Future Steel Forum 2018

� INDUSTRY 4.0 ARTICLES � SPEAKER BIOGRAPHIES � EXHIBITOR PROFILES � FULL PROGRAMME � FLOOR PLAN � A GOOD VIBE Future Steel Forum Supplement June 2018

cover.indd 1

22/05/2018 08:33:38

LEADING AUTOMATION EXPERTISE In times of increasing complexity for plants and functionality,

you

want

modular

yet

holistic

automation approaches with simple operating philosophies. To obtain highest-quality products. To achieve the best operating results using the most efficient processes. We listen. We learn. We deliver.

X-Pact® world-leading automation expertise makes your plant fit for the future – from energy supply and distribution through drive technology, instruments, and automation to production planning, and last but not the least – digitalization. Let’s add value along the entire value chain, together.

Leading partner in the world of metals

www.sms-group.com/x-pact

Automation Expertise_A4_e.indd 1

20.11.17 13:17

Contents � INDUSTRY 4.0 ARTICLES � SPEAKER BIOGRAPHIES � EXHIBITOR PROFILES � FULL PROGRAMME � FLOOR PLAN � A GOOD VIBE

2

Welcome by Matthew Moggridge

4

Future Steel Forum Conference Programme

8

Speaker Biographies

Future Steel Forum Supplement May 2018

cover.indd 1

18/05/2018 11:35:30

EDITORIAL/PRODUCTION 20561-18 LV Titel Steel Times DIJ LAY02 DRUCK.indd 1

20 Exhibitor Profiles

30.04.18 10:37

Editor / Programme Director Matthew Moggridge +44 1737 855151 matthewmoggridge@quartzltd.com Production Editor Annie Baker Advertisement Production Martin Lawrence

SALES International Sales Manager Paul Rossage +44 1737 855116 paulrossage@quartzltd.com Sales Director Ken Clark +44 1737 855117 kenclark@quartzltd.com

CORPORATE Managing Director Steve Diprose CEO Paul Michael

28 Floorplan 30 The digital transformation of steel, PwC 32 Business Ethics 4.0@Steel Industry, Primetals USA 36 AI – translating theory into reality, Steel Hub London 40 Data mining, modelling and smart manufacturing, ArcelorMittal 46 An AI-enabled future for metals? Accenture 50 Digital transformation of a steel company - first steps, NLMK 54 Open innovation and SPD, University of Liverpool, UK 60 Hi-tech steel production planning, SMS Group, Germany 68 How tech catalyses disruptive change, by Mick Steeper 72 Asset health, Endress + Hauser Messtechnik GmbH

Published by: Quartz Business Media Ltd Quart House, 20 Clarendon Road Redhill, Surry RH1 1QX, UK +44 1737 855000 www.steeltimesint.com © Quartz Business Media, 2018

contents fsf.indd 3

76 Through-process optimisation, Primetals Technologies 84 Is your plant digitally mature?, Danieli Automation 86 New economics of steel, Dastur & Co 90 Innovations

FUTURE STEEL FORUM

1 18/05/2018 11:53:28

Welcome “Welcome back, my friends, to the show that never ends, we’re so glad you could attend...” and I’ll stop there with the Emerson, Lake & Palmer lyrics, from the excellent Brain Salad Surgery album, although I will say that I’ve been making rock music analogies about the Future Steel Forum ever since the inaugural event closed on15 June 2017.

In short, if you’re reading this, you are now attending the Future Steel Forum 2018, the ‘difficult second album’ if you will; and I must say that this year’s programme is a classic in the making. I have brought together some of the biggest names in the business to discuss the principles of Industry 4.0. This year’s conference programme is jam-packed with leading global steelmakers: ArcelorMittal, Liberty House Group, Tata Steel and US Steel; and throughout the two-day programme, digitalisation experts will cover artificial intelligence, cyber security and consider ‘the digital dilemma’ facing the steel industry.

We have two excellent discussion panels, one entitled Great Expectations (bringing together plant builders and steelmakers) and the other looking at engineering education and how to qualify the Industry 4.0 workforce – both unmissable in my opinion.

The Future Steel Forum is a steel industry-specific conference that is focused entirely on the subject of Industry 4.0 and it’s relationship to steel production. My aim, as always, has been to develop a top-notch, uncompromising programme and I hope you will be pleased with the end result. Now, how about a US tour?

Matthew Moggridge, Programme Director

2 welcome.indd 2

FUTURE STEEL FORUM

Steel Times International

18/05/2018 12:57:22

METALS ORCHESTRA At Primetals Technologies, we don’t just build steel plants.

Instead, we ensure all your equipment is perfectly orchestrated for maximum performance and reliability. We like to continually innovate, to facilitate progress, and to pioneer new solutions.

We love steel, and we will change the way you produce it. primetals.com

Ad-297x210-FSF-magazine-Future-Steel-Forum-Warschau.indd 1

17.04.2018 07:54:00

CONFERENCE PROGRAMME

DAY ONE – WEDNESDAY 6 JUNE

REGISTRATION AND WELCOME 08:45 Welcome to the Future Steel Forum

Matthew Moggridge, Editor, Steel Times International and Programme Director

08:50:

Keynote: Future Steel: Industry 4.0 and more Dr Pinakin Chaubal, General Manager, ArcelorMittal Global R&D

INTRODUCTORY THEMES - CHAIRMAN KURT HERZOG, HEAD OF DEPARTMENT, PRIMETALS TECHNOLOGY INDUSTRIE 4.0 09:20:

Steel 4.0: Perceptions, current activities and expectations for Europe Dr. Marlene Arens, Senior Researcher, Fraunhofer Institute for Systems and Innovation Research

09:50:

Industry 4.0 for the Steel Industry - from a research point of view Dr Roger Andersson, Head of Research, SWEREA MEFOS, Sweden

10:20:

How Technology Catalyses Disruptive Change in the Steel Industry Mick Steeper, Former Chair, Iron & Steel Society, IoM3

10:50-11:20: Coffee Break and Exhibition Time

4

11:20:

The Digital Dilemma Facing Steel Dr Andrew Zoryk, Managing Director, Metals Practice, Accenture

11:50:

Data is the new currency in the steel industry: Hear how companies are cashing in Stefan Koch,Global Leads for Metals, SAP SE

12:20:

Lunch and Exhibition Time

13:30:

Keynote - Business Models - A Steel Industry Perspective Jayanta Banerjee, CIO, Tata Steel

FUTURE STEEL FORUM

conference programme.indd 2

Steel Times International

18/05/2018 13:00:44

BUSINESS MODELS - CHAIRMAN, MICK STEEPER, CHAIR, IRON & STEEL SOCIETY, IOM3 14:00: MARVEL 2.0 - Making Analytics Real, Valuable, Efficient and Logical Sarajit Jha, Chief of Digital Value Accelaration, Tata Steel

14:30: Beyond classical methods: Big Data, Machine Learning and Stochastic

Simulation for investment decisions in steel operations.

Diego Diaz, Global R&D Senior Specialist, ArcelorMittal

15:00: Steel 4.0 - Industry 4.0 and the Big Data Challenge:

Development of a Fully Integrated System for Real-Time Predictions

of Microstructure Evolution During Hot Rolling of Steel Bars

Dr. rer. Nat. Zeljko Cancarevic, Head of Innovation, GeorgsmarienhĂźtte Holding GmbH

15:00: Industry 4.0 - How the Internet of Things can improve steel plant safety

Jan Petko, General Manager, Process Technology Excellence, US Steel Kosice Peter Rusko, Global Automotive, IBM

15:30

Coffee Break and Exhibition Time

16:00

First Step in Digital Transformation Kirill Sukovykh, NLMK-SAP Co-Innovation Lab Lead

16:30: Closing Keynote: The Role of Industry 4.0 in Liberty House Group’s strategical green steel vision Eric Vitse, Chief Technology Officer, Liberty House Group

17:00: PANEL DISCUSSION: GREAT EXPECTATIONS Chairman Rizwan Janjua, Head of Technology, World Steel Association Panellists include: Kristiaan Van Teutem, Vice President, Steel Business Line, Fives Group Giovanni Bavestrelli, Digital Engineering Director, Tenova SpA Kurt Herzog, Head of Department, Primetals Technologies Prof. Dr.-Ing. Katja Windt, Member of the Managing Board, SMS group GmbH Jan Petko, General Manager, Process Technology Excellence, US Steel Kosice Dr Pinakin Chaubal, General Manager, ArcelorMittal Eric Vitse, Chief Technology Officer, Liberty House Group Marco Ometto, Executive Vice President, Danieli Automation The digital transformation is disruptive in nature. It holds promises of great returns and productivity, however: how does one justify investment in the face of uncertainties? This cannot be a leap of faith and rather requires clear understanding of facts, deliverables and limitations. The panel discussion brings together the two sides (steelmakers & OEMs) in order to debate the expectations and deliverables.

17:45: Conference closes.

18:30: Networking dinner sponsored by Fives Group

Steel Times International

conference programme.indd 3

FUTURE STEEL FORUM

5 18/05/2018 13:00:44

CONFERENCE PROGRAMME

DAY TWO – THURSDAY 7 JUNE REGISTRATION AND WELCOME 08:45: Keynote: Big River Steel - the world’s first ‘learning’ steel mill

Franck Adjogble, Chief Engineer, Process Control and Production Planning Systems, SMS group

ARTIFICIAL INTELLIGENCE FOR STEEL - CHAIRMAN, RAFFAEL BINDER, DIRECTOR, MARKETING, PSI METALS 09:15: Artificial Intelligence - translating the theory into reality

Emilio Riva, former member of the Executive Board of Riva Group, CEO & Founder of the Steel Hub, London & Valentina Colla PhD, Technical Research Manager, Scuola Superiore Sant’Anna

09:45: Towards an holistic AI strategy: Balancing vision with execution Jane Zavalishina, President & Co-Founder, Mechanica AI

10:15: Using AI modelling for predictive quality and routing

Heiko Wolf, Project Manager FutureLab, PSI

10:45: Coffee Break and Exhibition Time

11:15: Keynote: Open Innovation and Social Product

Eur Ing. Professor Dirk Schaefer, University of Liverpool

BUSINESS ORGANISATION - CHAIRMAN, DR LUC BONGAERTS, BUSINESS DEVELOPMENT MANAGER, OM PARTNERS 11:45: Integration of Industry 4.0 to Existing Steel Plant Facilities

Chris Oswin, Research Group Manager, Process Simulation, Materials Processing Institute

12:15: Simulation, Visualisation, and Data Analytics for Smart Steel Manufacturing

6

Dr. Chenn Q Zhou, Founding Director, Steel Manufacturing, Simulation and Visualisation Consortium, Purdue University Northwest, Indiana

FUTURE STEEL FORUM

conference programme.indd 4

Steel Times International

18/05/2018 13:00:44

12:45: Through-Process Optimization (TPO), a know-how based IT solution to

support best quality steel production and highest productivity at lowest cost

Wolfgang Oberaigner, Head of Through-Process Quality Control, Primetals Technologies

13:15:

Lunch and Exhibition Time

14:15: Keynote: Industry 4.0 and the Digital Future

Rizwan Janjua, Head of Technology, World Steel Association

14:45: New Technologies induce novel changes: How to implement Industry 4.0

in real-world applications

Dr. rer. nat. Marcus J. Neuer, VDEh-Betriebsforschungsintitut GmbH

15:15: Latest digital systems for process control, traceability and quality enhancement

in steelmaking

Kristiaan Van Teutem, Vice President, Steel Business Line, Fives Group

15:45: Smart Factories require Smart Planning

Dr Luc Bongaerts, Business Development Manager, OM Partners

16:00:

Coffee Break and Exhibition Time

16:30: Cyber Security: What can be learnt from other sectors?

Professor Chris Hankin, Co-Director, Institute of Security Science and Technology, Imperial College London

17:00: DISCUSSION PANEL: ENGINEERING EDUCATION AND QUALIFYING THE INDUSTRY 4.0 WORKFORCE. CHAIRMAN - DR. NILS NAUJOK, PARTNER, CONSULTING LEADER METALS INDUSTRIES EUROPE, MIDDLE EAST AND AFRICA, PWC STRATEGY Panelists include: Dr Chenn Qian Zhou, Founding Director, Steel Manufacturing, Simulation and Visualisation Consortium, Purdue University Northwest, Indiana Professor Dirk Schaefer, University of Liverpool Dr Joe Flynn, Assistant Professor in Manufacturing Engineering, University of Bath Dr Richard Curry, Director, Operations, Materials Processing Institute Henk Reimink, Director - Industry Excellence, World Steel Association

17:45: Closing remarks

Matthew Moggridge, Editor, Steel Times International, Programme Director

17:50:

Conference closes.

Steel Times International

conference programme.indd 5

FUTURE STEEL FORUM

7 18/05/2018 13:00:44

SPEAKER PROFILES

Matthew Moggridge, Editor, Steel Times International Matthew Moggridge has been editor of Steel Times International since January 2014 having previously edited Aluminium International Today, both published by the UK-based Quartz Business Media. During his time on both titles he has travelled extensively around the world interviewing and writing about leading figures in the metals industry and covering international steel and aluminium conferences. In addition to working

as a journalist in many different industrial sectors, he is also the driving force behind the development of the Future Steel Forum event, in particular the conference programme.

