealing the importance of using several sources to judge the true impact of a journal's or scientist’s work. In the web environment, impact factor is measured through the number of hyperlinks counts and number of WebPages. The concept of self-citation is replaced by self-links, i.e., the links within the websites and citation is replaced by in-links, i.e., the links coming outside the websites. As we know, WIF is the logical sum of external and self-link WebPages divided by number of web pages found on that particular websites. There are number of way to find the impact of journal, paper, and Web sites etc. In this proposed system is going to find impact factor of E-books by using EigenFactor and the links of E-books represented by using UCINET software.

The link of E-books can identify based on the degree, betweenness of the link. This system is used to measure the quality of E-books and to know how many of them referring the E-book. Most of the E-books are downloaded from the Web or require pages can read from the Web site itself. This was done by means of a citation analysis and a reader survey. For the citation analysis, impact factor, citing half-life, number of references per article, and the rate of self-references of a periodical were used as indicators. Webometric data have been collected through Yahoo! And Google search engines using special query syntax.

I. INTRODUCTION The World Wide Web has now become one of the main sources of information on academic and research activities, and therefore it is an excellent platform to test new methods of evaluating webometric activities.

However the world scientific community has not yet accepted the Web as a full supplement or a complement to traditional scientific publishing. The science of webometrics (also cybernetics) tries to measure the World Wide Web to get knowledge about the number and types of hyperlinks, structure of the World Wide Web and usage patterns.

According to BjĂśrneborn and Ingwersen (2004) [11], the definition of webometrics is "the study of the quantitative aspects of the construction and use of information resources, structures and technologies on the Web drawing on bibliometric and informetric approaches." Webometrics is (a) a set of quantitative techniques for tracking and evaluating the impact of web sites and online ideas and (b) the information science research field that developed these ideas.

Webometric techniques include link analysis, web mention analysis, blog analysis and search engine evaluation, but from the perspective of digital library evaluation the main method is link analysis. Why can analyzing web hyperlinks help evaluate digital repositories? The reason is that the links to a web site can reveal useful information about how popular it is, which pages or resources are the most popular, why it is popular and where it is popular. Whilst all this information can also be gained from web server log file analysis, the latter can normally only be

Keywords: Scientometrics, Webometrics, Cybermetrics, Bibliometrics, impact factor.

6

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 conducted with permission of a site’s webmaster. In contrast, link analysis can be applied to any web site. This means that link analysis can be used to evaluate a web site by comparing it to its competitors or to similar web sites and can also be used to identify missed audiences for a site. Links can reveal information about web sites because each link to a web site may be created to direct visitors to it. The link author believes that the target site is important or useful. For example, the course pages for an archaeology degree may contact links to the New Library of Alexandria for its images of ancient Egyptian artefacts. From the opposite perspective, discovering all the links to the New Library of Alexandria web site would give useful insights into who was using it and why. Of course, most people using a web site will not create a link to it but a link analysis can still give indicators about likely users and uses.

Fig 1: Taxonomy of Informetircs

7

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 scientometrics are two closely related approaches to measuring scientific publications and science in general, respectively. In practice, much of the work that falls under this header involves various types of citation analysis, which looks at how scholars cite one another in publications. This data can show quite a bit about networks of scholars and scholarly communication, links between scholars, and the development of areas of knowledge over time.

II. TAXONOMY OF INFORMETIRCS Informetrics is the study of quantitative aspects of information. This includes the production, dissemination and use of all forms of information, regardless of its form or origin. As such, informetrics encompasses the fields of • • • •

Scientometrics, which studies quantitative aspects of science.

Webometrics, which studies quantitative aspects of the World Wide Web. Cybermetrics, which is similar to webometrics, but broadens it's definition to include electronic resources.

Bibliometrics, which studies quantitative aspects of recorded information.

III.Webometrics Analysis In the wake of Internet/Web developments, some bibliometricians drew analogies between Webbased and research documents and came up with the idea that the scientific content of the Web could be analysed in the same way as the science journal system. Webometrics is the quantitative analysis of web phenomena, drawing upon informetric methods, and typically addressing problems related to bibliometrics. Webometrics [12] was triggered by the realisation that the web is an enormous document repository with many of these documents being academic-related [12]. Moreover, the web has its own citation indexes in the form of commercial search engines, and so it is ready for researchers to exploit. In fact, several major search engines can also deliver their results automatically to investigators’ computer programs, allowing large-scale investigations. One of the most visible outputs of webometrics is the ranking of world universities based upon their web sites and online impact [6]. Webometrics includes link analysis, web citation analysis, search engine evaluation and purely descriptive studies of the web. Webometrics, a modern, fast-growing offshoot of bibliometrics is reviewed in detail [9].

Informetrics can be classified based on the specialization as given below in Fig 1.

I. Bibliometrics Analysis Bibliometrics is the application of mathematical and statistical methods to publications (from biblos: book and metron: measurement). Bibliometrics is often used to assess scientific research through quantitative studies on research publications [7]. Bibliometric assessments are based on the assumption that most scientific discoveries and research results eventually are published in international scientific journals where they can be read and cited by other researchers. The number of citations to a journal article can be considered to reflect the article’s impact on the scientific community. Applied bibliometrics, as it is used today, analyses the number of scientific articles published by a selected number of authors, citations to these articles and connections between articles, authors and subjects.

III.THE PROPOSED ARCHITECTURE FOR DISCOVERING IMPACT FACTOR OF E-BOOKS

II.Scientometrics Analysis In order to properly to know, we first have to define the impact factor. The ISI (the Institute for Scientific Information) impact factor was designed by Eugene Garfield around 1960 as a means to measure the impact of a specific journal, and it gives an average value on how many times an article in the journal has been cited. It is defined as the average number of citations given in a specific year to documents

Scientometrics is the science of measuring and analyzing science. In practice, scientometrics is often done using bibliometrics which is a measurement of the impact of (scientific) publications. It is part of sociology of science and has application to science policy-making. It involves quantitative studies of scientific activities, among others, publication, and so overlaps bibliometrics to some extent [10]. Bibliometrics and

8

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

published in that journal in the two preceding years, divided by the number of documents published in that journal in those two years. In this proposed system is used to find out the impact factor of E-Books and determine the quality of their EBooks. The E-books available in various web server of the web are downloaded by the crawler using either depth first crawler or breadth first crawler and the document corpus collected are stored in the web page repository i.e., a web warehouse. The E-books URL addresses in the web repository are indexed using indexing methods. The URL addresses and the structural connectivity of the E-books URL addresses are stored in the link server and the indices with respect to their E-book ID are stored in the index server The E-Books of impact factor is calculated based some parameter such as E-Books count, Citations and impact factor, and Co-citation by using EigenFactor and related links of that E-books represented in diagrammatically by using UCINET. This is done by using UCINET Software [13], measure centrality of degree and betweenness in network. It is based on the set of the users most cited E-books and the number of citations that they have received in other users. The impact factor of E-books is returned to the client and it is possible to measure the quality of their E-books. Fig 2 gives an overview of this architecture.

9

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

presence of the included universities. The findings revealed that Jordanian universities represent 40 percent of the top ten universities with the revised web impact factor. However, this was not the case in terms of the universities’ web presence. Results indicated a strong correlation between external links and web presence.

IV.RELATED WORKS Jalal, Biswas and Mukhopadhyay [1] had shown in their hyperlink study for the state universities of West Bengal that IIT Kharagpur occupied the first rank among the universities based on WISER indicators and Uttar Banga Krishi Vishwavidyalaya got the last position from the point of view of webometric ranking Smith and Thelwall [5] calculated Web Impact Factors (WIFs) for Australian universities using a specially designed crawler and the AltaVista search engine. Links between UK, Australian and New Zealand Universities had been reflected. Both the number of pages at the site, and the number of academic staff members, were used as measures of the size of the universities.

V. RESULTS AND DISCUSSION Step 1: The social network is visualized as a graph having each user as nodes and their friend’s links as edges as shown in fig 2. The first target is to identify nodes which have maximum indegree. The indegree of each node is calculated by counting the number of links pointing towards a node. For this aspect we consider two factors, 1. Number of messages arrived to an user per a day (MPD) 2. Number of unique dispatchers(UD) 3. Number of identical dispatchers (ID).

The WIFs were compared with conventional measures of research output: rankings by Asiaweek magazine, the number of publications per staff member, and the number of citations per staff member. There is a good correlation between the crawler and AltaVista in estimating the link counts. The WIFs do not appear to correlate well with conventional measures of research output.

We find out a nodes indegree by using below given formula ((MPD/UD) – ID).We eliminate the inflowing links from identical users by subtracting the number of identical dispatchers from the computed result.

They also discussed some of the methodological issues in the calculation of WIFs. Mukhopadhyay [8] tried to explore the possibility of research in the field of webometrics in the educational institutions in India using Web Impact Factor (WIF). Li [3] studied hyperlinks extensively by applying existing bibliographic methods and made an exhaustive review the development of WIF. Li pointed out the origin of WIF and techniques for data collection using commercial search engines. The study also highlights the development of WIF - origin, traditional measures and its improvements.

Jayshankar and Babu [4] in a webometric study examined the websites of 45 universities in Tamil Nadu to analyse the number of WebPages, links, calculate various types of WIFs. The result found that although some universities of Tamil Nadu have quite large number of WebPages but very low number of inlinks and hence low WIF. Ravikumar [10] investigated the link pattern of selected academic libraries in India using UCINET computer software to visualize the network pattern that existed among peer group libraries.

Step 2: We explore the relationship between each user by finding out the relationship such as friend, Classmate, colleague etc.We find the whether there is a path from each node to every other node in the graph by finding the connectivity of the graph. We also find the neighbors of each user by considering the adjacency (those nodes connected with a path of length 1. We also explore how reach a node at an optimum speed and duration by finding out the shortest path using Dijkstra's algorithm.

Step 3: We use node-by-node matrix which has as many rows and columns as there are nodes in a network. The data entry is done by placing O’s or 1’s into a spreadsheet. The O’s represent no connection and 1’s signify a link which represents a binary matrix.

An input values are given manually in Table 1, the centrality of the each node by performing Degree Centrality and Betweenness Centrality in Table 2 and corresponding Web Graph in Fig 3.

Elgohary [2] made a study in order to investigate the WIF of Arab universities. The study included 99 universities representing 20 Arab countries.

The advanced search facility of AltaVista was used for data collection. Two rounds of data collection were conducted to retrieve the links as well as the web

10

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

REFERENCES 1. Jalal, Biswas, Mukhopathyay, Web Impact factor and link analysis of selected Indian Universities. 57(2010),109-121 2. Elgohary A, Arab Universities on the Web: A webometric Study, The Electronic Library, 26(3)(2008) 374-386.

3. Li X, A review of the development and application of the Web Impact Factor, Online Information Review, 27(6)(2003) 407-417.

4. Jayshankar R and Ramesh Babu B, Websites in universities in Tamil Nadu: a webometric study, Annals of library and information studies, 56(2)(2009), 69-79 5. Smith A and Thelwall M, Web Impact Factors for Australasian universities, Scientometrics, 54(1/2) (2002) 363-380.

6. WISER, Web Indicators for Scientific Technological and Innovation Research: A Survey of Practice, December 2005 43. Ingwersen, op. cit., 236-43 7. Jahaysnkar R and Ramesh Babu B,Narendra Kumar.A.M, Siruseri, Dr.Nageswara Rao.P, Web Credibility of Engineering College Websites Affiliated to Anna University, Tirunelveli (Tamil Nadu, India), October 19-22, 2010.

8. Mukhopadhyay P S, The calculation of Web Impact Factors for educational institutes of India: A Webometric analysis In. Information Management in e-Libraries, 26-27 February 2002.

9. Vishal Kumar Saxena, Webometrics, Informetrics, Scienctomerics, October 19-22, 2010.

10. Ravikumar S, A webometric study of selected academic libraries in India using link analysis, Proceedings of Fifth International Conference on Webometrics, Informetrics and Scientometrics. Dalian, September 13-16, 2009.

11. http://en.wikipedia.org/wiki/Webometrics 12. http://webometrics.wlv.ac.uk/ 13. www.analytictech.com/ucinet/.

Table 1: Sample Input values for UCINET

Table 2: Betweenness & Degree Centrality Measures

Fig 3: Web Graph for Link Structure

VI. CONCLUSION This study has been exploratory and there is scope for future Webometrics research in this area. It would be useful to carry out a more comprehensive study comparing more E-Books and identify the quality of E-Books. There was a lot research in measuring the quality of Web sites for few Universities and institutions.

Webometrics is concerned with measuring aspects of the web: web sites, web pages, parts of web pages, words in web pages, hyperlinks, web search engine results. The importance of the web itself as a communication medium and for hosting an increasingly wide array of documents, from journal articles to holiday brochures, etc.

11

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

Toward a 21st Century Crutch Design for Assisting Natural Gait Hyosang Moon #1, Herbert Baumgartner *2, Nina Patarinsky Robson *3 #

Department of Mechanical Engineering, Texas A&M University College Station, TX 77843, USA 1

*

hsmoon@neo.tamu.edu

Department of Engineering Technology and Industrial Distribution, Texas A&M University College Station, TX 77843, USA 2

herbert.baumgartner@neo.tamu.edu 3 robson@entc.tamu.edu

Abstract—In order to resolve the disadvantages of conventional crutch designs (i.e. underarm and forearm), st a novel crutch design is presented with the name of 21 Century Crutch. Prior the design, human natural treadmill walking is monitored by a 3D Motion Capture System and acquired a reference end–foot trajectory with a ‘teardrop shape’. Considering the design objectives, natural human walking and comfort, and other factors such as load capacity and weight of the device, the final design was determined. In order to satisfy the design objectives, a kinematic synthesis previously worked by Robson and McCarthy [14] is applied to test if the end–foot trajectory of designed crutch smoothly follows the desired reference, ‘teardrop shape’ in the vicinity of two specified task points, heel strike and toe off. For that reason the leg was synthesized and animated in Mathematica as a RR planar kinematic chain, where the first hinge joint/fixed pivot was located at the hip and the other hinge joint/moving pivot is at the knee joint. A prototype of the final design was fabricated and its performance was tested by 2 mph treadmill walking.

1917 (US patent No. 1244249) [16]. From these patented designs, numerous modification works have been patented for comfort and safety.

Though the current crutch designs are inexpensive solutions to fulfil their main function, body weight support, they have non–negligible disadvantages follows: 1) since the armpit or forearm are not supposed to support a large load, normally, continuous stress on them can cause degraded motion control and even nerve damages [2], 2) important upper limb functional motions in daily living (e.g. arm reaching motion, hand grasping and manipulating) are limited, 3) decreased degrees of freedom (DOF) can induce compensatory joint motions and unnatural locomotion, which can be possibly developed as chronic pathologies after long term of usage, and 4) ambulation with underarm and forearm crutches demands high cost of energy consumption [17].

The iWalk–Free (iWALKFree, Inc., USA) [18] shows a typical example of novel crutch design to overcome the disadvantages described above. With a stable load support through 90° flexed knee, the device could reduce discomfort significantly by freeing upper limb motions. However, it is only applicable to injuries on below tibia and fibula region (i.e. foot and ankle). Also, due to lack of knee joint DOF, the hip and pelvic joints tend to make an abnormal motion pattern to ensure the foot clearance during the swing phase of gait.

The idea behind the proposed crutch design is at assisting ambulation of patients who have injured one of their lower legs (i.e. below femur, including knee joint). These injuries include ankle sprain, fracturend/or cracks on tibia, fibula, foot bones, or any combination of them. The skeletal anatomy of human leg is shown in Fig. 1.

Keywords: Crutch design, Human walking, Planar chain I. INTRODUCTION The crutch is the simplest and reliable way to compensate the mobility of people with lower limb injuries by supporting their body weight during locomotion in daily life (e.g. ascending/descending stairs and walking). It provides a stable environment for recovery by allowing the injured body part in a load free condition. It is known that the crutches have been used for 5,000 years [15]. People used fallen tree branches as supporting sticks to help balancing or ambulating wounded body. From its primitive forms, the current configurations of underarm and forearm crutches have been evolved through various empirical designs. The first US patent for a crutch was issued to Tuttle (US patent No. 332,684) [16]. The first commercialized form of forearm crutch design was patented by a French mechanical engineer, Schlick, as a walking stick in

12

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

Hip Joint

guide the foot through multiple number of separate positions [8,9]. The required kinematic specifications in the synthesis are number of task positions with specified end–foot velocities and accelerations. The Section IV introduces the preliminary design and is followed by the Section V which describes the final design and its prototype fabrication. The experimental result is shown in the Section VI. Finally conclusion and future works are discussed in the Section VII.

Fibula Ankle joint

Knee joint Femur

II. DESIGN OBJECTIVES AND SPECIFICATIONS Before the actual design work, the details of design objectives (i.e. natural walking motion and comfort) and other factors are described in this section.