Matthew’s career as a business journalist has spanned many leading titles covering other industrial sectors including food processing, foodservice, bulk handling and transportation and computers.

Dr Pinakin Chaubal, General Manager at ArcelorMittal Dr. Chaubal obtained his PhD in Metallurgical Engineering from the University of Utah in the USA. He joined the steel industry in 1988 and has been involved in various aspects of process technology developments from raw materials to finishing processes. He is currently general manager of ArcelorMittal Global R&D responsible for worldwide

programmes in process technology development.

ArcelorMittal has placed a strong emphasis on leveraging developments in measurements, modeling, information technology to automate processes, and today is focused on further leveraging the various concepts under the umbrella of digitisation on an enterprise-wide basis.

Kurt Herzog, Head of Department, Primetals Technologies Industrie 4.0 Kurt Herzog studied control engineering at the Technical School Hollabrunn and industrial automation at the Technical University in Vienna. He recently attended the Linz Management Academy (LIMAK) where he studied engineering management. He joined Siemens VAI in 1997 as an E&A project engineer and has progressed through the company, becoming head

of process control systems for ironmaking, steelmaking and continuous casting and later becoming head of E&A product development, engineering and project execution for ironmaking, steelmaking and continuous casting. His present position is head of Industrie 4.0 (electronics and automation).

Dr Marlene Arens, Senior Researcher at Fraunhofer Institute for Systems and Innovation Research Since September 2009, Dr. Arens has been a research associate at the Fraunhofer Institute for Systems and Innovation Research ISI in the Competence Centre, Energy Technology and Energy Systems in Karlsruhe.

She finished her PhD at the University of Utrecht in January 2017 on “Technological change and

8 speakers bio.indd 2

FUTURE STEEL FORUM

industrial energy efficiency – Exploring the low-carbon transformation of the German steel industry”. Her research focuses on data analysis in the field of energy efficiency in industry, technology assessment and scenario based evaluation of future developments in energy consumption and CO2 emissions.

Steel Times International

18/05/2018 13:03:36

Strong in Performance.

Strong in efficiency.

Energy savings up to 30%!

BICOM_20111.01 1.08.2017

DIJ pumps from Leybold are the most advanced, most powerful, most power-saving and most reliable state-ofthe-art diffusion pumps available on the world market.

©

They excel with superior vacuum performance. Highest pumping speed and throughput, ideal for industrial high vacuum applications, like degassing of steel and special metals or heat treatment.

■

High and stable throughput at pressures >10-3 mbar

■

High pumping speed, stable forevacuum pressure

Most innovative heating concept providing an optimized energy transfer into the pump fluid ■

30% Energy

saving

■ Choice of ISO or ANSI flange variants for improved connectivity

Innovative ECO Energy control unit for energy savings up to 30% without vacuum performance loss ■

The new DIJ Oil Diffusion Pumps Best performance and utmost efficiency

Leybold GmbH Bonner Str. 498 · D-50968 Köln T +49 (0) 221-347-0 F +49 (0) 221-347-1250 info@leybold.com

www.leybold.com

Anz_DIJ_Diffusion_Coating_EN_210x297.indd 1

31.07.17 15:27

SPEAKER PROFILES

Dr Roger Andersson, Head of Research at SWEREA MEFOS Sweden Roger Andersson, born on 27 April 1966 in Boden, Sweden, finished his PhD at the Department of Production Technology in 2005 after his MSc degree in material science at Luleå University of Technology.

Andersson joined the steel research institute Swerea MEFOS in June 2016. He is in charge of a research

department and is part of the management team for the entire Swerea MEFOS. In his last position Andersson was CEO at Duroc Special Steel, a rerolling company for flat products of special grades.

Mick Steeper, Former Chair at Iron & Steel Society, IoM3 Mick Steeper has worked in metals processing technology companies, first in projects and latterly in R+D, for most of his career. His main expertise is in steel rolling, and straddles the boundaries of mechanical engineering, process control and metallurgy. Among his employers was Siemens Metals Technologies (2005-2014) throughout which time Industrie 4.0 was a significant and evolving research

field. Mick now works as an independent industrial technology consultant.

Professionally, Mick is a Fellow of the Institute of Materials, Minerals and Mining and has recently completed his tenure as chair of the Iron and Steel Society having been an active member of the Division’s Technical Committee and latterly its Board since the 1990s.

Dr Andrew Zoryk, Managing Director, Metals Practice at Accenture Dr. Zoryk is a managing director with Accenture’s global metals practice based in Vienna. He started his career with British Steel and Corus in the UK and since then has worked with many leading steel companies globally in areas of core enterprise processes, supply chain and manufacturing. Accenture is a global professional services company, providing a range of

services and solutions in strategy, consulting, digital, technology and operations. Accenture helps its clients improve their performance and create sustainable value for their stakeholders. With more than 435,000 people serving clients in more than 120 countries, Accenture drives innovation to improve the way the world works and lives.

Stefan Koch, Global Lead for Metals at SAP SE Stefan Koch is responsible for SAP solutions for the metal industry globally. In this role, he looks closely at all aspects of how technology can be applied to drive efficiency, innovation and growth across the metals industry. He is in frequent discussions with leading metals companies, industry user groups, technology implementation partners and independent

10 speakers bio.indd 4

FUTURE STEEL FORUM

software vendors. Presently Stefan is guiding a number of ongoing discussions with metals companies on how to drive the Digital Transformation in Metals and to identify the role of Industry 4.0 and IoT in this context. Stefan has been involved in the application of technology in manufacturing industries for more than 20 years.

Steel Times International

18/05/2018 13:03:37

Measure Your Profile In-line light section profile measurement systems for hot and cold steel applications.

Unique active and passive measures are taken to assure system operational temperatures are optimized, assuring precise measurements.

Comprehensive, sophisticated and well-proven software capabilities.

Industry 4.0 ready.

OPC UA interface.

6 – 7 June 2018 Stand A09

www.zumbach.com sales@zumbach.ch Zumbach_PM-SPS_A5.indd 1

02.05.18 08:11

SPEAKER PROFILES

Jayanta Banerjee, CIO at Tata Steel Jayanta Banerjee is the CIO at Tata Steel Limited, focusing on IT, Digital and Automation. He has been a business leader in TCS, focussed on the products and platforms business strategy and governance.

Jayanta has been the business leader and global head of TCS’ energy and resources business and has been with TCS for more than 24 years, leading businesses, sales, delivery and operations globally.

Jayanta has played key leadership roles in industry initiatives specific to life sciences, CPG and natural resources. He has spearheaded strategic entries into these verticals and also lead major transformational programmes. He has been instrumental in nurturing many strategic partnerships for TCS and has been an extensive traveler across the globe, taking a keen interest in ground breaking initiatives.

Sarajit Jha, Chief of Digital Value Acceleration at Tata Steel Sarajit Jha is chief digital value acceleration, Tata Steel, based out of Jamshedpur. Prior to this he was chief corporate strategist. Sarajit leads Tata Steel’s digital transformation initiatives and is focused on value creation, linkage to strategy, building a cross-company agile culture and experimentation with new business models.

Sarajit has 15 years of general management experience in the TAS – Leadership Development Cadre of the 100 bn USD+ Tata group in technology, international business development, brand strategy, turnaround strategy, business vertical creations, joint ventures, start-ups, merger & acquisitions (M&A), delivery centre and delivery model optimisation.

Diego Diaz, Global R&D Senior Specialist at ArcelorMittal Diego has been a researcher at the Business & technoEconomic Department (KiN) of ArcelorMittal Global R&D since its inception in 2004. This is a corporate division that provides service to ArcelorMittal globally. It is a multi-disciplinary team that brings advanced analytics and artificial intelligence to the business side of the company. Within this team and in collaboration with domain specialists in each field,

Diego has developed solutions across the value chain of the steel industry: line scheduling, internal and external logistics, yard management, strategy, purchasing, sales, and so on. His main focus is on mathematical optimisation, metaheuristics, and machine learning; and to a lesser degree on simulation, algorithmic game theory, and other fields of artificial intelligence.

Henk Reimink, Director - Industry Excellence at World Steel Association Henk Reimink is director of Industry Excellence at the World Steel Association (worldsteel) based in Brussels.

His role embraces corporate governance, strategic direction setting, international project management, safety and occupational health improvement and process safety management and change management engineering design, operational benchmarking on

12 speakers bio.indd 6

FUTURE STEEL FORUM

reliability, energy intensity, process yield, CO² emission intensity, contract management, iron and steel making, shaping and coating processes, steel in construction information, education and promotion, and staff development. He specialises in engineering and project management and maintenance strategy and execution.

Steel Times International

18/05/2018 13:03:40

Jan Petko, General Manager, Process Technology Excellence at US Steel Kosice Jan Petko has metallurgical education. He finished technical university in 1989 and started his career in the steel industry in 1996 as a shift manager with US Steel.

2005 director of power engineering. In 2008, he was appointed director of the coke plant and then general manager of reliability in 2010. He became general manager of process excellence. Jan is leading a Six

Sigman group in US Steel Košice s.r.o. with a strong focus on process excellence implementation. Jan has 20 years of experience in management, continuous improvement, operation and reliability. At this year’s Forum he will be speaking about how the Internet of Things (IoT) can improve steel plant safety and US Steel’s partnership with IBM.

Kirill Sukovykh, Co-Innovation Lab Lead at NLMK-SAP Kirill Sukovykh graduated from the Far Eastern Transport University, Khabarovsk, Russia, and has been head of NLMK-SAP’s Co-Innovation Lab since January 2017.

Prior to taking up the role Mr Sukovykh worked as a global account director at Rosatom, the State Atomic

Energy Corporation where he also worked as a master data management programme manager. Throughout his career he has participated in various large IT projects within international companies, such as Total, Accenture, InBev and ABB (2006 to 2012).

Dr Rizwan Janjua, Head of Technology at World Steel Dr. Janjua graduated from the University of Dalarna, Sweden, and got his PhD in material sciences from the TU Bergakademie Freiberg, Germany. He joined the steel industry in 2002 and has been involved in process technology, product application, energy, recycling, process yield and maintenance and reliability. He is head of technology at the World Steel

Association (worldsteel) and is responsible for leading activities in the field of technology, manufacturing excellence, expert groups and benchmarking systems.

Prior to this he led steeluniversity, an industry platform delivering education and training to current and future employees of steel companies and related businesses.

Franck Adjogble, Chief Engineer and Business Development Manager at SMS Group Franck Adjogble is on track to complete a doctorate degree in technology and innovation management at Fraunhofer Institute in Germany. His field of interest is in technology forecasting, on which his research work, “Towards a dynamic technology intelligence system”, is based. At the University of Siegen in Germany, Franck

Steel Times International

speakers bio.indd 7

studied computer science and completed a Master of Science degree. He has worked in the Department of Applied Informatics at the University of Siegen as assistant to the research director, where he participated actively in the project MODICAS (Efficient Integration of a 3D Digitising System and a Surgical Robot).

FUTURE STEEL FORUM

13 18/05/2018 13:03:45

SPEAKER PROFILES

Raffael Binder, Director, Marketing at PSI Metals Raffael Binder finished his innovation and product management studies with a diploma thesis about lead user methodology in the field of applied research.

He worked for the Austrian Institute of Technology in Vienna before joining PSI Metals as sales manager in 2009. After taking over management responsibility in 2012 as sales director for the Austria division, he took over the marketing department in 2015 and immediately positioned the business within the frame

of Industry 4.0 and digitalisation, bringing together relevant experts from PSI and outside the company.

Since 2017 Binder has been responsible for the PSImetals Academy, the company’s internal training programme, key accounts, sales support and product management. With this overall responsibility he is perfectly able to co-ordinate all activities around Industry 4.0 together with the company’s recently built FutureLab team.

Giovanni Bavestrelli, Digital Engineering Director at Tenova Giovanni Bavestrelli graduated from the Politecnico di Milano, Italy, with a master’s degree in software engineering, after attending high school in Johannesburg, South Africa. He joined Pomini Tenova in 1994, working on roll grinder automation, developing and commissioning software systems for running roll shops in steel plants. In 2000 he joined

Unisys, leading the development team for the Hermes editorial system, and returned to Pomini Tenova in 2004 as software engineering director. Since then he has worked for Pomini Tenova, leading the software development team. In 2016 he contributed to starting the new Tenova Digital Transformation initiative and later joined the new Tenova Digital Team as Director.

Professor Dr.-Ing. Katja Windt, Member of the Managing Board at SMS group Katja Windt finished her PhD at the Institut für Fabrikanlagen und Logistik (Institute of Production Systems and Logistics) in 2000 after her degree in mechanical engineering at Leibniz University Hannover. She joined SMS group as member of the managing board in January 2018 and is in charge of the Electrics/Automation division. In her last

position Windt headed Jacobs University Bremen, Germany, as President for four years. She took on the professorship of global productions logistics at Jacobs University, and in the same year was awarded the Alfried Krupp Prize for Young Professors. Windt was named “Professor of the Year 2008” by the German Association of University Professors and Lecturers.

Eric Vitse, Chief Technology Officer at Liberty House Group Eric is leading LHG’s move towards clean, lowcarbon methods across all of its industrial operations.

He was previously CTO for Erdemir, Europe's third largest steel producer, after a three-decade career with ArcelorMittal. As CTO for ArcelorMittal in the Americas he was responsible for capital investment strategy and technical oversight across more than

14 speakers bio.indd 8

FUTURE STEEL FORUM

40 steel-making facilities with a combined capacity of over 40Mt/yr. At Erdemir he was responsible for their engineering transformation programme, devising and implementing innovative technologies, processes and solutions to support LHG's growth strategy. This included securing environmental best practice solutions and vertical integration opportunities.

Steel Times International

18/05/2018 13:03:46

Marco Ometto, Executive Vice President at Danieli Automation Marco Ometto DANIELI Automation in 1994 as project leader for the development of manufacturing execution systems (MES) in the metals industry. As manager, he had 13 engineers working with on analysis, development and commissioning of MET systems. He was appointed manager of process control systems in 1998 and by 2001 he had been

appointed manager of automation systems design and development for electric meltshops and CCMs for long and flat steel products. Two years later he was appointed executive manager with the task of studying the expansion of the company abroad. He is currently executive vice president of Danieli Automation.

Dr Ing. Valentina Colla, Co-Ordinator of the ICT Centre at Scuola Superiore Sant’Anna-TeCip Institute Dr. Colla holds a masters degree in engineering and a PhD in Robotics and is now technical research manager at Scuola Superiore Sant'Anna in Pisa. She is responsible for the Centre of ICT for Complex Industrial Systems and Processes (ICT-COISP) of the TeCIP Institute of SSSA, and has over 20 years of research experience in the steel sector. Dr. Colla is a member of the European Steel Technology Platform (ESTEP) and a

representative of SSSA within SPIRE. Co-author of around 280 papers published in international journals and conferences, her expertise concerns simulation, modelling, control and optimisation technologies for industrial applications as well as data pre-processing, data analytics and mining through traditional and Artificial Intelligence-based techniques.

Emilio Riva, CEO & Founder at Steel Hub, London Emilio Riva is CEO and founder of London-based Steel Hub, a consultancy boasting the highest performing team of technical steel consultants in Europe. A former member of the executive board of Riva Group, Riva is a steel industry leader with a determination for solving complex and challenging problems in steel plants.

Riva has an impressive track record of driving global success in reducing the cost of steel manufacturing and continuous casting processes using state-of-the-art modelling and optimisation techniques in steel plants across the world.

Jane Zavalishina, President & Co Founder at Mechanica AI Jane Zavalishina is president and co-founder of Mechanica AI, a provider of AI-based solutions for the industrial sector. For over 15 years, she held various executive positions within Yandex, one of Europe’s largest internet companies, being in charge of using innovative technologies to create new and

Steel Times International

speakers bio.indd 9

transform traditional businesses. Jane is a regular voice at international events on AI-related topics. She also serves on the World Economic Forum’s Global Future Councils. In 2016, Jane was named in Silicon Republic’s Top 40 Women in Tech as an inspiring leader.

FUTURE STEEL FORUM

15 18/05/2018 13:03:47

SPEAKER PROFILES

Heiko Wolf, Project Manager, FutureLab at PSI Metals GmbH Software developer, PSI Metals GmbH, Berlin (2006 – 2010).

Technical project manager for Logistics @ SSAB Mobile, PSI Metals North America, Pittsburgh (2011 – 2012).

Project manager framework development, PSI Metals GmbH, Berlin (2013 – 2014).

Project manager FutureLab, PSI Metals GmbH, Berlin (2015 – now).

According to Heiko Wolf, Industry 4.0 offers developed countries the opportunity to keep or regain a leading edge in production and manufacturing.

Professor Dirk Schaefer, International Thought Leader in Cloud-Based Design and Technology for Digital Manufacturing at University of Liverpool Professor Dirk Schaefer holds the Chair in Industrial Design at the University of Liverpool in the UK. He previously held academic positions at the University of Bath (UK), the Georgia Institute of Technology (United States), and the University of Durham (UK).

Prof. Schaefer is spearheading internationally leading research on cloud-based design and manufacturing

(CBDM) and social product development (SPD) in the context of Industry 4.0. His research group coined the term CBDM, and in 2014 he published two books on cloud-based design and manufacturing. A third volume in Springer’s Advanced Manufacturing Series, Cybersecurity for Industry 4.0: Analysis for Design and Manufacturing, appeared in 2017.

Dr Luc Bongaerts, Business Development Manager at OM Partners, Belgium Luc Bongaerts is business development manager at OM Partners. His PhD in mechanical engineering focused on the integration of scheduling and shop floor control of holonic manufacturing systems. Holonic manufacturing was part of the intelligent manufacturing initiative that focused on autonomous and co-operating

agents organising themselves to form agile, adaptive and high-performance production systems for the 21st century. Luc was active in supply chain management for 25 years. His experience includes several SCM projects at metals companies, focusing on delivering true value through integrated supply chain planning.

Chris Oswin, Research Group Manager, Materials Processing Institute Chris Oswin is manager of process simulation and materials engineering at the Materials Processing Institute, a research and innovation centre serving organisations that work in advanced materials, low carbon energy and the circular economy. He is responsible for the delivery of research and technical activities including physical modelling, finite element

16 speakers bio.indd 10

FUTURE STEEL FORUM

analysis, computational fluid dynamics and the Forensic Metallurgy carried out in the Institute’s Metallurgy Laboratory.

A graduate in physics from Durham University, Chris joined what was then British Steel in 1992 and has worked in industrial research and development ever since.

Steel Times International

18/05/2018 13:03:49

Dr Chenn Q Zhou, Founding Director, Steel Manufacturing Simulation & Visualisation Consortium at Purdue University Northwest Dr. Zhou is founding director of the Steel Manufacturing Simulation and Visualisation Consortium (SMSVC) and the Centre for Innovation through Visualisation and Simulation (CIVS), and Professor of Mechanical Engineering at Purdue University Northwest. She is also Professor by

Courtesy at Purdue University West Lafayette. Her BS and MS degrees are in power engineering from Nanjing University of Aeronautics and Astronautics, China. Her Ph.D. in mech’ eng’ is from Carnegie Mellon University, USA. She joined Purdue University Northwest in ‘94 after 3 years industrial experience.

Wolfgang Oberaigner Head of Through-Process Quality Control at Primetals Technologies Wolfgang Oberaigner is heading the Through-Process Quality Control (TPQC) team at Primetals Technologies and is responsible for product development, project execution and sales for TPQC systems. Starting as a software developer for caster level 2 systems he

continued to work as a project and commissioning engineer and later as a project manager for level 2 projects. In 2004 he took over the caster level 2 team for which he has been responsible for more than a decade.

Dr. rer. nat. Marcus J. Neuer, Project Manager at VDEh-Betrebsforschungsintitut Marcus J. Neuer was born in Germany in 1977. He received the Dipl.-Phys. degree, with distinction, from the Heinrich-Heine-University Düsseldorf in 2003.

From 2003 to 2006 he was researcher at the Heinrich-Heine-University in different educational and scientific roles, where he received the degree Dr. rer. nat. in 2006, following a Ph.D. thesis on a stochastic theory

about anomalous transport in plasmas. Since 2012, he has worked for VDeh Betriebsforschungsinstitut GmbH (BFI), initially as project manager for algorithms and artificial intelligence. Marcus has more than 15 years experience in software architecture and development.

His interests include multi-agent systems, holonic manufacturing and Monte-Carlo simulations.

Kristiaan Van Teutum, Vice President, Sales & Marketing at Fives Kristiaan is responsible for sales and marketing of Fives’ Steel business line which covers engineering, process expertise, strip processing line and reheating furnace design and supply, including mechanical equipment, thermal technologies and induction heating solutions. Graduating as a mechanical engineer in 1988 from Portsmouth University (UK), he spent five

Steel Times International

speakers bio.indd 11

years in the UK in technical sales then moved to Italy where, over 20 years, he held executive commercial roles in two multi-national engineering companies related to the steel sector for both long and flat products before joining Fives in Paris.

FUTURE STEEL FORUM

17 18/05/2018 13:03:52

SPEAKER PROFILES

Professor Chris Hankin,Co-Director, Institute for Security Science and Technology at Imperial College London Professor Hankin joined Imperial in1984 and was promoted to Professor in1995. He is co-director of the Institute for Security Science and Technology – succession planning. His research is in cyber security and data analytics. He leads multi-disciplinary projects focused on developing advanced visual

analytics and providing support to defend against cyber attacks. He is director of NCSC/EPSRC Research Institute on Trustworthy Interconnected Cyber-Physical Systems, which involves 5 universities developing a better understanding of the cyber threat to industrial control systems.

Dr Nils Naujok, Partner, Consulting Leader Metals Industries Europe at PwC Strategy& Nils Naujok is EMEA metals consulting leader and partner with PwC Strategy&, which reinvents strategy consulting as the world’s leading strategy-throughexecution firm. Based in Berlin, where he specialises in strategy development, operating model development, operations and innovation strategies for the metals and process industry. He is the leader of PwC’s EMEA

steel and metals consulting practice and of Strategy&’s Innovation and Development Excellence practice. In this role he is in close contact with leaders of the European steel and metals industry, industry associations, media representatives and technology partners. One key element of his consulting focus is the advice related to industry 4.0 in the steel and metals industry.

Dr Joe Flynn, Assistant Professor at University of Bath Dr. Flynn graduated from the University of Bath with honours in integrated mechanical and electrical engineering. He has since completed a PhD in manufacturing metrology and automated testing, and conducted postdoctoral research on additive manufacturing. He became an assistant professor at the University of Bath in December 2016. Since then,

his research includes design for additive manufacturing and various themes around the Industrial Internet of Things (IIoT) and Industrie 4.0 (I4). He is currently working with start-up companies implementing datadriven business models within the power generation sector, while exploring the impacts of IIoT and I4 on UK education and technological (un)employment.

Dr Richard Curry, Director, Operations, Materials Processing Institute Dr Richard Curry is director, operations, for the Materials Processing Institute. The Institute carries out industrial research and innovation in advanced materials, low carbon energy and the circular economy and has been supporting the materials, processing and energy sectors for over 70 years.

Richard was part of the team driving the divestment of

18 speakers bio.indd 12

FUTURE STEEL FORUM

the Institute from its then parent company, Tata Steel, returning the organisation to independent ownership in 2014. Richard’s background is in harsh environment sensors, novel analogue integrated circuit design for biological integration and holographic photolithography.

Steel Times International

18/05/2018 13:03:55

®

With over 45 years in the Steel Industry, we have a wide variety of solutions to keep your mill rolling.

Traversing Restbars allow for quick and accurate pass changes

· Honours existing mill attachment points and guide base, no machining on your stands required · Made from stainless steel and specifically tailored to your mill ensuring perfect fit, operation and longevity

Our roller entry guides keep your product on the pass · Single point, centralized adjustment during operation · 2, 3 and 4 roller configurations · Rigid, stainless steel construction · Broad size range

® spiral cooling system

Maximize your mill speed while meeting your quenching requirements on all bar sizes. Not just controlled cooling, but correct cooling

box configuration

· Fewer surface defects and better scale control · Available in Box, Trough and Restbar-mounted configurations

Contact Us www.fabris.com

info@fabris.com

905.643.4111

trough configuration

EXHIBITOR PROFILES

A.L.B.A S.R.L Cutting Technology Stand A18

AMETEK Land (Land Instruments) Stand A21

Contact details Phone: +39 3346883938 Email: e.dottavi@albacut.com Website: http://www.albacut.com/

Contact details Stubley Lane, Dronfield, S18 1DJ, United Kingdom Tel: +44 1246 417691 Email: land.enquiry@ametek.com

ALBA is an organised group of four companies capable of satisfying any customer requirements in the steelmaking industry connected with torch cutting, gas processing and special equipment.

The company directly manages design, raw materials purchasing, in-house manufacturing, assembly, testing, distribution and service operations.

ALBA is the family-owned head company of the group. Based in Genova, Italy, and established in 1956, it is a leading global manufacturer of torch cutting systems and special equipment for the steelmaking industry.

ALBA Meccanica is specialised in on-drawing steel constructions as well as complete machinery assemblies.

ALBACUT Korea is the service centre and distributor for east Asia.

ATES is responsible for automation and service.

The range of products and services offered includes – but is not limited to – the following: • complete torch cutting machines for continuous casting of billets, blooms and slabs • deburring systems for billets, blooms and slabs • special gas cutting machines • gas control stations • heating/drying equipment • oxygen lances • gas burners • torches and nozzles • safety valves, flash back arrestors and regulators • on-drawing steel constructions and machinery assemblies

AMETEK Land is a leading manufacturer of monitors and analysers for industrial infra-red non-contact temperature measurement, combustion efficiency and environmental pollutant emissions.

Through its trusted range of leading-edge technologies, it is chosen the world over to deliver highly accurate measurement solutions that meet all process needs.

Successful steel production requires accurate measurements across a wide range of temperatures and under a variety of different conditions. AMETEK Land provides comprehensive temperature measurement solutions supported by more than 70 years’ experience serving the steel industry.