Tibia

Fig. 1. Skeletal anatomy of human leg (image source [5])

A. Natural Walking Motion Like iWalk–Free design, the load supporting structure should be transferred from the upper limbs (e.g. armpit or forearm) to the lower limbs. This allows the user to keep more natural walking motion compare to conventional crutch designs (see the disadvantages of conventional crutch design described in Section I). For realizing even more natural walking, the proposed novel crutch in this paper is designed to mimic the normal end–foot trajectory so that important gait events and properties (e.g. heel strike, toe off and foot clearance) can be performed smoothly and effectively.

In order to acquire a reference gait data, a person’s normal treadmill walking with 2 mph speed was collected via 3D Motion Capture System (Vicon MX, Vicon Inc., UK). Eight reflective markers were attached on the subject’s right leg as shown in Fig. 2(a) with highlighted circles. Fig. 2(c) represents the single gait cycle obtained by the 3D Motion Capture System. For visualizing the actual end–foot trajectory, the positions of ankle marker on the sagittal plane is plotted (see Fig. 2(b)). Note that the geometrical shape of the reference trajectory looks like a ‘teardrop’ for each gait cycle.

In this paper, a novel crutch design is developed with st the name of 21 Century Crutch. In order to cope with significant disadvantages of conventional crutch designs, two distinct design objectives, natural walking motion and comfort, are set. On its way to the final design, a preliminary concept was developed and tested for comfort and natural gait trajectory of the foot.

Next, after the preliminary concept, a planar chain with RR (Revolute– Revolute) joint was tested in Mathematica. A kinematic synthesis of linkage design with acceleration specification is adopted to achieve the design objectives (i.e. natural walking motion and comfort). The goal of this synthetic approach is to obtain the solutions to a given task specification (i.e. end–foot trajectory of normal treadmill walking) in order to test the design of the mechanical linkage that can move the end–foot smoothly through the specified task.

Unlike syntheses with position and velocity specifications, research in the synthesis of serial chains to achieve acceleration requirements is limited. It is primarily found in the synthesis theory for planar RR chains and the work by Chen and Roth [10] for spatial chains. The use of second order effects first appears in the analysis of grasping in a work by Hanafusa and Asada [11], where planar objects are grasped with three elastic rods.

Rimon and Burdick [12,13] showed that acceleration properties of movement can be used to effectively constrain a rigid body for part–fixing and grasping applications. Robson and McCarthy [14] presented a technique for deriving geometric constraints on position, velocity and acceleration from contact and curvature task requirements. These constraints yield design equations that can be solved to determine the dimensions of the serial chain. In this paper, despite the fact that the dimensions of the serial chain are pre– determined (i.e. size of the crutch must be customized to the human subject), the developed kinematic synthesis theory is applied to test the end–foot trajectory and its smoothness.

The detailed design objectives and required specifications are described in Section II. After that, the Section III considers the synthesis of planar chains to

500

z-position (mm)

400 300 200 100 0 -100 -200

(a) Attached marker positions

0

200 400 y-position (mm)

600

(b) Acquired foot trajectory

(c) A cycle of walking motion in the 3D Motion Capture System Fig. 2. Capturing a normal treadmill walking cycle

13

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

B. Comfort Comfort is an important factor which directly affects the subject’s ADL (activities in daily living) performance.

It is deeply related with the ergonomic design, which pursues to reduce the user’s fatigue. In this paper, the secure attachment mechanism and its location on the subject’s body are considered to maximize the comfort.

4) High Mobility In order to maximize the performance on ADL of the proposed design, a high mobility is needed. The high mobility will allow the subject to be able to explore challenging environments (e.g. uneven terrain and obstacles) even after the injury.

C. Others

5) Usability Due to the crutch’s high frequency of use, its usability should be considered in the design stage. Easy put on and off mechanism and adjustable size mechanism can be typical examples of this factor.

1) Load Bearing Capacity Since the injured patients solely depend on the crutch to support their body weight during locomotion, it becomes the most critical design factor which is directly related to the safety issue.

The crutch should be able to assist most of locomotion in daily living such as level walking, sitting down, standing up and ascending/descending stairs. In order to specify the load bearing capacity, the empirical data on maximum loading of the knee joint during ADL is adopted. According to the study of Kutzner et al. [1], average peak resultant forces on the knee joint was 261% BW (body weight) for level walking while the maximum loading was applied during descending stair (346% BW). Therefore, for instance, the crutch should be designed to sustain at least 692 lb to support a 200 lb person without counting a safety factor.

III. KINEMATIC SYNTHESIS OF PLANAR LINKAGES The kinematic synthesis of planar linkages is adopted to test the final crutch design (i.e. RR chain, see Section V). The pattern of one gait cycle consists of swing and stance phases. Since the leg dynamics of each phase is totally different, two phase transition events, heel strike (i.e. when the foot initiate its contact with the ground at the beginning of a stance phase) and toe off (i.e. when the foot terminates the ground contact at the ending of a stance phase), contain useful information to characterize the dynamics and kinematics of gait.

In this paper, the kinematic specifications (i.e. position, velocity and acceleration) at those two events in the reference gait trajectory will be derived by a planar synthesis proposed in [14]. The derived kinematic specifications are utilized to manipulate the end–foot trajectory of the designed crutch to mimic the desired one (i.e. natural human walking). Also, by applying the derived kinematic specifications to the geometric design equations of RR planar chain, an appropriate position of RR chain’s base frame (i.e. attachment location of the crutch on the subject’s thigh) can be determined. The purpose of this section in the context of design objectives can be described as follows: 1) obtaining required kinematic specifications at heel strike and toe off events to realize a natural gait pattern (i.e. ‘teardrop shape’, see Fig. 2(b)) of the designed crutch, and 2) obtaining an accurate attachment location of the designed crutch on the subject’s thigh to enhance its fit, dynamic stability and comfort.

In what follows, a planar synthesis approach recently developed by Robson and McCarthy [14] is presented.

2) Light Weight As the proposed design is intended to be attached on the lower limb, it imposes an additional mass on the subject’s body and alters weight of the limb which induces changed dynamics during locomotion. Also the varied inertia can cause compensatory motions of other DOF (e.g. pelvic and hip joint) during a leg swing motion and they can be further developed as pathological abnormal joint patterns. Therefore in order to minimize these side effects, the device is required to be designed in light weight as possible. As a reference, according to an anthropometric rule, weight of a leg is approximately 16.1% of BW [19].

3) Passive Actuation The design can contain any moving mechanisms to enhance its functionalities (e.g. less energy consumption, impact force absorption and increased dynamic stability). Higuchi et al. [3] adopted linear actuators with telescopic links to reduce user’s efforts while using a pair of underarm crutches. Mori et al. [4] added a mobile platform which can interact with a sensor integrated underarm crutches to enhance the mobility. However, mechanisms with active actuation systems incorporate complicated control and monitoring algorithms which can possibly cause unintended malfunction. Therefore, for the safety and simplicities in design, a passive actuation mechanism is considered in this paper.

A. Geometric Design of Planar Mechanical Linkages with Task Acceleration Specifications The walking motion can be assumed as a planar task on the sagittal plane which consists of positioning of the j foot, at the point, M (j=1,…, n), located on the reference

14

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 j -1

trajectory. If the kinematic specifications at the start (i.e. heel strike) and end points (i.e. toe off) are acquired, it is possible to manipulate the designed RR chain (i.e. crutch) to mimic the reference trajectory pattern. The required kinematic specifications are derived from contact and curvature constraints between the foot and the ground in the specified event positions. The derivation of these constraints is discussed in details in [14]. The movement of foot can be described by the parameterized set of 3x3 homogeneous transform matrix

 T ( t )  =  R ( t ) , d ( t ) 

−1 1   P j ( t ) = T0 j + T1 j t + T2j t 2 + ... T0 j  P j 2   1   =  I + Ω j t + Λ j t 2 + ... P j 2  

j

(3)

(4)

where

Ti j  =

d T  dt i j

(5)

0

j

j

1   P ( t ) = T ( t )  p = T0 j + T1 j t + T2 j t 2 + ... p 2  

(9)

j

2

(10)

Note that the link length, R, and the length of tool frame, H, which has its tip located at the ankle are known values.The first and second derivatives of (10) provide the velocity constraint equation

t =0

j

j

−φ12

d x 2 + d x 0φ12 + d y 0φ2   d y 2 + d y 0φ12 − d x 0φ2   0 

( P (t ) − B ) ⋅ ( P (t ) − B ) = R

The matrices [T0 ], [T1 ] and [T2 ] are defined by the position, velocity and acceleration of the foot in the j vicinity of each task position M , respectively.