The company offers dedicated solutions for key applications and flexible instrumentation that can be customised for specific processes.

The products support higher quality, lower costs, and greater safety.

Built to operate in the harsh environments found in steel production, these instruments are designed to the highest performance standards, optimised for making temperature measurements at every important stage of the process.

AMETEK Land’s bestselling SPOT pyrometer is a fully-featured, high-performance solution for fixed non-contact infrared spot temperature monitoring. SPOT offers powerful processing, communications and control functions that deliver accurate single spot measurements which users depend on to optimise their application processes; helping them to maintain high product quality and protect against costly inefficiencies.

AMETEK Land is part of the process & analytical instruments division of AMETEK, Inc., a global supplier of high-end analytical instrumentation.

ALBACUT4.0 is the technology that takes automatic torch cutting into a new era, responding to the latest demands of smart factories. It offers a solution for predictive maintenance and machine management that will allow steel plants to reduce downtime, increase productivity and minimise costs.

Beda Oxygentechnik Armaturen GmbH Stand 23

Contact details An der Pönt 59, D-40885 Ratingen, Germany Phone: +49/2102/910914 Email: jan.matthias@beda.com Website: https://beda.com/en/

20 profiles exhibitors.indd 2

FUTURE STEEL FORUM

Steel Times International

18/05/2018 13:06:11

BEDA Oxygentecknik Armaturen GmbH is leading global supplier of safety oxygen lancing solutions for steel plants and foundries. The company supplies optimised solutions for blast furnaces, converters, EAF meltshops, casting areas and scrapyard applications. Beda is committed to the highest safety and best cost efficiency, and offers leakage free argon stirring solutions, contributing to process stability and quality in secondary metallurgy and delivering substantial savings in operating costs. With various products for oxygen and carbon injection, BEDA contributes to the safety and productivity of EAF operation.

BM SPA

Stand A14

Contact details Email: sales@bmgroup.com Website: http://www.bmgroup.com/

Endress+Hauser

BM Group (BMG) is an Italian company operating globally as a supplier of process automation equipment and customised robotic solutions for industry. PolytecRobotics is a BMG brand and is a success story that begins in 2012. Thanks to know-how acquired in the steel sector, BMG focused on the production of highly technological robotic cells for the steel and tube/pipe sector. A deep knowledge of the steelmaking process together with a constant investment in R&D have been key to understanding the increasing needs of the steelmaker. In a few years, Polytec has developed a range of more than 15 robotic cells that integrate the steelmaking process, from the furnace to the finishing mill, long and flat as well as tube and pipe.

Each solution is manufactured based on customer needs, tested in the Polytec workshop and rapidly installed in new and existing steel plants. Advanced machine vision systems collect data and verify product quality and process performance.

CMI INDUSTRY

Stand A03

Contact details MI (headquarters), Avenue Greiner 1, 4100 Seraing, Belgium Email: welcome@cmigroupe.com phone: +32 4 330 2444 Website: www.cmigroupe.com

CMI designs, installs, upgrades and services equipment for energy, defence, steelmaking, the environment and other industry in general.

The company assists clients throughout the entire life-cycle of

Steel Times International

profiles exhibitors.indd 3

their equipment in order to improve the economic, technical and environmental performance of the equipment. The benefits of CMI are numerous: a unique combination of expertise in engineering, maintenance and management of international technical projects, a vast geographic and technological scope, and an ability to innovate in accordance with the needs of its customers. The company employs 4 ,700 experienced people in Africa, Brazil, China, Europe, India, New Caledonia, Mexico, Russia and the USA. CMI Industry, a division of CMI Group, designs, supplies and modernises cold rolling mills, processing lines, chemical and thermal treatment installations for the steel and non-ferrous industries. It provides state-of-the-art heat treatment technologies for the aviation, forging and casting industry, as well as surface treatment installations for all types of industries. CMI Industry supplies greenfield and brownfield installations and equipment, and provides related services, plus training and technical assistance. CMI’s reliable and cost-effective, yet innovative, solutions are adapted to the specific needs of each and every customer.

Stand A10

Contact details Colmarer Str. 6, Weil am Rhein D-79576 Germany Phone: +49 7621 975 935 Email: jens.hundrieser@de.endress.com Website: http://www.de.endress.com/de

Endress+Hauser is a leading supplier of products, solutions and services for industrial process measurement and automation. It offers comprehensive process solutions for flow, level, pressure, analysis, temperature, recording and digital communications across a wide range of industries, optimising processes with regard to economic efficiency, safety and environmental protection.

The company is focused on seven strategic sectors, one being the metals industry. With more than 60 years of experience, Endress+Hauser helps its customers improve process efficiency, cost savings, plant safety and sustainability.

Endress+Hauser owns more than 6,500 patents and spends approximately 7% of sales on R&D – it is also at the forefront of the digitalisation trend, supporting its customers along the first step towards Industry 4.0.

Where digital communication is concerned, it enables advanced measurement sensor diagnostics which can form the basis of effective process condition monitoring and preventative maintenance measures or calibration requests, which can be triggered in an ERP system. Many of the company’s smart measurement sensors can monitor process condition and verify measurement integrity. But that’s one example for how an Industry 4.0 approach can improve everyday business. For further details, ask our colleagues on booth A10.

FUTURE STEEL FORUM

21 18/05/2018 13:06:12

EXHIBITOR PROFILES

Fives

Stand A05

Contact details Phone: +33 1 45 18 65 35 Email: steel@fivesgroup.com Website: http://www.fivesgroup.com/

Fives’ Steel business line provides steelmakers with turnkey process expertise and technologies for the carbon, stainless and silicon steel sectors of the long and flat products markets, as well as tube and pipe, including strip processing, reheating and rolling.

As a provider of turnkey solutions and a full range of equipment, Fives offers: • Metallurgical intelligence and competitive strategy development • Design and supply of strip processing lines and reheating furnaces • Thermal and mechanical technologies and proprietary equipment • Automation systems • Smart maintenance Fives puts innovation at the core of its development strategy and invests in R&D to design and create pioneering technologies that meet the performance criteria of industrial companies across a broad range of sectors. Fives has adopted a collaborative approach, built on partnerships and joint ventures with public and private players to contribute to building the factories of the future.

Today’s plants are becoming “smart” and more agile. Digital technologies are powerful tools to improve operational efficiency.

Fives combines its process expertise with digital tools – such as data and flow management, modelling and simulation of production line equipment, digital control and robotisation – to offer industrial companies solutions that facilitate production system management and maintenance.

IBA AG

Stand A13

Contact details Email: connected@iba-ag.com Website: https://www.iba-ag.com/en/germany/processanalysis/

IBA AG is measurement and automation technology specialist with a mission to bring transparency to the world of industrial systems. By using an IBA system steelmakers can be sure that their plants and

22 profiles exhibitors.indd 4

FUTURE STEEL FORUM

machines are recorded by 360 ° and thus all individual processes can be recorded and made completely visible.

Our acquisition systems and software solutions to measure, check and analyse machines and plants are scalable and thereby expandable at any time. They not only grow with increasing demands, they also get along with almost every industrial control system.

IBA systems can document the complete production chain – and provide a comprehensive view. All recording results are available to our customers in a data pool from which they can optimise function and design of their systems. With modern tools for data analysis, root-causes can also be found for high-sporadic faults and thus permanently improve availability.

LAP GMBH LASER APPLIKATIONEN Stand A01

Contact details Zeppelinstr. 23, 21337 Lüneburg, Germany Phone: +49 4131 9511 12 Email: d.meuser@lap-laser.com Website: https://www.lap-laser.com/

LAP supplies laser-based systems for high-precision measurements of geometric dimensions, such as width, thickness, length, diameter and flatness. LAP systems provide ultra-precise measurement results under the harshest operating conditions. Hundreds of LAP systems are tried and tested every day in rolling mills worldwide.

The aim of the company’s systems is to save resources in the process of long products production using intelligent laser-based measuring gauges for automatic inline detection of rolling defects.

So far, gauges that measure the geometrical dimensions and shape of wires and bars, profiles or tubes, have become a standard in the rolling production of long products. Yet, although there are significant benefits, there is still room for improvement with respect to excess production and producing scrap that needs to be recycled to save resources in terms of materials, energy, workforce and operational time. The ability to intelligently detect rolling defects automatically supports current efforts to automate the long products production process and initiate the benefits of Industry 4.0. The LAP solution thus contributes to supporting the aims of achieving higher production efficiency and resource savings through scrap minimisation, and reducing energy consumption in the overall production processes.

Baosteel, POSCO, ThyssenKrupp Steel, Vallourec and other leading steelmakers are customers of LAP. LAP staff support the company’s worldwide customer base from company headquarters in Lüneburg/Germany and through an international network of branches and technical agencies.

Steel Times International

18/05/2018 13:06:13

Materials Processing Institute

Stand A08

Contact details Eston Road, Middlesbrough TS6 6US United Kingdom Phone: +44 (0)1642 382000 Email: enquiries@mpiuk.com Website: https://www.mpiuk.com/

The Materials Processing Institute is a research and innovation centre serving organisations that work in advanced materials, low carbon energy and the circular economy. The MPI provides technology and R&D based services and consultancy. Scientists and engineers apply their expertise to progress innovation, develop materials and improve products and processes.

The Institute is equipped with state-of-the-art steel making and refining equipment together with ingot and continuous casting facilities. The Institute has continued to be at the centre of innovation and new product development for over 70 years with many modern high performance steels and their process routes being developed in the laboratories and the Normanton Plant before mass production at the larger steel manufacturing sites.

The Materials Processing Institute possesses a 7T pilot facility capable of producing plain carbon and alloy steel types by continuous casting and ingot casting. The plant is used primarily for development of new steel alloys, but also supplies highly specialist steels. This facility is being reconfigured as a ‘Future Steel’ plant, providing a test bed for industry 4.0 technologies to be developed.

At the Institute a four-stage model has been developed and successfully applied for full application of Industry 4.0 technologies.

The four stages are: measurement, monitoring, expert system and closed loop control. At the conference the approach being taken at the Institute for development and integration of industry 4.0 will be outlined, with emphasis on the routes through the pilot and demonstration scale, for safe and reliable integration into steel plants.

Novatech APS

Stand A20

Contact details Skudehavnsvej 3, 9000 Aalborg Denmark Phone: +45 9816 5009 Email: sale@novatech.dk Website: http://www.novatech.dk/

Heavy transportation solutions specialist Novatech is a Danish sales and engineering company with its own production facilities in Poland.

Steel Times International

profiles exhibitors.indd 5

The company specialises in horizontal transport equipment, delivering prime solutions for special industrial purposes and for the shipping industry, among others.

The product portfolio includes: Roll trailers, flexmasters (translifters), cassetes, hydraulic lift trailers, magnetic lifters, trailers with weight systems, heat-resistant trailers, trailers dedicated to transport steel coils, slag and others.

The equipment can be produced in different variants and for various purposes, in accordance with the actual requirements of the customer, and after-sales services are provided too. Novatech products are manufactured in the company’s own production plant and meet the highest quality standards. An experienced team of engineers design and develop the equipment, as well as optimise the production process and introduce innovations with the aim of meeting customer requirements and enhancing their productivity.

NT Liftec Oy

Stand A20

Contact details Sorkkalantie 394, 33980 Pirkkala Finland Phone: +358 3 3140 1400 Email: sales@ntliftec.com Website: https://www.ntliftec.com/

NT Liftec Oy, previously TTS Liftec Oy, is located in Pirkkala, Finland, from where it designs and delivers cost-efficient horizontal transportation systems for ports and heavy industry. The company’s portfolio includes systems for containers and cargo cassette handling along with extensive service and after sales capabilities.

The company is market leader and is widely known as an innovative and reliable supplier of high quality products and solutions.

As a development-oriented company, NT Liftec Oy is driven by customer needs, offering systems and services that are designed to maximise the profitability of customers’ and cargo owners’ businesses.

OM Partners

Stand A04

Contact details Phone: +32 3 650 2211 Email: LBongaerts@ompartners.com Website: https://ompartners.com/en/solutions/omp-for-metals

OM Partners is a software and consulting company delivering supply chainplanning solutions for mill products (metals, paper and packaging, floor covering) and semi process industries (chemicals, pharmaceuticals and consumer products).

FUTURE STEEL FORUM

23 18/05/2018 13:06:14

EXHIBITOR PROFILES

With more than 250 customers and 650 implementations, OM Partners has established solid partnerships with customers all over the world. With annual group sales revenues of over 46.1 million EUROs and a workforce of over 400 people in offices in Antwerp, Atlanta, Shanghai, Dubai, Sao Paulo, Paris, Rotterdam, Cologne and London, the company has developed into a top player in the supply chain planning market.

OM Partners’ OMP Plus is an integrated solution for all planning related issues, from the strategic to operational level. It is aimed at reducing logistics costs and throughput times and at increasing the reliability of delivery dates and customer satisfaction. The revolutionary technology of OMP Plus makes integrated demand planning, supply planning and scheduling a reality.

Primetals Technologies

Stand A06

Contact details Email: contact@primetals.com Website: http://primetals.com/en/Pages/Home.aspx

Primetals Technologies, headquartered in London, UK, is a worldwide leading engineering, plant-building and lifecycle partner for the metals industry. The company offers a complete technology, product and service portfolio that includes integrated electrics, automation and environmental solutions covering every step of the iron and steel production process that extends from raw materials to finished product – in addition to the latest rolling solutions for the nonferrous metals sector.

As Primetals Technologies has been a provider of automation solutions of all levels to steel producers for decades, the digitalisation of the metals industry has been one of the company’s main focus areas for a long time.

Primetals Technologies is a joint venture between Siemens, Mitsubishi Heavy Industries (MHI) and partners.

PSI METALS

Stand A07

Contact details Heinrichstraße 83-85, 40239 Düsseldorf Germany Phone: +49 211 60219-271 Email: info@psimetals.com Website: www.psimetals.com

PSI is the leading partner for digital production in the metals industry combining SCM, APS and MES within one software platform –

24 profiles exhibitors.indd 6

FUTURE STEEL FORUM

PSImetals. The company’s software solutions enable aluminium and steel producers to ensure their competitive edge by delivering products as agreed in quantity, quality and time while considering inventory, productivity and performance targets.

The PSImetals software line is an end-to-end approach for the overall supply chain caring for all the needs of the primary metals industry. From supplier to customer, PSImetals offers powerful and highly configurable standard products to support all processes from planning to execution while respecting the complexity of metal production: • Planning level to support all planning processes from business planning via production planning to detailed scheduling, • Execution level to monitor and control production activities as well as to assure quality, • Level of material- and transport logistics to optimise all transports requested to keep production running, • Energy management level, • Cross-application KPI and production monitoring functions.

All information is based on PSImetals Factory Model – a digital twin of the whole supply chain providing consistent real time plant status information.

As market leader PSI claims technology leadership as well.

Therefore, PSImetals FutureLab investigates and develops the solutions of tomorrow taking into consideration • latest developments around Industry 4.0 • a collaborative approach with customers, partners and experts • leading edge IT based on PSI Java Framework.

Combining 45 years of experience in implementing production management software with innovation, PSI supports numerous metals producers around the globe in achieving their competitive edge.

PRISMA Impianti S.P.A

Stand A22

Contact details Via Asti, 7, 15060 Basaluzzo (AL), Italy Phone: +39 0143.48.98.91 Email: manuel.alfonso@prismagroup.it Website: http://www.prismagroup.it/

PRISMA Impianti is a system integrator highly specialised in process control for the steel industry. We have realised plants throughout the entire steel production chain: mineral parks, coking plants, blast furnaces, continuous casting, hot and cold rolling mills, reheating and heat treatment furnaces, processing lines for long and flat products (plus coils and tubes), such as annealing, pickling, galvanising, coating, skin pass mills and presses. The competence and reliability of PRISMA Impianti are globally recognised by major steel producers.

Steel Times International

18/05/2018 13:06:15

PRISMA Impianti provides a complete package that includes feasibility study, risk analysis from the company’s functional safety engineers certified by TUV, basic and detailed engineering, equipment supply, drives configuration, as well as the development of control software (PLC and SCADA) and higher-level systems such as manufacturing operation management (MOM). During the event we will present our Cyber Security suite SEC.R.A (SECure Remote Access) to protect data and industrial network from cyber attacks.

making a decision in case of quality or process deviations – throughout the entire value chain.

Quinlogic GmbH

Contact details Email: info@sap.com Website: https://www.sap.com/industries/mill-products.html

Stand A11

Contact details Heider-Hof-Weg 23 52080 Aachen Germany Phone: +49 (2405) 47 999 40 Email: info@quinlogic.de Website: https://www.quinlogic.com

The company was established in Aachen, Germany, in 2008 with the aim of harnessing large quantities of complex measurement data for quality management in the steel and aluminium industries so as to facilitate significant increases in production efficiency.

QuinLogic’s customers include a large number of well-known steel rolling mills throughout the world. The SMS Group has been a majority shareholder in the company since 2016.

The outstanding and easy-to-use technology of the QES – Quality Assurance System – is increasingly applied in premium steel rolling mills.

Industry 4.0 QES - A PRAGMATIC STEP TOWARDS INDUSTRY 4.0 The discussion about the future transformation in industrial production processes is ongoing. Industry 4.0 does not have one standardised definition, but there are major criteria to describe it: • Better use of data • Merge production processes with information technology • Real-time availability of relevant information • Mass customisation • Technical decision assistance • Finally it is an application that reveals know-how, preserves it and makes it available to anybody in the plant 24/7 • Link the real production world with the virtual world QuinLogic’s QES is a 4.0-compliant application that: • Makes production data transparent • Supports the consideration of individual customer specifications in a mass production environment • Provides relevant information to support the human being in

Steel Times International

profiles exhibitors.indd 7

SAP

Stand A19

A leader in enterprise application software, SAP helps companies of all sizes and industries run at their best. From back office to boardroom, warehouse to storefront, desktop to mobile device – SAP empowers people and organisations to work together more efficiently and use business insight to stay ahead of the competition. SAP applications and services enable more than 378,000 business to operate profitably, adapt continuously, and grow sustainably.

Massive shifts in the steel industry are forcing dramatic changes in global and local markets. Steel companies need to find solutions to be profitable in oversupplied, competitive, and constantly changing markets. SAP has 40 years’ experience working with hundreds of steel manufacturers globally to use digital innovation to anticipate real-time demand and supply, enhance process excellence for operational efficiency, operate resilient supply chains, and innovate the customer experience. SAP is in tune with how to apply the newest technologies to challenges and opportunities facing the steel business.

One of SAP’s newest innovations solves challenges in the area of reducing the high costs of asset maintenance. Organisations manage thousands of assets to keep plants operational. Better pooling and sharing of all up-to-date asset-related content is one of the key problems facing organisations. Maintenance can be a difficult challenge, due to heterogeneous systems, or incomplete operators’ manuals. SAP’s asset intelligence network (AIN), solves these issues.

AIN is a cloud-based collaborative network that allows companies to collect, track, and trace equipment information in a central repository.

Operators can access up-to-date maintenance strategies, manuals, and more from manufacturers – and manufacturers can automatically receive asset usage and failure data from operators.

SMS digital

Stand A16

Contact details Erkrather Str. 234 b, 40233 Düsseldorf Germany Phone: +49 211 881-5332 Email: hello@sms-digital.com Website: https://sms-digital.com/#home

SMS digital GmbH is a start-up of the SMS group, the market leading

FUTURE STEEL FORUM

25 18/05/2018 13:06:16

EXHIBITOR PROFILES

constructor of metallurgical plants and machinery for the industrial processing of steel, aluminium, and nonferrous metals. With more than 13.000 employees worldwide SMS group generates a revenue of about 3.3 billion EUR a year.

SMS digital takes up the challenges of ‘Digitalisation’ and ‘Industry 4.0’ within the SMS group. With the use of state-of-the-art innovation methods, metallurgical processes know-how and technological expertise, we create new digital products, that, from the very beginning, are developed in close collaboration with our customers and end-users as well as with SMS group experts. This partnership results in the best possible solutions, which are perfectly tailored to customer needs, with immediate added value.

Founded in May 2016 as an independent business unit, SMS Digital is still under development – creating ideas, interviewing customers, recruiting staff. Based in Düsseldorf’s Schwanenhöfe, the continuously growing, young and dynamic team is tackling the challenge of developing digital products for the steel industry. The aim of SMS digital is to become a leading provider of industrial IT services and digital solutions. All of our products are the first steps of the vision of an intelligent steel plant that takes full advantage of the 21st century technology to increase productivity and user friendliness.

Spraying Systems Co

Stand A17

TMEIC

Stand A02

Contact Details Website: https://www.tmeic.com/industry/metals

TMEIC drives industry around the world through a comprehensive offering of unique systems solutions including variable frequency drives, motors, photovoltaic inverters and advanced automation systems for a wide range of industrial applications. Established in 2003, Toshiba Mitsubishi-Electric Industrial Systems Corporation (TMEIC) resulted from the integration of Toshiba and Mitsubishi Electric Corporation’s industrial systems divisions.

The company’s committed approach to collaborative solutions development ensures every industry throughout the globe can benefit from the world’s brightest minds. As a result of this powerful combination of resources, TMEIC is positioned to develop innovative technologies, quickly respond to industry trends and apply solutions to a wide variety of industrial market segments around the world.

UVB TECHNIK S.R.O

Contact Details Oddział w Polsce, Wyczółkowskiego 23, 44-109 Gliwice Poland Phone: +48 / 32 / 238 81 11 Website: http://www.spraying.pl/

You’ll find spray technology solutions for every area of your mill in SSC’s extensive product line. Our CasterJet, DescaleJet Pro, VeeJet and FullJet nozzles are industry standards. But that’s just the beginning – we have nozzles and systems for precision oil application, selective roll cooling and heating, dust suppression, descaling and more.

Our steel industry experts work with mills around the world optimising spray operations. Services include on-site evaluations, impact and wear testing in our spray labs, descale header design using proprietary software, gas cooling calculations and process modeling for pollution control operations.

26 profiles exhibitors.indd 8

FUTURE STEEL FORUM

Stand A12

Contact Details Phone: +420 595 044 444 Email: sales@uvbtechnik.cz Website: http://www.uvbtechnik.cz/en/

UVB TECHNIK s.r.o. is a producer of measuring and cleaning systems for metal strip, that is to say high precision continuous strip thickness contact gauge, strip profile gauge, wiping equipment, shapemeter roll and automatic detection of strip defects.

UVB supplies Surfscan systems for cold rolling mills and process lines such as cleaning lines, slitting lines, tension levelling lines and strip edge trimming lines.

The company has multiple references from across the world, such as ArcelorMittal, Jindal Stainless, ThyssenKrupp Electrical Steel, Voestalpine Stahl, MINO, Primetals Technologies France, Wickeder Westfalenstahl, Compania Valenciana de Aluminio, Enersys, US Steel, El Zinc, Nippon Cross Rolling and Osaka Heat Treatment.

UVB’s Surfscan continuous strip inspection system is being exhibited at Future Steel Forum 2018.

SURFSCAN is a visual inspection system designed for defect detection of metal plates, non-woven textiles, foils, paper and so on. It was developed as a highly modular system and, therefore, is fully configurable to customer requirements and can be used in

Steel Times International

18/05/2018 13:06:16

PREDICT TO PREVENT ALBACUT4.0, the new solution for predictive maintenance and machine management.

The best solution for TORCH CUTTING applied in Smart Steel Factories.

RNET OF THI INTE NG

S

ZUMBACH ELECTRONIC AG Stand A09

CLOUD ANALYTIC SOFTWARE BIG DATA

Contact Details P.O. Box, CH-2552 Orpund Switzerland Phone: +41 32 356 0400 Email: sales@zumbach.ch Website: http://landingpage.zumbach-news.com/?l=1

FIELD SENSOR MACHINE DIAGNOSTIC

SEMI-AUTO SETTINGS PROBLEM ANALYSIS REMOTE SUPPORT

HM

ZUMBACH manufactures a comprehensive range of non-contact, on-line measuring and control instruments. Its technology is in use world-wide for such dimensional parameters as diameter, thickness, eccentricity, out-of-round and for physical or electrical parameters.

Zumbach technology is being used around the world by customers who rely on the quality and reliability of our instruments and systems. In the steel industry they are used on hot rolling mills as well as in cold processes.

ZUMBACH equips its powerful instruments with the recognised OPC UA standard. Using this key technology, measurement solutions provide easy, scalable and secure information exchange with different systems in the production line. Instruments with the new OPC UA standard meet the technical requirements for intelligent data networking.

The Zumbach group is represented on all continents with 11 company-owned enterprises as well as over 40 sales and service centres around the world.

IA

ND M

O BILE AP

P

CPLWEB.IT

various production phases, including slitting quality inspection. For surface scanning, the system uses the fastest line-scan cameras in the business, with a bandwidth over 20 GB/s. The system was designed to meet demand for high quality illumination of material and was specially developed for this purpose with an advanced LED unit. Thanks to state-of-the-art components SURFSCAN can detect defects as small as 0.1 mm2 at speeds up to 2000 m/min and a strip width up to 5 m.

SMART MACHINE MANAGEMENT

With our smart technology your plant will reduce downtimes, increase productivity and minimize costs.

Be smart, be ALBACUT 4.0

YOU CAST, WE CUT

WWW.ALBACUT.COM

Steel Times International

profiles exhibitors.indd 9

FUTURE STEEL FORUM

27 18/05/2018 13:06:18

FLOOR PLAN

EXHIBITOR LIST A1

LAP GmbH Laser Applikationen

A2

TMEIC

A3

CMI Groupe

A4

OM Partner

A5

Fives Group

A6

Primetals Technologies

A7

PSI Metals GmbH

A8

Materials Processing Institute

A9

Zumbach Electronic AG

A10

Endress+Hauser GmbH

A11

Quinlogic GmbH

A12

UVB Technik s.r.o.

A13

IBA AG

A14

BM SpA / Polytec

A15

SMS group GmbH

A16

SMS digital GmbH

A17

Spraying Systems Co.