Therefore, a point p in M has the trajectory P(t) defined by the equation j

−φ2

B. Design Equations In this paper, the end–foot kinematics is presented as a planar RR mechanical linkage. The design parameters for the planar RR chain are the coordinates B=(Bx, By) of the fixed pivot, located at the hip joint and 1 the coordinates P =(Px, Py) of the moving pivot, located at the knee joint, when the floating link is in its start position. The ankle is assumed to be rigid since the final design does not include an ankle joint (see Section j V). In each task position, the moving pivot P is constrained to lie at the distance R (i.e. length of the moving link, which connects the hip and knee joints) from B, which yields,

The next goal is to determine the movement of the foot, as defined by [T(t)]. The movement of the foot, relative to a world frame in the vicinity of a reference position, defined by t=0 can be expressed by the Taylor series expansion,

i

0

( 8)

[Ω ] and [Λ ] are the planar velocity planar acceleration matrices, respectively, which are defined by the foot kinematic specifications in the vicinity of the task j positions M .

(2)

( j = 1,..., n )

d x1 + d y 0φ1  d y1 − d x 0φ1   0

0

 −φ12   Λ j  =  φ2  0 

or

1 T j ( t ) = T0 j  + T1 j  t + T2j  t 2 + ...

2

−φ1

0 Ω  = φ1   0 j

where R(t) and d(t) represent a rotation matrix and a translation vector, respectively. A point p fixed in themoving body traces a trajectory P(t) in a fixed coordinate frame F such that

P ( t ) = T ( t )  p

(7)

where

(1)

 Px ( t )  cosφ ( t ) − sinφ ( t ) dx ( t )   px       Py ( t )  = sinφ ( t ) cosφ ( t ) dy ( t )   py   1   0 0 1   1    

j

Let p=[T0 ] P , which yields

d P ⋅(P − B) = 0 dt and the acceleration constraint equation

(6)

15

(11)

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

d2 d  d  P ⋅(P − B) +  P  ⋅ P  = 0 2 d t  dt   dt 

( )( ) V : ( Ω   D  P ) ⋅ (  D  P − B) = 0, A : ( Λ   D  P ) ⋅ (  D  P − B) + ( Ω   D  P ) ⋅ ( Ω   D  P ) = 0 Pj :  D1 j  P1 − B ⋅  D1 j  P1 − B = R2 ,

(12)

j

1

j

In order to determine the five design parameters, five design equations are required. Choosing one of the task positions to be the first and applying the relative displacement matrices

 D1 j  = T0 j  T01 

−1

1

1

2

1j

(i =1,...,n)

In the crutch design, n=2. Notice that [D11] is a 3 x 3 identity matrix. From the previously defined 3 x 3 velocity matrix, (8), the below can be derived.

(16)

By substituting (14) and (16) into (11), the velocity design equations for the two specified positions can be represented.

( Ω   D j

1j

)(

)

 P1 ⋅  D1 j  P1 − B = 0

( j = 1,...n )

(17)

Similar to (16), the below can be derived from the defined 3 x 3 acceleration matrix, (9).

d2 j P =  Λ j   D1 j  P1 dt 2

(18)

Also the substitution of (14) and (18) into (12), the acceleration design equations for the start position are derived.

( Λ   D j

1j

)(

) (

)(

( j = 1,..., n)

A. Design The design of TAMC1 is shown in Fig. 3. In order to protect the injured body part (i.e. lower leg region including knee joint), it is assumed that the user is in a cast, which covers the leg from the foot to just above the knee joint, with a slightly flexed knee (see Fig. 3), or is missing a part of or the entire one’s lower leg.

The user’s body weight is supported through a seat structure located within the thigh region (see Fig. 3(a)).

The large and flat thigh seat allows the body weight to be evenly distributed over the entire contact area. At the same time, not to interrupt the natural hip joint motion, the size and location of seat is carefully selected through experimental trials.

In a gait cycle, stance leg supports the whole body weight by keeping the knee joint nearly full extension state. Therefore the dynamics of entire body during the stance phase can be assumed as a rigid link inverted pendulum. Though there is a slight knee flexion during a stance phase, its range is small and its purpose is on diminishing vertical fluctuation of COM (center of mass) [6]. Adamczyk et al. [7] focused on the inverted pendulum like dynamics of the stance leg and showed that the metabolic cost can be reduced by adopting a rolling foot to take advantage of the stance leg dynamics. The TAMC1 design consists of two closed– loop linkages, 5–bar linkage for the upper link (see Fig. 3(b)) and 4–bar linkage for the lower link (see Fig. 3(c)), respectively. The upper link motion mimics the inverted pendulum motion while the lower link emulates the rolling foot motion. In order to reduce any impact forces and to propel the body forward in toe–off state (i.e. when the foot lose its contact with the ground at the end of a stance motion), a compression spring shock absorber is installed at toe region (see Fig. 3(d)).

(15)

d j P = Ω j   D1 j  P1 dt

1

1j

IV. PRELIMINARY DESIGN AND RESULT Based on the design objectives and specifications (see Section II), a closed–loop linkage type crutch, TAMC1, is developed as a preliminary design. The main purpose of TAMC1 is to test the performance based on comfort and natural gait.

By substituting (14) into (10), the constraint equation becomes the position design equation below.

1j

j

The algebraic solution to the set of four bilinear equations for an RR chain is presented in [20] and can be applied without any changes for the case of five position synthesis (i.e. design equations (20)).

(14)

( D P −B) ⋅( D P −B) =R

1j

1

1j

j

P =  D1 j  P

1

1j

j

(13)

1

1

j

(20)

1j

j

allows the coordinates P to be defined by the moving pivot as follow j

1

1j

)

1 1 j 1 j 1  P ⋅  D1 j  P − B + Ω   D1 j  P ⋅ Ω   D1 j  P = 0

(19) Therefore, for each of the n task positions, the position, velocity and acceleration design equations become

16

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 C. Experiment The fabricated prototype of TAMC1 was worn by a volunteered subject weighed 155 lb. Each link length is adjusted to fit the crutch to the subject. During the stand still posture, the TAMC1 could sustain the body weight.

However, the attachment mechanism was not secure enough to allow any locomotion.

(a) (b)

V. FINAL DESIGN CONCEPT From the tests on the preliminary design, the final design is determined as a serial chain with a thigh cuff attachment. This design is inspired by the passive walker, which was first introduced by McGeer [21]. The passive walking device is a human–like walking device which is driven by gravity only. Due to its anthropomorphic shape, it is able to approximate the natural human walking very closely. Also, the serial chain can be fabricated in a light weight due to its simple design.

(d) (c)

Fig. 3. The design of TAMC1 (a) Thigh seat (c) Closed–loop four bar linkages

(b)Closed–loop five bar linkages (d)Shock absorber

B. Prototype Fabrication Based on the design, a prototype was built to verify the performance. For the material selection, an aluminium alloy was selected for its low cost, relatively easy manufacturability, light weight and moderate strength. The prototype is shown in Fig. 4. In order to constrain the ROM (range of motion) of upper link (see Fig. 4(b)), two limiters (see Fig. 4(g)) and a tension spring (see Fig. 4(c)) are installed. Each link has multiple joint holes for adjusting the size of device.

A. Design and Prototype Based on the passive walking device design, several factors are modified according to the design objectives and specifications (see Section II) as follows: 1) For a secure attachment, thigh cuff is adopted and the subject’s thigh is regarded as the first link of the chain. This also allows controllable and more stable locomotion by incorporating the subject’s hip joint motion into the system.

2) Since the knee joint requires two different movements according to the walking phase (i.e. knee lock for a stance phase and free swing for a swing phase), an additional locking mechanism is adopted.

3) In order to guarantee a firm stability during a stance, a high stiffness ankle and foot mechanism is applied.

The final design of crutch, TAMC2, is shown in the Fig. 5. The entire thigh segment is covered by the cuff (see Fig. 5(a)). The thigh cuff consists of two parts and can be easily assembled by Velcro straps (i.e. usability– easy put on and off). Also for a secure attachment, each part has S–shaped interface (see the highlighted line in Fig. 5(a)) which can prevent relative translation. For the safety from unintended knee buckling, a prosthetic knee donated by a local prostheses and orthoses place is utilized (see Fig. 5(c)). The knee joint can be locked in any angle when a vertical force is applied (i.e. whenever the subject collapses). Also it is designed to be easily flexed with a small amount of force at the toe off posture and to be retracted when the foot loses its contact with the ground. The first link, thigh cuff, and the second link, knee prosthesis, are joined by a steel bracket to ensure the structural strength.