A18

A.L.B.A. Srl Cutting Technology

A19

SAP

A20

Novatech ApS

A21

AMETEK

A22

Prisma Impianti SpA

A23

Beda-Oxygentechnik Armaturen GmbH

A24

Burkert Austria GmbH

28 FLOOR PLAN.indd 2

FUTURE STEEL FORUM

Steel Times International

17/05/2018 12:27:16

A16 A15

A17

Conference Room

A13

A14

A12

Foyer 2

Lisbon (Speakers Room)

Buffet

A9

A10

Registration

A22

A8 A7

A23

A18

A21

Foyer 1 A1

Cloak Room

Coffee

A19

Brussels

A6

Coffee

A20

A11

Foyer C

A24

A2

Safes Office

A3

A4

A5

Amsterdam Refreshment Area

GmbH

Steel Times International

FLOOR PLAN.indd 3

FUTURE STEEL FORUM

29 17/05/2018 12:27:16

DIGITAL MANUFACTURING

The digital transformation of steel The digitalisation of the steel value chain could lead to a productivity revolution in steel making, argues Nils Naujok* and Holger Stamm**

T

he global steel industry is the critical backbone of the industrialised value chain. As an essential base material for significant sectors such as automotive and aerospace, it is a centrepiece of innovation and economic growth. However, despite steel’s importance for the industry today, uncertainty seems to be the only certainty. With high efforts the sector has recovered from the aftermath of the last decade’s recession, when the number of new construction projects worldwide decreased significantly and investments in infrastructure and machinery plummeted.

Besides, to manage volatile demands, companies have to contend with increasingly stringent environmental regulations, volatility in raw materials and steel prices, and greater competition from producers in developing economies such as China. And if that were not challenging enough, in times of increasing protectionism in the US with planned tariffs on imports of steel, the global steel market is bracing for disruption.

Beyond the macroeconomic forces, the steel industry is facing a raft of external challenges. Companies are compelled to transform their operations to satisfy calls for ever stronger and more durable steel, a more diversified and specialised portfolio of products and grades, and shorter innovation

30 PWC.indd 2

FUTURE STEEL FORUM

cycles, which enable enhanced flexibility in mill capacity and prioritise reliable delivery and service to downstream partners or customers.

These disruptions will catalyse the digital revolution in the steel industry. In the past years the steel industry has successfully implemented first digital solutions and use cases such as data analytics to steer the furnace or predictive maintenance solutions in the rolling mills. The challenge now is to further develop the digital strategy and to convert these use cases into digital business models. The results of our CEO survey in 2017 indicate in metals that CEOs expect nothing more than a burst of growth: in their estimate digital business models could lead to a 2.9% growth per annum in revenue and offer a significant potential of 3.6% per annum in cost reduction and performance improvements.

Combining the already existing use cases, the digitalisation of the steel value chain could lead to a productivity revolution in steel making. Productivity could be increased in an end-to-end approach from supplier to customer by using data integration, track and trace solutions, sensors for steel making and casting, flexible steel processing and additive manufacturing. The integrated steel value chain provides significant performance improvements and cost reduction and is a

prerequisite for digital business models.

Digital business models can be split into the dimensions of a digital product portfolio, digital service offerings and digital go-tomarket. Digital product portfolio means the digitalisation of the “physical” product with value added through new and improved product characteristics (quality information of the coil).This provides new options for the steel producer to engage with the customers around the physical product.

To create a digital service offering the steel companies have to expand their existing product portfolio with new digital services such as in product development, design and quality management. A digital go-to-market approach transforms the customer interface and customer experience. This could lead to increased performance through higher customer integration as well as customised customer communication and sales approaches.

In the future these new steel business models will have a significant impact on the integrated steel value chain and will lead to virtual and agile production networks from supplier to customer. Steel companies as key drivers of these networks will play a crucial role in building up digital ecosystems. The end-to-end integration creates transparency along the whole network based on the real-

Steel Times International

16/05/2018 10:22:51

Raw Materials Recycling

Steel production

Component manufacturing

Steel processing

Use autonomous trucks and drones Get real time access to data and digitally enable workers for internal transport [drones can already lift 1 ton today] with augmented reality hardware & software

End customer

E2E process and data integration up to the Track and trace and control of Data mining for remote customer and end customer product the material flow of individual control of critical proas well as multifactorial process steel staffs or coils with RFID cesses and cyber security control, sequencing and planning

Forging

Raw materials BOF Casting

Hot or cold rolling

Surface refinement

Transform press

Scrap metal

EAF Integrated temperature management Decentral production across the whole production process planning based on software agents through sensors

Flexible production adjustment based on real time cutomer data

3D print and additive manufacturing Customer satisfaction through reduction of for rapid prototyping and production rework through dynamic adjustments of steel of components, tools, molds product requirements based on actual data

Logistics and transport Planning and maintenance time data exchange with partners. The further development of digital products, services and a new customer experience on these digital platforms is leading to a new ecosystem for steel.

Looking forward into 2025 the integrated steel value chain and the role of the steel industry could be described by three models:1. Supplier in a digital ecosystem – applying the business model of the big internet platforms like Amazon to the steel industry will create an ecosystem that consolidates and centralises sales, distribution, and planning for various steel producers.

2. Digital system supplier – applying the business model of the new integrated companies for electric cars for the steel industry will increase the integration into the customer industry supporting the customer with design, development, production planning, and component production.

3. Digital network supplier – applying the business model of internet platforms in the steel industry will create a virtual and digital network of partnerships in production, development, and sales.

A steel company can combine elements of these three models depending on their product and customer segments. For example, a producer of higher grade steels for the automotive industry could act as digital system supplier. They will have a close integration into the value creation of their customers providing a platform for

Steel Times International

PWC.indd 3

tool producers, steel producers and car construction. These platforms will allow an integrated part and process design with the full transparency of the value chain. As a result the steel company can extend the forward integration up to the production of pre-finished parts. For the OEM this has significant benefits: reducing waste (ramp up parts), quality costs and lead times.

The steel industry has to manage the digital transformation and the digitalisation of the integrated value chain. A guiding principle for this transformation is that each steel company has to start with the big picture about its role in the value chain and the relevant digital business models for their segments. To identify the best business model the steel industry has to take the perspective of its customers and has to be prepared to fail – and win by experience. The best way to create and develop new business models is to collaborate with partners e.g. customers, other suppliers and technology companies.

Finally, digital may be a buzzword or a hype at the moment, nevertheless it has to be anchored in the DNA of the steel industry as well as in the governance and culture of each steel company. The steel sector is embarking on a digital journey, the companies should take a minute to think about the route, their luggage and their fellow travellers.

Nils Naujok, partner at PwC Strategy& Deutschland – PwC’s strategy consulting business. He is based in Berlin, where he specialises in strategy development, operating model development, operations and innovation strategies for the metals and process industry. He is the leader of PwC’s EMEA steel and metals consulting practice and of Strategy&’s Innovation and Development Excellence practice.

In this role he is in close contact with leaders of the European steel and metals industry, industry associations, media representatives and technology partners. One key element of his consulting focus is the advice related to Industry 4.0 in the steel and metals industry. He supports his clients in the needed transformation to digital operations and digital business models. Nils has more than 20 years of experience enabling metals organisations to deliver substantial benefit by improving their performance.

Holger Stamm, director at PwC Strategy& Deutschland, is co-leading the EMEA metals team of PwC. He is based in Düsseldorf, where he specialises in operating model development, digital and M&A strategies in the metals and process industry.

Holger is responsible for the digital and Industry 4.0 solutions and services for the steel and metals industry. In this role he is in close contact with leaders of the European steel and metals industry, companies along the steel value chain and technology partners. He supports his clients in the needed business and organisational transformation to digital operations and digital business models for more agility, efficiency, digital capabilities and forward integration. Holger has more than 20 years of

* Partner, PwC Strategy& Deutschland.

experience enabling his clients to deliver substantial

**Director, PwC Strategy& Deutschland

benefit by improving their business performance.

FUTURE STEEL FORUM

31 16/05/2018 10:22:54

FUTURE SHOCK

The steel industry is a traditional business. In the past, contracts could get signed on a napkin or sealed by a handshake.

Even if these times have passed, personal trust and experience remain the backbone for long-term success. By Patrick Henz*

Business Ethics 4.0 @ Steel Industry T oday, new generations of managers have introduced Industry 4.0. We are in a continuous technological revolution.

Different experts predict that by 2025, 30% of today’s jobs will be performed by robots, including white collar positions in all departments. This creative disruption will also,

32 primetals Heinz.indd 2

FUTURE STEEL FORUM

on the other hand, require that completely new job positions be created.

Following computer-aided automation, Industry 4.0 puts its focus on the smart workplace. Intelligent machines should understand human employees and adapt to them, not the other way around as before.

Voice-based recognition devices will especially become the employee’s standard method of controlling the different machines.

Even if not all job positions will work like this, steel mills in the future will become high tech workplaces, for both blue and white collar employees.

Steel Times International

16/05/2018 10:26:33

Beyond this positive vision, the development will create new ethics and compliance challenges, which have to be managed in order not to jeopardise the company’s future. Today, the focus is on the US Foreign Corrupt Practices Act from 1977, which enables the US Department of Justice to prosecute companies for global corruption and bribery cases. The situation changes as more and more countries create similar new anti-corruption laws, like the UK Bribery Act or the Brazilian Clean Company Act. In addition to this, the different responsible governmental departments began to cooperate in the investigations.

Today’s compliance systems are tasked

with protecting the company against corruption, antitrust, data privacy, and money laundering cases. Furthermore, these systems demonstrate how the company respects not only clients, but also its own employees, providers, partners and society. Besides clear processes, principles guide employees where new scenarios are not defined by guidelines anymore.

The compliance risk assessment identifies different risk factors to develop protection measures; such as workshops, effective processes and related controls. Imperative is an adequate “tone from the top” to ensure that no one is above the guidelines. Besides numerous opportunities, Industry 4.0 will also add new risks to the steel industry.

Processing of information Economic and technical environments have

Steel Times International

primetals Heinz.indd 3

become more complicated and “information” is the critical resource. This is also valid for compliance departments. The team’s goal is to ensure adequate behaviour and sustainability of the business. Behaviour is based on the perception of information and its correct processing. Based on this idea, the compliance department is responsible for the protection, gaining and usage of information.

Transparency minimises the cost to gain information.

To ensure compliance with guidelines and laws, employees receive information from their employers; such as mission, vision, principles, values and business conduct guidelines. Furthermore, employees have access to a vast amount of external information, which may or may not be accurate. In a worst-case scenario, employees live in “information bubbles.”

Based on culture and experience, or diversity in general, individuals perceive information differently and process it based on learned examples. Correct and clear information, including its perception, is imperative to understand: • laws and guidelines, • the cost of corruption (developing empathy for its victims) and • the consequences of potential violations.

Artificial intelligence What is true for individuals is even more important for Artificial Intelligence (AI). Such software is already capable of beating humans at chess, poker and Ms.Pac Man. But, for adequate predictions and decisions, it requires a correct input of information. This is a risk factor, as hackers no longer have to hack the organisation itself. Instead, damage can be done by manipulating the stream of incoming information so that the company’s AI makes wrong decisions based on incorrect data.

Even if AI is capable of autonomous learning, its base and perception depend on the algorithms that were coded by the human software designers. Due to this, the algorithms may be flawed because of different human biases. That is why these employees become a new focus group for ethics and compliance departments. Steel producers

have to be certain that the programmers are an integrated part of the company, and that they share its vision and values. Software designers share the same “must win” pressure as sales employees, as deadlines are often tight. A risk example for this group would be forgotten “temporary fixes” written into software. These can lead to big problems if they are not replaced before the project is handed over to another person or team.

Autonomous learning includes two steps: 1) Programmed algorithms make decisions based on received information.

2) Based on the monitored results, the AI adds experience to the algorithms and, based on this, adapts the decision-making process.

After his famous “Three Laws of Robotics”, Isaac Asimov added a Law 0: “A robot may not harm humanity, or, by inaction, allow humanity to come to harm.” As corruption is not a faceless crime and there is a real cost of corruption for the victims, an intelligent machine or robot would have to avoid any kind of corruption.

The use of AI inside a company leads to an interesting question: Who inside the organisation will be responsible for ensuring that intelligent self-learning software will act according to laws and regulations? Thinking about AI and automation, we have to dismiss the romantic concepts about robots. Such machines are not individuals, but similar to apps, are connected to the Cloud and act as one big system. Furthermore, there is no classic self-awareness. Rather, AI acts based on a programmed morality. With a focus on programmers, but also directly controlling AI behaviour; intelligent software will be included into the compliance system. This is a logical consequence, as the department is already responsible for fostering adequate human behaviour.

Because AI learns on its own, it is difficult, if not impossible, to understand its compliance by analysing the software’s code or archives.

It is essential, therefore, to monitor the intelligent software and test its behaviour and decisions in different scenarios. At the next level, AI could even learn compliance and ethics via an AI coach. Expressed in mathematical formulas, transparent business maximises long-term success, even if less-

FUTURE STEEL FORUM

33 16/05/2018 10:26:39

FUTURE SHOCK

than-ideal results are achieved in the shortterm. If the AI analyses different scenarios, it will come to the conclusion: “Compliance is a salesadvantage!” In this example, ethics and mathematics combine to create job-enrichment for the next generation of compliance officers.