The fabricated prototype is shown in Fig. 6(a) and an enlarged picture of the prosthetic knee joint is represented in Fig. 6(b). Thermoplastic is utilized for the

(a)

(c)

(b)

(f )

(g)

(d) (e)

Fig. 4. The prototype of TAMC1 (a) Thigh straps (c) Tension spring (e) Shock absorber (g) Upper link limiters

(b) Closed–loop five bar linkages (d) Closed–loop four bar linkages (f) Thigh seat

17

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 The kinematic specifications obtained from the Fig. 7 are applied to the design equations to test the smooth movement of the RR chain throughout the task. Fig. 8 shows the RR chain going through the specified task, consisting of two positions, two velocities and one acceleration, defined in the first position.

(a)

(b) (c)

500

400

z position (mm)

300

(b) (a)

200

100

0

-100

Fig. 5. The design of TAMC2

-200 0

100

200

300

400

500

600

700

y position (mm)

(a) Thigh cuff (c) Prosthetic knee joint

(b)Joining bracket Fig. 7. The reference foot trajectory of one normal gait cycle (b)Toe off/second position (a) Heel strike/first position

B P

Fig. 8. The test result of the kinematic synthesis of RR chain B and P represent the fixed and moving pivot, respectively (a) Final prototype (b) Prosthetic knee joint Fig. 6. The final prototype, TAMC2

For the trajectory tests, equation (20) is used for finding locations of the fixed pivot (i.e. hip joint) and the moving pivot (i.e. knee joint) with respect to a fixed frame, located at the hip joint. The kinematic specifications at task points assist in shaping the foot trajectory. The trajectory shown in Fig. 8 represents the foot path after the contact and curvature constraints have been implemented into the task. Note that the tested trajectory with velocity and acceleration specifications is close to the desired one (i.e. ‘teardrop shape’) in the vicinity of the two specified positions, heel strike and toe off.

thigh cuff. For stable and stiff ankle–foot mechanism, a frame of bike seat is adopted. A tracking shoe is worn on the structure for shock absorption and natural look.

B. Testing the Foot Trajectory of the Final Design in the Vicinity of the Specified Positions As mentioned in the Section III, the two task positions of a gait cycle are determined as heel strike and toe off.

In order to test the foot trajectory of the final design, the kinematic specifications at two task points on the reference trajectory are derived. Fig. 7 shows a foot trajectory of one cycle of normal treadmill walking which is a part of Fig. 2(b).

18

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 natural walking motion and comfort, a kinematic synthesis of RR chain was applied. The kinematic specifications (i.e. position, velocity and acceleration compatible with contact and curvature constraint between the foot and the ground) at two important gait events (i.e. heel strike and toe off) are derived from the normal foot trajectory. By imposing the obtained kinematic specifications to the synthesis design equations, both factors, the locations of the fixed and moving pivots, as well as the foot trajectory have been considered, i.e. comfort and natural gait motion.

As future works, the knee joint of the crutch will be modified to more closely resemble the natural human walking motion. In order to maintain the knee retracting function while more natural knee flexion is required, a semi–active actuator such as an electromagnet can be adopted.

VI. TREADMILL W ALKING EXPERIMENT The volunteered human subject performed treadmill walking experiments with and without the TAMC2 prototype. In order to emulate the lower leg injured condition, the subject tied one’s leg to the waist (see the right side figure in Fig. 9(a)). To test comfort, equation (20) was used. According to the proper position of fixed frame from the kinematic synthesis, attachment location of thigh cuff was determined. The speed of treadmill was set as 2 mph for both experiments and the end–foot (i.e. ankle marker) positions are recorded by the 3D Motion Capture System. The acquired trajectories are compared as shown in Fig. 9(b). Unlike the normal trajectory or the tested trajectory (see Fig. 8), the result with crutch was formed flatter than the desired one. The main reason for that is the self–retracting mechanism of the prosthetic knee joint. Because of its stiffness, the knee joint extended faster than the hip flexion. This phenomenon eliminates and flattens the swing phase trajectories.

ACKNOWLEDGMENT The authors greatly acknowledge the support of NASA TSGC design challenge program, NASA mentor, Regnald Berka, Central Texas Orthotics and Prosthetics, as well as the members of the design team, Andra Arnold, John Bonnette, Tom Carney and Daryll Zalesak.

REFERENCES [1] I. Kutzner, B. Heinlein, F. Graichen, A. Bender, A. Rohlmann and A. Halder, “Loading of the Knee Joint During Activities of Daily Living Measured in vivo in Five Subjects,” Journal of Biomechanics, vol. 43, pp. 2164–2173, 2010.

[2] F. Ginanneschi, F. Filippou, P. Milani, A. Biasella and A. Rossi, “Ulnar Nerve Compression Neuropathy at Guyon's Canal Caused by Crutch Walking: Case Report With Ultrasonographic Nerve Imaging,” Archives of Physical Medicine and Rehabilitation, vol. 90, no. 3, pp. 522– 524, 2009.

[3] M. Higuchi, M. Ogata, S. Sato and Y. Takeda, “Development of a Walking Assist Machine Using Crutches,” Journal of Mechanical Science and Technology, vol. 24, pp. 245–248, 2010.

[4] Y. Mori, T. Taniguchi, K. Inoue, Y. Fukuora and N. Shiroma, “Development of a Standing Style Transfer System ABLE with Novel Crutches for a Person with Disabled Lower Limbs,” Journal of Systems Design and Dynamics, vol. 5, no. 1, pp. 83– 93, 2011.

[5] www.Anatomy–Resources.com [6] J. Perry, Gait Analysis, SLACK Inc., NJ, 1992.

[7] P. Adamczyk, S. Collins and A. Kuo, “The Advantages of a Rolling Foot in Human Walking,” The Journal of Experimental Biology, vol. 209, pp. 3953–3963, 2006.

[8] D. Tesar and J. W. Sparks, “The Generalized Concept of Five Multiply Separated Positions in Coplanar Motion”, J. Mechanisms, 1968, 3(1), 25-33.

[9] H. J, Dowler, J. Duffy, and D. Tesar, “A Generalised Study of Four and Five Multiply Separated Positions in Spherical Kinematics—II”, Mech. Mach. Theory, 1978, 13:409-435.

[10] Chen P. and Roth B., “Design Equations for the Finitely and Infinitesimally Separated Position Synthesis of Binary Links and Combined Link Chains”, ASME Journal of Engineering for Industry, 1969, Vol. 91: 209-219.

[11] Hanafusa H., and Asada H., “Stable Prehension by a Robot Hand with Elastic Fingers”, Proc. 7-th Int. Symp. Ind. Robots, 1977, pp. 384-389.

(a) Pictures of experiments (left: without crutch, right: with crutch) 500

without crutch with crutch

400

z position (mm)

300

200

100

0

-100

-200

-300

0

100

200

300

400

500

600

700

800

x position (mm)

(b) Foot trajectory compairment Fig. 9. Treadmill walking experiments and their results

VII. SUMMARY AND FUTURE W ORKS In this paper, the disadvantages of conventional designs, underarm and forearm crutches, are studied.

In order to overcome them, required design objectives and specifications are setup. Closed–loop linkages design is considered as a preliminary design. From the preliminary design tests, the final design was determined as a serial chain inspired by McGeer’s [21] passive walker. To fulfill the two main design objectives,

19

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 [12] E. Rimon and Burdick J., “A Configuration Space Analysis of Bodies in Contact - I. 1-st Order Mobility”, Mechanism and Machine Theory, 1995, Vol.30(6) : 897-912.

[13] E. Rimon and Burdick J., “A Configuration Space Analysis of Bodies in Contact - II. 2-nd Order Mobility”, Mechanism and Machine Theory, 1995, Vol. 30(6): 913-928.

[14] N. Robson and McCarthy J. M., “Kinematic Synthesis with Contact Direction and Curvature Constraints on the Workpiece”, ASME Int. Design Engineering Conference, 2007.

[15] S. Epstein, “Art, History, and the Crutch,” Ann Medical History, vol. 9, pp. 304–313, 1937.

[16] M. Emami and S. Jamali, “Investigation of Ergonomic Issues in Crutch Design and Present an Innovation,” in Proc. of APIEMS Asia Pacific Industrial Engineering & Management Systems Conference (APIEMS), pp. 2939–2943, 2009.

[17] S. Fisher and R. Patterson, “Energy Cost of Ambulation with Crutches,” Archives of Physical Medicine and Rehabilitation, vol. 62, no. 6, pp. 250–256, 1981.

[18] www.iwalk-free.com [19] D. Winter, Biomechanics and Motor Control of Human Movement, Wiley–Interscience Publication, 1990.

[20] McCarthy, J.M., Geometric Design of Linkages, SpringerVerlag, 2000, New York.

[21] T. McGeer, “Passive Walking with Knees,” in Proc. of the IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 1640–1645, 1990.

20

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

Optimization of machining parameters and tool selection in 2.5D milling using Genetic Algorithm Arun Kumar Gupta#1, Pankaj Chandna2, Puneet Tandon3 1,2

Department of Mechanical Engineering, National Institute of Technology Kurukshetra, Haryana, INDIA 3 Mechanical Engineering and Design Discipline, PDPM Indian Institute of Information Technology, Design and Manufacturing Jabalpur, Madhya Pradesh, INDIA #

(Corresponding Author E-mail: arun_gupta@engineer.com)

Material removal rate cm3/min Diameter in (mm) Tool path length (mm) Cutting speed in (mm/min) Feed per tooth in (mm per tooth) Number of cutting edges on cutting tool C,C1,C2,C3 Constants a Chip cross-section Area mm2 Cu Unit Cost in US$ Cv Cost per unit volume in (US$/cm3) Cmat Cost of work piece material in (US$) Ct Cost of tool in (US$) Ctc Tool changing cost per unit in (US$) c1 Labour cost in (US$/min) co Over head cost in (US$/min)

Abstract— Optimization of machining parameters for improving the machining efficiency is become important, when high capital cost NC machines have been employed for high precision and efficient machining. The strategy is to minimize the production time and cost by optimizing feed per tooth, speed, width of cut, depth of cut and tool diameter by satisfying all the constraints such as maximum machine power, maximum cutting force, maximum machining speed, feed rate, tool life and required surface roughness. The optimal End milling cutter diameter and radial depth of cut (step over) are also the key issues for minimization of total production cost. Therefore, in this paper an attempt has been made to include all major parameters such as feed per tooth, speed, width of cut (Step-over) and depth of cut along with diameter of tool for minimising the time and production cost during 2.5 D milling. Hence, a mathematical model has been developed and Genetic Algorithm (GA) has been proposed to solve the problem.

Optimal values of machining parameters have been calculated for benchmark problems and compared with handbook recommendations. It has been found that approximately 13% of production cost can be reduced by choosing optimal cutter diameter and width of cut.

Besides this 50% reduction in cost per unit volume and 61% increment in material removal rate has also been reported by selecting optimal cutting parameters over the handbook recommendations.

MRR d L V f z

I. INTRODUCTION Milling is one of the most common metal removing processes in manufacturing. The application of milling has been increased with the introduction of high speed machining (HSM) and improvement in the milling equipment. In today’s competitive environment, optimizing machining parameters for increase in the total profit rate and quality product are the vital issue.

Generally, the handbook references or human experiences have been used to select the machining parameters. The productive time, cost and quality of production is highly influenced by machining parameters such as cutting speed, feed rate, width of cut (step over), depth of cut and tool diameter. Besides these parameters, the 2.5 D milling operation has also affected by the capability of machine tool, tool material and type of coolant used to a great extent.

Keywords: Optimization of Machining Parameters, 2.5D Milling, End Milling, Genetic Algorithm NOMENCLATURE

ttc Tool changing time per component (min) ts Setup Time in (min) tm Machining Time in (min) tnp Time spent during Non productive movement T Tool Life in (min)

Tu

Higher chip thickness is the indication of high material removal rate (MRR). But the chip thickness is

Unit time in (min)

21

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 dependent upon feed per tooth, cutting speed and number of cutting flutes [7]. In general practice the feed per tooth is maintained at maximum whereas cutting speed is maintained at minimum to increase MRR under the limitation prescribed for the particular tool.

The surface finish is solely depends upon feed per tooth and tool geometry. Therefore, the required surface finish is the major constraint for the value of feed per tooth [10]. Beside above the MRR can also be improved by increasing feed per tooth and by maintaining cutting speed at a certain level keeping in view of tool life. The tool life is highly affected by cutting speed [18]. It is also affected by the feed per tooth and depth of cut. Lower tool life might be the cause of higher cost of production. But higher feed rate and cutting speed is responsible for higher MRR and lower tool life [8], [17]. Therefore, the machining parameter selection is the compromise between tool life and cost of production. Tondon et al, [13] have optimized machining parameters feed and cutting speed for NC end milling operations by Particle swarm evolutionary computation technique. Kiliq et al, [3] considered a computer-aided graphical technique for optimization of machining parameters that is cutting speed and feed rate under consideration of machine power, surface finish and tool life.

In 2.5D milling, the production time and cost during roughing is also influenced by width of cut. Higher width of cut leads to higher MRR, but cutting forces become predominant and might be the cause of tool deflection with lesser heat dissipation time per tooth. Hence, there is rise in temperature of cutting edges, which leads to built-up edges. Therefore, in rough cut usually low width of cut with higher depth of cut has been considered.

The range of width of cut depends upon type of operation performed. Very few researchers have considered width of cut along with other machining parameters to optimize the problem. Hinduja et al, [11] optimized the problem by choosing appropriate ratio of width of cut to tool diameter for machining 2.5 D milling.

Gopalsamy et al, [18] have conducted some experiments to optimize width of cut along with other machining parameters by Taguchi method. Ibraheem et al, [16] and Saffar et al, [17] have optimized machining parameters in prospective of cutting forces on end milling cutter by using Genetic algorithm.

The machining cost might also be reduced by selection of optimal tool diameter. A lot of work has been reported for selection of appropriate diameter of tool. Lee and Chang, [5] calculated the largest possible diameter circle that can be inscribed the whole 2.5D pocket. They concluded that the diameter of cutting tool should approach to this largest possible diameter circle.

Bala and chang, [2] optimized tool path length by selecting multiple tool diameter selection. Hinduja et al, [11] studied other cutting parameters along with diameter of tool and obtain optimum width of cut to tool diameter ratio. Ding et al, [15] developed an approach to identify the feasible regions for the candidate cutters without tool path generation. The machining time for different cutter combinations has been estimated based on the areas of the feasible regions and the cutter feed rates.

MRR can also be improved at same cutting speed and feed per tooth by increasing the number of cutting flutes. But cutting flutes at smaller pitch results in material clogging and hence rubbing might be occurred rather than cutting. Beside these parameters, the cost, time and quality of production are highly sensitive to depth of cut and number of passes [9], [10]. Therefore, the selection of optimal depth of cut is great concerned before a part is put in to production.

A lot of work has been done to optimize the cutting parameters such as cutting speed, feed per tooth and depth of cut. Red and Bidhendi, [7] optimized the machining parameters under the consideration of surface finish, power and cutting force for milling operations. Similar work has been done by Ahmad et al, [14] to optimize these machining parameters for end milling operation by soap based genetic algorithm.

Whereas, Dereli et al, [12] have also used Genetic Algorithm to solve the similar problem. Yang et al, [22] optimized the feed rate, cutting speed and depth of cut for multi-pass face milling operation with constraints of machine power, cutting force, machining speed and surface roughness using particle swarm intelligence technique.

The selection of optimum tool and cutting parameters is an important activity in process planning of 2.5 D milling and is responsible to a great extent for production cost/time. The cost has been minimized by optimizing the parameters such as speed, feed per tooth and depth of cut in past practices ([1], [4], [8], [9], [13], [14], [19], [20], [22]), Whereas the selection of optimum tool and width of cut have been studied separately ([2], [5], [6], [11], [15], [16], [17]). But, the smaller diameter tool possesses longer tool path length as compeered to larger diameter tool [6], [11]. Higher forces are predominant on larger diameter tool due to high linear velocity at same RPM. To minimize theses forces, low spindle speed is assigned, which affect the MRR and productivity. Hence, from review of literature,

22

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 it can be concluded that both the approaches are interrelated. Therefore, the present study concurrently considered the optimization of cutting parameters and selection of optimum tool diameter along with optimum value of width of cut using Genetic Algorithm.

The cost of 2.5 D milling involves material cost, cost of machining during non productive movement of tool, productive cost and tool cost.

(04)

II. PROBLEM FORMULATION Optimum values of feed per tooth, speed, depth of cut, tool diameter and width of cut have been calculated for minimum cost in minimum time and simultaneously satisfying the constrained such as maximum power, maximum force, required surface finish, available speed and feed. MRR and cost per unit volume have been calculated for comparison of objectives.

(05) Substituting the equation 2a and 2b in equation 05:

Unit Time Unit time comprises of productive and non productive time for the case of single tool 2.5 D milling operation.