Liability also has to be considered. Can the company that sold the AI to the client be held liable for what the AI does? Or does liability belong to the company that used the AI? You may tend to believe the latter, as intelligent software learns inside its environment. The exception would be, if the software included an algorithm with a bias in perception or processing of information.

This concept is similar for dog breeders and owners. If a dog attacks someone, the liability is not with the breeder, but the owner.

The dog itself has no liability, but may face the consequences of being put to sleep.

Furthermore, even if the breeder has no direct liability, legal decrees may prohibit the sale of certain dog breeds if the legislature identifies them as a public risk.

Cognitive hacking The world is not black and white. There is no clear line between humans and AI. We have practically outsourced parts of our memory to the Internet. Scientists found out that humans

Mission vision

today take less time than in the past to try and answer a question from memory. Instead, the tendency is to give up more quickly and rely on smart phones to find the answer in Google or on the Wikipedia-app. Even if we don’t have hardware directly integrated into our bodies, we have practically become cyborgs in a sense, vulnerable to the manipulation of stored information or the use of biased online portals.

The term “cyborg” was first coined in 1960 by the scientists Manfred Clynes and Nathan Kline. It describes a human individual with artificial parts in order to replace missing ones or to achieve enhanced abilities.

Hackers today are not the cliché nerd sitting at a desk with tons of empty pizza boxes around. Anybody can be a hacker.

The weakest brick in the firewall is often the employee. For this, the most dangerous attacks are not autonomous, but cognitive hacks. With highly personalised phishing

Unit “Grey box”

values BCGs guidelines

Processing

comm.

Compliance

Perception

Forecast Behaviour

Information

34 primetals Heinz.indd 4

FUTURE STEEL FORUM

emails (for example, an email in the name of a superior manager or a provider) employees can be provoked to open an attachment or send transfers to a fake bank account. Names and job-titles can easily be found on social media or inside the company’s communication system.

Creating an effective attack on the human brain requires less computer skills, but more knowledge about different psychological biases.

Hackers can use the effects of time or authority pressure on these biases to influence inadequate behaviour. In one case, they used human curiosity by leaving infected USB drives throughout a corporate parking lot. Such an attack is not limited to the transfer of money or deletion of information. If hackers have internal knowledge, it is possible to stop automated processes and shut off machines.

One such attack led to physical damage at a steel mill.

Anybody can fall victim to psychological pressures and cognitive hacking. The best protection is to create awareness and present different scenarios in workshops.

Such an event is similar to a vaccination.

Normally the employees will not need it. But, if they experience a similar situation, they can remember the scenarios and execute the recommended behaviour, such as ‘do not open suspicious attachments’ and ‘always confirm changes of vendor information’. As cognitive hacking is more a human risk than an IT one, it makes sense that the Compliance department takes responsibility to prepare the employees.

Antitrust Remote administration of equipment in a steel mill is a logical next step. It requires a solid agreement regarding who owns information and/or has the rights to use it. AI can learn from the continuous flow of information. One example is knowing the best time to replace a spare part, and synchronising the change with planned down times to minimise costs.

The AI learns such patterns by analysing information coming from the steel mill.

Based on this, equipment run times can be optimised.

The quality of learning can improve, if the AI not only analyses the information from this particular mill, but also from others with

Steel Times International

16/05/2018 10:26:52

similar machinery – including competitor mills.

With improved AI and a more exact forecast, all mills with access to the AI become more efficient. This could be a competitive advantage for mills that participate. Even if local laws allow for such co-operation, it is a fine line as the AI may analyse patterns in machine usage and predict strategies. Such co-operation may be seen as an illegal cartel, and subject to anti-trust laws.

Today, antitrust laws target the misbehaviour and conspiracy of human employees. This may soon change as, for example, the European Parliament has started a discussion about whether AI and robots should be considered “electronic persons” with civil rights. As a consequence, existing anti-trust laws would apply to them.

Data privacy The connection of all kinds of machines, called the Internet of Things, creates Big Data. This should not be confused with Smart Data. Data itself is a collection of information without any kind of intelligence.

Only the connection of data with statistical methods and logical theories, to predict future behaviour or situations, makes it smart.

Digitalisation plays an important part in Industry 4.0. Production and test facilities, as well as office infrastructure, can be simulated inside the computer. These virtual structures can run in parallel to real-world locations.

As the “digital twin” receives continuous information from its counterpart, the simulation gets highly precise.

The digital twin is used to analyse the efficiency of the complex system, especially to predict how it reacts to changes. This is an essential tool to synchronise maintenance and updates with the mill.

Employees are part of the system. So changes to hardware and software will not only determine the output, but also how employees switch between different positions based on skills and experience. Similar to athletes, corporate employees do not consistently work at their highest levels, but are influenced by mood, distraction, health and other factors. For this, companies may not only include long-term skills into the digital twin, but also the actual performance of the employee, including medical leaves and results from their last performance evaluation.

Steel Times International

primetals Heinz.indd 5

Voiced-based interfaces are expected to become standard for employees to control machines.

Further uses for these gadgets will pop up all over organisations, like monitoring when an individual walks through the door in the morning, for example. As all information will eventually be stored in the cloud, it is an ideal source to make predictions about employee development, including the development of digital twins.

As long as information exists, there will always be a temptation to use it. If a company has sensible information, it is difficult to prevent an organisation from using it. This may potentially lead to a better understanding of the employees. This could be used to establish a more efficient communication with them. Or, it could even be used to manipulate them. The compliance department has to ensure that the different local data privacy laws, such as the 2018 EU General Data Protection Regulation, are obeyed. In doing so, not only is the physical location of the server relevant, but also the country of the client and the nationality of the individuals.

Big Data + Statistical Methods + Logical Theories = Smart Data.

If the data is related to individuals, data privacy may limit smart data or at least require robust controls to avoid leaks or questionable statistic connections.

A company should regularly review which information should be stored and which could be eliminated. Data that is not stored anymore cannot get stolen and misused. It is important to remember that 60% of all data attacks have been carried out by company insiders. Such attacks can be on a high- or low-risk level; entering a virus from inside the system or just downloading the relevant information to an USB-stick. It is important to note that damage can not only be caused by theft. Data manipulation may also lead to non-efficient processes – a relevant disadvantage in a competitive situation.

Conclusion Industry 4.0 will disrupt and evolve the steel industry. As with all new technologies, it

offers opportunities and pitfalls. The key to a successful implementation is building up the required trust levels with internal and external stakeholders, as the risks are known.

As a non-tangible good, AI requires a high level of trust between technology suppliers and steel producers. It is up to all partners to respectfully manage Industry 4.0 by implementing an adequate business culture as well as the required controls and processes. To that end, legal concerns and ethics have to be addressed right from the beginning in the R&D phase and not when AI solutions are available to the market. Personal trust and experience remain the foundation for longterm success, as steel is and can be used for robots, self-driving cars and spaceships to colonise Mars. Properly handled, intelligent processes will ensure tomorrow’s relevance of the steel industry.

Bibliography Deming, William Edwards (2000): “The New Economics for Industry, Governance, Education” EUGDPR.org (fetched 28.08.2017) FCPAmericas Blog (2015): “Highlight of Brazil’s Regulation on the Clean Company Act”: http://fcpamericas.com/english/anticorruption-compliance/highlights-brazilsregulation-clean-companies-act/# Elkins, Kathleen (2015): “Experts predict robots will take over 30% of our jobs by 2025 — and white-collar jobs aren’t immune, http://www.businessinsider.com/ experts-predict-that-one-third-of-jobs-will-bereplaced-by-robots-2015-5 Emerging Technology from arXiv (2014): “Do we Need Asimov’s Laws?”: https://www. technologyreview.com/s/527336/do-weneed-asimovs-laws/ The United States Department of Justice (2017): Foreign Corrupt Practices Act ,https://www.justice.gov/criminal-fraud/ foreign-corrupt-practices-act UK Legislation (2010): “Bribery Act”, (http:// www.legislation.gov.uk/ukpga/2010/23/ contents * PTUS Head of Governance & Compliance, Primetals USA; Regional Compliance Officer Americas

FUTURE STEEL FORUM

35 16/05/2018 10:26:54

ARTIFICIAL INTELLIGENCE

AI – translating theory into reality The focus of this article is to give top managers in steel plants some guidelines, a clear vision, and practical advice about the application of Industry 4.0 in steel manufacturing. We will explore the current perception of artificial intelligence (AI), which methods could be used to build a clear vision about this technology, and we shall offer some practical examples. By Emilio Riva*

A

lthough this application is already used in many industrial sectors such as the car industry, the benefits to the steel industry are not yet clearly defined. Based on a recent survey by Price Waterhouse Coopers of 2,000 CEOs in the metal industry there is still a lot of confusion regarding the possible applications of AI especially in two particular areas. Firstly, what exactly is Industry 4.0 for steel plants and secondly, how to actually implement the developments with the ultimate objective of value creation. From an overall industry sector view point the results so far are disappointing.

An external observer would be surprised how many articles and material are available talking about steel industry 4.0, AI, and neural network. Yet, their actual application is far from widespread. As a result, it is still not yet clear how to really apply these new technologies.

Despite these considerations, the expectations of CEOs are to reduce operational costs by 3.6 % and add additional revenue of 2.7% as a result of these new applications. Clearly these figures are substantial. There is agreement, too, that companies which fall behind in this process will soon become distanced from the first movers and are at high risk of paying the

36 steel hub.indd 2

FUTURE STEEL FORUM

price of lost competitiveness in the long term.

The principle areas of concern that represent obstacles in how to achieve these objectives are as follows: • Unclear economic benefits • Lack of digital culture • Absence of a clear digital operations strategy AI – handle with care First of all, we would like to introduce the concept that it is mandatory to keep a clear vision of the main objectives of the steel business. Taking in to consideration that every steel plant is different there is no one fit-for-all approach. Every steel plant experiences different problems, operates in different markets, and the life cycle of the plants’ installation could vary. For these reasons the technology that we will analyse requires preliminary due diligence to make sure that resources are invested wisely.

The second simple but very important concept that we want to introduce is that it is vital to start with the problem rather than the technology. The first step is not to ask where to apply certain tools that AI makes available, but to start from the production/ business problem and understand if AI can do

something to help solve the problem.

Therefore, it is a question of reversing the process and starting from the need which may not yet be fully clear to those inside the organisation. It is necessary to examine the need in the search for the solution of one or more problems and then ask what the tool can do to help resolve the issues. Undertaking the operation in reverse means that we risk letting ourselves be guided by the AI; in other words, guided by an algorithm.

Thus, AI should not be used as an oracle with an expectation of answers that will always be perfect. It is essential to understand that AI must be seen as a support for decisions that remain for humans to make. In order to support decisions, the tool must be able to show what it learned and how it learned it.

Currently there is too little focus on this point.

Some big players – given a considerable amount of data – propose to identify relationships between information sets almost completely automatically, but this strategy is yet to be proven.

The “black box” approach can also be a first step, but in order to fully exploit the potential, tools are needed that allow us to understand the main factors that lead an algorithm to perform a specific descriptive or predictive analysis, also returning an accurate

Steel Times International

16/05/2018 10:28:57

portrayal of which items are under our control and which are not.

It is absolutely necessary for managers in the steel business to govern the process rather than passively accept the results. In fact, there is no need for the end user to have any technical knowledge of the underlying models and methods, provided he is aware of the logic with which the tool provides answers and that someone has verified that the answer is produced correctly.

Possessing large amounts of heterogeneous data and feeding them into the software without an appropriate strategy still might not solve the problem.

Links could be discovered by accident that would not seem to be related logically, and the line between cause and effect

Steel Times International

steel hub.indd 3

could become blurred. We must have a strategy to formalise the problem and look for answers; to mention a practical example in the real steel plant environment. After long discussion it has been decided that to increase the production rate a new and more sophisticated IT system based on AI, and scheduling of production flow, would have been necessary to solve the problem. The discussion was held among the top managers without discussing the improvements with the local operator. After costly investment the programme was really well designed, but operators found the system too complex and, most importantly, they didn’t see any advantage.

The third aspect that we want to mention is again very simple but still vital for the

success of the implementation of these new technologies. Industry leaders need to communicate clearly the benefits of the digital transformation inside their organisations and the strong need for closer integration among the different production areas in the steel plant. The main challenge of the organisation is not buying the technology but transforming the culture, the IT infrastructure, and its employees’ working habits.

Models like AI and neural networks are extremely powerful. However, they remain a “black box” and very few people are able to understand how these models work. So, how can operators and managers rely on them?

Companies that work on the installation of these models should explain how they operate in simple terms using an explanation tool.

FUTURE STEEL FORUM

37 16/05/2018 10:29:00

ARTIFICIAL INTELLIGENCE

Fig 1

Practical examples of AI in the steel business Application of AI in the steel industry is still a “work-in-progress”, but already some trends and patterns are evolving. The fact that these technologies are still very new allows us to mention that a feasibility study needs to be performed before implementing the technology to ensure that steel plants are capable of installing the technologies in a timely manner.