Further productive time can also be segmentized as setup time, time spent during machining and tool changing time. Since, damaged tool needs to be replaced in a single tool milling operation.

(1)

(06) The tool gets retracted and repositioned several times in multi pocket jobs during rough machining which consumes 15 to 30% of total machining time depending on the complexity of job [21]. The problem considered in the paper has lesser number of retraction points.

Therefore, the time spent during rapid movement or non productive time of tool is assumed to be negligible.

Beside this cost of material, setup cost and tool changing costs are not influencing the machining and therefore excluded from equation 6.

(2) The machining time and tool life can be calculated as.

(2a)

(2b)

(07)

Where n is 0.15 for Carbide tools, C is 100.05 for carbide tools, Q is the contact proportion of cutting edge per revaluation with work piece, G is Slenderness ratio , A is Chip cross-section area A , exponent g = 0.14 and exponent w = 0.28. By substituting the equation 2a and 2b in equation 2:

A. Constraint Function There are certain limitations of cutting and machine tool such as maximum spindle speed, feed rate, maximum power, maximum machining force and required surface finish. To avoid built up edges and smooth running of cutting tool, manufacturers have been provided a definite range of speed, feed rate and depth of cut. Therefore, the parameters values have to be optimized in the specified range for satisfying the constraint regarding the available spindle speed, maximum power, maximum force and required surface finish as shown in the table 1.

(3) Where constant C1, C2 and C3 can be calculated as (3a) (3b)

Unit Cost

23

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 TABLE 1 Constraints

S. No.

Name of Constraint

Step 3: The algorithm then creates a sequence of new population. At each step, the algorithm uses individuals in current generation to create the next population. To create the new population, the algorithm performs the following steps:

Constraint

1

Machining Speed in m/min

50 < V < 150

2

Feed per tooth in mm

0.05 < f < 0.2

3

Machining Force in N

F < 300

4

Machining Power in KW

P < 5.5

5

Surface Finish in mm

Sf < 0.5

a. Scores each member of the current population by computing fitness i.e. minimizing. b. Select members, called parents, based on their fitness. c. Some of the individuals in the current population that have lesser fitness are chosen as elite. These elite individuals are considered to the next population. d. Produces offspring from the parents. Offspring are produced either by combining the vector entries of a pair of parentsâ&#x20AC;&#x201D;crossover or by making random changes to a single parentâ&#x20AC;&#x201D; mutation. e. Replaces the current population with the children to form the next generation.

Step 4: The algorithm stops after running specified number of generations which is considered as stopping limit.

III. METHODOLOGY In the present 2.5 D milling problem, five parameters i.e tool diameter, width of cut, speed, depth of cut and feed rate have been considered. The parameters have been calculated for optimized objective function with all the constrained. A set of five possible tool diameters have been taken for a specified job, whereas the width of cut has been varied from 0.2 D to 0.7 D. To optimize the problem, Genetic Algorithm has been proposed.

The unit cost and unit time have been taken simultaneously by converting the two objectives into a single objective optimization problem by considering the weighted sum of these objectives.

Step 5: The final optimum solution in binary string, given by G.A. is decode in to decimal number values.

Step 6: Repeat the above five steps for each possible depth of cut.

B. Genetic Algorithm The Genetic Algorithm (GA) uses probabilistic selection as a basis for evolving a population of problem solutions. An initial population is created and subsequent generations are generated according to a pre-specified breeding and mutation methods inspired by nature. GA generates initial population randomly according to constrained mentioned. Best solution is selected from the population as evaluated by fitness function. This best solution is termed as elite solution.

The new population is again passed from the same process and the process is repeated to calculate best solution. The process remains continue till the stopping limit has not been achieved.

The detail of each step Genetic Algorithm is explained below:

C. Coding The data processed by the proposed GA considered a binary string of 27 element length. Out of which, first twelve elements of binary string represent speed and next ten elements represent feed per tooth, whereas remaining five elements represent index of width of cut and diameter of tool. All solutions are converted into single binary codes after generation of possible parameters combination of specified population size.

The fractional value of any parameter is converted into a whole number by multiplying and dividing the values by a multiplying factor and converted into a binary string.

D. Parameters for proposed Genetic Algorithm The considered parameters of proposed GA are shown in table 2.

Step1: Generate random possible combination as per population size of three parameters values (speed, feed, & width of cut) and tool diameter index.

Step2: Convert each combination into a single binary string.

24

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

TABLE 2 PARAMETERS FOR GENETIC ALGORITHM

Parameter Population Size Crossover Function Mutation Function Elite Count Crossover Fraction Mutation Fraction Stopping Condition

Fig. 1 shows the three islands problem in which the whole pocket has been machined with maximum of 40 mm diameter tool. Therefore, five end-mill cutters with best possible diameters 25, 32, 30, 35, 40 has been considered for this problem by varying width of cut from 0.2d to 0.7d. The tool path lengths for these tools have been calculated on different width of cut as shown in figure 2.

Value 100 Partially Matched Crossover (PMX) Reciprocal Exchange (RX) 2 0.85 0.15 100 Generations

A. The tool path length calculated for 40 mm diameter tool is found to be smallest for each width of cut as clearly shown in fig. 2. Whereas the largest tool path length has been observed for 25mm diameter tool. It is also evident that the tool path length reduces with increase in width of cut. The curves drawn for each diameter of tool get converged with increase in width of cut. Therefore, it can be concluded that length of path is less sensitive with higher value of width of cut.

B. Unit cost and unit time is the function of tool path length and also affected by combination of speed, feed and width of cut. Therefore, a Genetic Algorithm has been proposed for selecting optimum parameters at each possible diameter of tool. The observation has been shown in table 4 and the corresponding objective function values are shown in table 5. From comparing, it has been found that optimum values obtained by Genetic Algorithm are better than Catalogue values for all possible diameters of tool. The results also shows that 25 diameter tool has minimum machining cost per unit -2 3 volume of 2.83X10 US$/cm , and maximum MRR 3 of 74.57 cm /min. The 40 mm diameter tool also gives comparable optimal results.

IV. RESULTS AND DISCUSSIONS A pocket with three islands (benchmark problem) has been considered in the present work as shown in fig. 2 [11]. It is machined by 10 mm depth using CNC milling Machine. The desirable machining parameters and optimum tool diameter has been calculated using GA. A CNC machine with 5.5 KW of maximum machine power and 90% motor efficiency has been considered.

Unalloyed steel grade 1075 with hardness of 225 HB has been considered which is generally used in most of mechanical components, Dies and moulds etc. The various values considered have been given in table 3.

Fig. 1. A pocket with three islands [11] TABLE 3 SPECIFICATIONS AND CONSTRAINTS

Specification and Constraint co c1 Pmax Fmax Work Piece Material Material Hardness Machine Efficiency Tool Material amax amin Specification and Constraint

Value 1.45 US$ 0.45 US$ 5.5 KW 300N SAE 1075 225 BHN 90% Carbide 10 mm 1 mm Value

Fig. 2. Tool Path Length at all possible width of cut and diameters of tool.

25

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 TABLE 4 COMPARISON OF CATALOGUE PARAMETERS WITH THE OPTIMUM PARAMETERS

Diameter of tool 25 30 32 35 40

Catalogue value V fz B 100 0.15 0.3d 100 0.15 0.3d 120 0.15 0.2d 120 0.15 0.2d 120 0.15 0.2d

Optimum value V fz 76.6 0.191 71.2 0.189 102.1 0.191 78.9 0.184 77.4 0.188

b 0.4d 0.4d 0.3d 0.4d 0.4d

TABLE 5 COMPARISON OF OBJECTIVES

Catalogue value

Diameter

-3

Optimum value -2

of tool

Cv in 10 3 US$/cm

MRR 3 cm /min

Cv in 10 3 US$/cm

MRR 3 cm /min

25

39.5

57.33

2.83

74.57

30

39.5

57.35

3.08

68.58

32

56.0

45.85

3.28

74.52

35

55.9

45.86

2.87

73.98

40

56.0

45.84

2.84

74.12

Fig.3. Comparison of Cost per Unit Volume and MRR at different depth of Cut

E. The variation of cost per unit volume and MRR with depth of cut has been shown in fig. 3. From the analysis, it has been found that MRR increases and machining cost per unit volume decreases with increase in depth of cut. The average value of force estimated for optimum parameters on tool is 98% of maximum permissible value, which shows the machining at maximum force. All the parameters have been calculated within the specified range. Therefore, machine should be run at maximum depth of cut for achieving maximum optimum machining cost and MRR.

C. From table 6. It has been observed that feed rate is being varied from 0.184 mm/tooth to 0.191 mm/tooth as depth of cut increases from 1 mm to 10 mm. whereas, speed decreased appreciably from 150 m/min to 76.5 m/min. This shows that minimum cost and time can be achieved if the feed per tooth calculated by GA is nearer to maximum value of feed rate (0.2mm/tooth). Also, due to constraints of maximum power and force the value of speed decreasing gradually with depth of cut.

V. CONCLUSIONS A Genetic Algorithm has been applied for optimizing machining parameters during 2.5 D milling operation for minimizing machining cost and machining time.

Machining parameter such as depth of cut, width of cut, spindle speed and feed per tooth have been considered along with the selection of optimal tool diameter. The optimization is subjected to satisfying certain constraints such as maximum available power, maximum cutting force, maximum spindle speed, feed per tooth and required surface finish.

The optimum machining parameters obtained by GA has made a significant increase in machining efficiency over the tool Catalog recommendations. The table 7 shows that the cost per unit volume is decreased by approximately 50% and MRR is improved by 61%.

TABLE 6 COMPARISON OF OPTIMUM PARAMETERS OBTAINED BY GA AT VARYING DEPTH OF CUT

Depth of Cut in mm

D

1 2 3 4 5 6 7 8 9 10

25 25 25 40 40 40 40 40 40 40

V mm/mi n 150.0 150.0 145.9 146.1 144.1 127.7 113.7 108.5 86.2 76.5

Optimum value F b Cost/c mm mm ut US$ 0.20 0.7d 4.91 0.20 0.7d 5.42 0.191 0.7d 5.91 0.199 0.4d 6.27 0.191 0.4d 6.62 0.191 0.4d 6.67 0.182 0.4d 7.05 0.193 0.3d 7.29 0.188 0.3d 7.84 0.191 0.4d 8.37

Time/ cut min.

1.88 1.88 2.02 1.99 2.11 2.37 2.80 3.17 3.58 3.97

TABLE 7 COMPARISON OF OPTIMUM RESULTS

Cv in US$/cm Catalog value Optimum values Improvement over catalog

D. The value of width of cut is also decreased from 17.5mm (0.7d, 25mm) to 16mm (0.4d, 25mm) with increase in depth of cut.

56.0 28.4 Decrease by 50.71%

3

MRR in 3 cm 45.84 74.12 Increase by 61.84%

The GA has been applied on three island job with different diameter of tools and it has been found that machining cost per unit volume and machining time varies depends on tool diameter. From table 8, It has

26

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011 been shown that 25 mm and 40mm diameter of tools, produces minimum machining cost per unit volume and maximum material removal rate.

Therefore, it can be concluded that combination of optimum diameter of tool and width of cut significantly improve the machining efficiency. It has also been found that maximum depth of cut under the permissible range provides optimum cost per unit volume and MRR.

[9]

[10]

[11]

[12] TABLE 8 COMPARISON OF MRR AND COST PER UNIT VOLUME OBTAINED AT DIFFERENT POSSIBLE DIAMETERS

D 25 30 32 35 40

MRR in 3 cm 74.57 68.58 74.52 73.98 74.12

-2

Cv in 10 3 US$/cm 2.83 3.08 3.28 2.87 2.84

Decrease in MRR -------8.02% 0.07% 0.78% 0.60%

[13]

Increase in Cv -------8.23% 13.84% 1.31% 0.49%

[14]

[15]

[16]

References [1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[17]

V. A. Ostafiev, A. V. Globa, L. S. Globa, “Integrated End Milling Optimization Development” Annals of the CIRP Vol. 33, No.1, 1984.

M. Bala, and T. C. Chang, “Automatic cutter selection and optimal path generation for prismatic parts, International journal of Production Research, vol. 29, No. 11, pp 2163-2176, 1991.

S.E. Killic, Can Cogun and D. Tayyar Sen “A computer-aided graphical technique for the optimization of machining conditions”, computers in Industry, vol. 22, pp. 319-326, 1993.

S. Engin Kiliq, Can Cogun and D. Tayyar Sen, “A computeraided graphical technique for optimization of machining conditions”, Computers in Industry vol. 22, pp. 319-326, 1993.

Y. S. Lee, T. C. Chang, “Application of computational geometry in optimization 2.5D and 3D NC surface machining”, computers in industries, vol. 26, No. 1, pp. 173-187, 1995.

D. Veeramani, Y.S Gau. “Selection of an optimal set of cuttingtools for a general triangular pocket”, International Journal of Production Research, vol. 35, No. 9, pp. 2621-2638, 1997.

M. Tolouei-Rad and I. M. Bidhendi, “On the optimization of machining parameters for Milling operations on the optimization of machining parameters for Milling operations” International Journal Mechanical Tools Manufacturing, vol. 37, No. I, pp. 1-16, 1997.

Ihsan Sonmez, Adil Baykasoglu, Turkay Dereli, I. Huseyin Fılız “Dynamic optimization of multi pass milling operations via

[18]

[19]

[20]

[21]

[22]

27

geometric programming”, International Journal of Machine Tools & Manufacture vol. 39, pp. 297–320, 1999.

TuÈ rkay Dereli, I. HuÈ seyin Filiz, “Optimisation of process planning functions by genetic algorithms” Computers & Industrial Engineering, vol. 36, pp. 281-308, 1999.

John L. Yang & Joseph C. Chen, “A Systematic Approach for Identifying Optimum Surface Roughness Performance in EndMilling Operations”, Journal of Industrial Technology, vol 17, No.2, pp 1-8, 2001.

S Hinduja, A Roaydi, P Philimis, “Determination of optimum cutter diameter for machining 2 1/2-O pockets” International Journal of machine tool & manufacture, vol. 41, No. 5, pp. 687702, 2001.

T. Dereli, I. H. Filiz and A. Baykasoglu, “Optimizing cutting parameters in process planning of prismatic parts by using genetic algorithms”, International journal of production research, Vol. 39, No. 15, pp. 3303- 3328, 2001.

V. Tandon , H. El-Mounayri, H. Kishawy, “NC end milling optimization using evolutionary computation”, International Journal of Machine Tools & Manufacture, vol 42, No. 5, pp 595605, 2002.

Nafis Ahmad, Tomohisa Tanaka, and Yoshio Saito, “Optimization of cutting parameters for end milling Operation by soap based genetic algorithm”, Proceedings of the International Conference on Mechanical Engineering, pp.28-30, 2005.

X. M. Ding, Y. Q. Lu and P. L. Liu, “Optimal cutter selection for complex mould machining”, SIM Tech. technical reports, vol. 6, No.1, pp.84-88, 2005.

Abbas Fadhe Ibraheem, Saad Kareem Shather, “Prediction of Cutting Forces by using Machine Parameters in end Milling Process”, Eng. &Tech.,vol.26,.No.11, 2008.

R. Jalili Saffar, M.R. Razfar, A.H. Salimi and M.M. Khani, “Optimization of Machining Parameters to Minimize Tool Deflection in the End Milling Operation Using Genetic Algorithm”, World Applied Sciences Journal Vol. 6, No. 1, pp. 64-69, 2009.

Bala murugan Gopalsamy, Biswanath Mondal, Sukamal Ghosh, “Taguchi method and ANOVA an approach for process parameter optimization of hard machining while machining hardened steel”, Journal of scientific & Industrial research, vol.

68, pp. 686-695, 2009.

R.E. Zapata-Ramos and T.L. Schmitz, M. Traverso, “Value of information and experimentation in milling profit optimization”, International journal of Mechatronics and Manufacturing Systems, Vol. 2, No. 5/6, pp. 580-599, 2009.

Sanjit Moshat, Saurav Datta, Asish Bandyopadhyay and Pradip Kumar Pal, “Optimization of CNC end milling process parameters using PCA-based Taguchi method”, International Journal of Engineering, Science and Technology, vol. 2, No. 1, pp. 92-102, 2010.

A.K. Gupta, P. Chandna, P. Tandon, “Hybrid genetic algorithm for minimizing non productive machining time during 2.5 D milling”, International Journal of Engineering, Science and Technology, vol. 3, No. 1, pp. 183-190, 2011.

Wen-an Yang & Yu Guo & Wenhe Liao, “Multi-objective optimization of multi-pass face milling using particle swarm intelligence”, International Journal of Advance Manufacturing & Technology (DOI), vol. 56 No. 5-8, pp. 3187-88, 2011.

INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY & CREATIVE ENGINEERING (ISSN:2045-8711) VOL.1 NO.8 AUGUST 2011

28