There are three main areas where AI has been implemented: exploitation of a large amount of heterogeneous data, extraction/ preservation of knowledge, and support to human decision making.

The advantage of using AI is included in a number of skills such as: • Models and algorithms with selflearning capabilities • Capability of dynamically evolving • Infer relationships among variables by pointing out the factors which mostly affect a given phenomenon or system behaviour • Solve complex optimisation problems by finding the optimal trade-off among counteracting objectives.

Here below a series of examples: (Fig 1) Blast furnace consumption.

In the blast furnace process, the hearth is subject to chemical and physical erosion. A mathematical model has been created for the prediction of the refractory material thickness in the wall and bottom area of the hearth.

Final hydrogen content in stacked blooms.

Simulation of thermal patterns and

38 steel hub.indd 4

FUTURE STEEL FORUM

hydrogen diffusion in bloom/billet stacks for embrittlement avoidance. Thermal history of each bloom is simulated according to product characteristics and piling strategy.

Integrated energy management.

Development of dynamic approaches for electricity demand monitoring and timely reactions to reduce energy consumption.

Prediction of Jominy profile from chemical composition and process parameters.

Hardness predicted point-wise by means of a set of artificial neural networks (ANNs).

Exploit both plant data and knowledge on the relation between hardness at different points of the profile. Provides a measure of the confidence on the prediction for each point for that peculiar steel type. Among the achievements we can mention, high accuracy, destructive and time-consuming tests can be avoided and process parameters and can be suitably tuned (manually or automatically) in order to obtain the desired curve.

different. So, despite all the efforts very few companies are ahead in implementation.

In addition, the companies that are ahead of the field have been working on these innovations for many years to develop an adequate base platform from which to install these technologies.

An increase in the amount of data does not necessarily lead to increased insight unless you are able to process and analyse the data to obtain the valuable information.

Sometimes it is very useful to glance at other industries such as quantitative finance in which enormous amounts of data is processed by AI algorithms, but still the human factor is necessary in most of the operation.

AI and neural networks are not in themselves the final objectives. They are only the means to achieving the real objective, which is to optimise EBITDA and ROI when compared to competitors.

Another, and very important, theme to mention is data security and the implementation of rigorous methods to ensure that the IT architecture is sufficiently protected.

To conclude, considering Moore’s law, the possible implications of AI are endless.

Since the beginning of the human race there have been winners and losers but as Charles Darwin said, “It is not the strongest species that survive, nor the most intelligent, but the ones most responsive to change.”

One thing we can be sure of is that the further introduction of these technologies in the steel industry will bring considerable changes. Firms which are not able to embrace such change put their survival at high risk.

Artificial neural networks for the prediction of mean flow stress in hot rolling of steel.

The problem of the estimation of mean flow stress within a hot rolling mill plant for flat steel products is faced, as the correct estimation of this measure can improve the quality of the final product.

* CEO and founder of Steel Hub London

Final considerations.

To resume our final consideration, we noticed that many suppliers of these applications for steel plants are still working to figure out the best way to implement these technologies.

The reason is not a lack of technological capacity but, rather, because each factory is

Emilio has practical knowledge on how AI could

About the Author Emilio Riva MBA Former member of the Executive Board of Riva Group, CEO and founder Steel Hub London

benefit steel companies using his experience in the biggest integrated steel plants. In particular, where Emilio has overseen the application of AI in the blast furnace, with regard to consumption and optimisation of raw material, AI has delivered savings of Millions of Euros.

Steel Times International

16/05/2018 10:29:01

The challenge of new globalized market and the current steel market outlook characterized by plant underutilization, are the elements leading metal producers to seek for low capital investments, aiming at improving the efficiency of the production facilities, the quality of the products, the health and safety of the workers as well as the environmental sustainability.

SIMPLIFYING METALS COMPLEXITY CROSS-FUNCTIONAL BUSINESS UNIT TO DEVELOP AND IMPLEMENT NEW PLANT DESIGN CONCEPTS BASED ON DIGITAL INNOVATION

684

Process Control Systems

Process Control Systems

www.dca.it

Since 1969 Danieli Automation designed and implemented

DANIELI TEAM A CENTURY OF PARTNERSHIP EXPERIENCE

DATA MINING AND MODELLING

Data mining, modelling This article presents a robust and flexible framework that combines theoretical development and several technological solutions based on the RapidMiner software to data mining of time-series, Kasem software for knowledge management and MongoDB for data management. Finally, we demonstrate the use of the developed theory and the tools on an industrial case for prediction of the scattering of temperature in annealing furnaces.

By Amaury De Melo Souza1, David Arnu2, Fabian Temme2, Edin Klapic2, Ralf Klinkenberg2, Marcus Neuer3, Xavier Renard1, Patrick Gallinari4, Christophe Mozzati5, Claudio Mocci6, Gabriel Fricout1

40 Arcelor Mittal.indd 2

FUTURE STEEL FORUM

Steel Times International

16/05/2018 10:31:03

a

and smart manufacturing I

n recent years, state-of-the-art data mining approaches have provided invaluable insights for business. Where process experts can provide no obvious solution statistical models based on artificial intelligence techniques have gained considerable attention and nowadays they are very often used to drive the process, increase its reliability and can even be applied to predictive maintenance applications. Most of the time, in the steel industry, the object studied in data mining is the coil, and average values per coil are considered (average temperature, average rolling force, average pickling time). Besides the aggregated data that specifies each coil and its production life cycle, time-series data is largely available in steel plants due to the vast amount of sensors collecting data for process control. It is desirable, therefore, to build both methodology and tools to refine this “classic” data mining approach to exploit “time-series” that can describe more precisely the behaviour of process parameters over the length of a coil rather than only average values. In order to overcome challenges posed by the large volume of data in time-series databases, new software tools are needed from both the theoretical and computing efficiency point of view.

Methodology: Developed framework Shapelets-based approach for information discovery Information extraction from time-series is a well-known problem and several approaches have been developed in the past decades[1-5] to tackle this challenge.

Among those, the most classical approach is to compute one or many statistical quantities from the time-series (mean values, standard deviation, correlation function, Fourier transformations), and use them as input variables (or ‘global

Steel Times International

Arcelor Mittal.indd 3

features’) for machine learning tasks. In order to go beyond the classical approach of “global features” computation, we present our most recent development within the “shapelet” paradigm called “EAST” (Enumerate And Select discriminant Temporal patterns[5]).

The Shapelet algorithm (SA) is based on the presence of particular “motifs” or “patterns” (also called shapelets) in the time-series that contain valuable information regarding the process and, therefore, of the product quality. From the physics point of view, these “events” revealed by specific behaviour (for instance strong drop or rise, oscillations, or even more subtle patterns) of specific time-series, can trigger or be related to the occurrence of defects. One of the key challenges is how to identify such patterns in gigabytes of time-series. The traditional SA is based on the exhaustive search of motifs for all the time-series present in the training data set. For each metal unit and for each candidate pattern found, the distance information between the metalunit time-series and the candidate pattern is calculated. The distance is a measure of how likely there exists a sub-sequence in a given time-series of a given metal unit that is close to the candidate pattern, or how much a metal unit “contains” the specific pattern. The pattern is relevant only if it can discriminate between “good” and “bad” metal units in the training set.

Thus, if the pattern is only present in a ‘bad’ metal unit, then the presence of this pattern is a key feature for classification.

Nonetheless, the traditional SA becomes computationally prohibitive for large data sets. The EAST algorithm is a successful attempt to circumvent this problem (see Ref. 5 for a detailed description of the EAST algorithm), i.e. by combining random sampling with variable selection methods prior to the shapelet selection, while improving the performances of

FUTURE STEEL FORUM

41 16/05/2018 10:31:07

DATA MINING AND MODELLING

EAST representation Minimal Euclidean distance between each drawn candidate and every time series

Time series (training set)

Draw candidates among every subsequence

Min Eucl. distance between TS & candidates

Min Eucl. distance between TS & candidates #1

#1

#1 #2 #3

Set of candidates selection with advanced feature selection

Representation as a feature space

#2 #3 #4 ...

9.2 1.5

9.8 8.6

9.6 1.1

#2

8.5 1.3 ...

9.9 8.7

8.8 1.4

#4

...

...

#3 ...

9.2 1.5

9.8 8.6

9.6 1.1

8.5 1.3 ...

9.9 8.7 ...

8.8 1.4 ...

#4 Typical classifier training Fig 1. Schematic picture showing the discovery of discriminant patterns for time-series (TS). (1) Subsequences are enumerated from the data set; (2) the distances between subsequences and the time-series are computed; (3) a feature space is generated from the distances and (4) the discriminant set of patterns are used fed into feature selection schemes and subsequently used on classification tasks

Fig 2. Data model concepts. Various possibilities of “metal unit” and various associated process data collected in different process steps

classification tasks when compared to benchmark (“global features”) for time-series processing. Fig. 1 schematically shows how the EAST algorithm is used to transform the raw time-series information into a set of features that can be used by machine learning algorithms.

Technological Solutions: RapidMiner, KASEM and MongoDB The EAST algorithm was implemented in the data science platform RapidMiner[6], which offers a versatile framework to designing of data models for industrial cases. The server version of the platform contains a shared repository of data mining scripts and dashboard tools to visualise data and display

42 Arcelor Mittal.indd 4

FUTURE STEEL FORUM

model outputs[7]. Time-series data mining scripts have also been developed, including pre-processing, time-series transformations (derivative, filtering), feature extraction (including shapelets, and python interface), model building, model validation and crossvalidation for robustness estimates.

One key part of our framework is a smart recording of the knowledge generated during the model building process with RapidMiner. This is accomplished through knowledge management tools based onn ontologies[8], which provide a formal way to describe industrial cases, where concepts from the metallurgy or steel production process are added (e.g., crystallography, phase transform, typical process conditions

for different steel grades, steel defects).

This information can be linked to data mining concepts, such as machine learning, classification tasks; variable selection, training type, algorithm and so on. Moreover, it benefits from past studies in order to speedup the data mining process on new industrial problems. For example, past studies/models could have proven that certain variables, and their transformations and relevant features extracted, significantly improved the results from the statistical modelling. The software used for ontology management and similarity search is Kasem[9] and it was integrated with the RapidMiner platform.

In the context of time-series, the data base is really seen from a “big data” perspective, where data mining software interacts with a very large instance of the data base, in a parallel way, storing considerable additional information in the database for further re-use. The NoSQL MongoDB technology[10] is well suited for such interactions, allowing a very flexible data model (as presented in the next section) suitable for many different industrial cases.

Fig. 2 illustrates the framework developed for data mining and modelling of time-series within the context of steel production. Data coming from different parts of the production process in different formats are fed into the MongoDB database. All the data preparation and transformations (including the shapelets development) as well as model creation and deployment are handled by the RapidMiner Server that is connected to the database. The last stage is handled by the knowledge management layer, where the most important inputs and

Steel Times International

16/05/2018 10:31:10

findings are stored in Kasem tools that can be used for speeding up future analysis of similar industrial problems.

Application to industrial case Prediction of the scattering of temperature in annealing furnaces The scattering of annealing temperature for specific steel grades is known to be correlated to mechanical properties scattering. Therefore, it is highly desirable to be able to predict these scatterings that affect the quality of the product. Furthermore, the underlying physical mechanisms leading to temperature scattering are highly complex and depend on surface properties (cleanliness, emissivity), microstructure as well as on metallurgical properties that will impact the calorific capacity of the steel and, therefore, its temperature during annealing.

Surface and metallurgy can be impacted by more or less all process steps from hot-rolling to annealing; this also leads to difficulties in terms of data quality and data collection. For instance, the length of the product can change during rolling, where parts of the coil can be cut, and head and tails are switched at each process step. This has to be taken into account while treating the data.

Our strategy is two-fold, firstly, we aim at predicting the temperature scattering at the beginning of the annealing furnace in the direct flame furnace pre-heating section. A quality label is defined and “bad” products are those that show high standard deviation (STD) for the pre-heating temperature (when compared to a defined average STD). Secondly, to predict the temperature scattering in the final step of annealing, i.e., the “soaking” area, and the same strategy

Fig 3. (a) Time-series for cold-rolling speed extracted from the data set with a selected shapelet superimposed in red. (b) Distributions of shapelet distances for the population of “good” (green) and bad (red) products.

for labelling the product is used. In both cases the goal is to identify several process conditions or events that lead to a high probability of being a “bad” product. The collected variables come from three process steps: hot-rolling (rolling temperature, cooling actuators, coiling temperatures, reduction, forces, speed), pickling (pickling speed, acid bath information), cold rolling (speed, traction and strength for each stand, thickness) and galvanising (speed, different annealing temperatures). The database contains ~500 products.

Fig. 3 shows a schematic view of the data structure with all the transformations needed to take into account data phasing for the different steel making process steps.

It starts with the “metal units”, which can be, depending on the situations, heat, slabs or coils. Each metal unit evolves through different process steps, and for each of these process

steps, data are collected. These data can be either time-series and/or “static variables”.

Each process step has its own referential related to the metal unit, and each time point of each time-series can be associated to one particular position of the metal unit.

For instance, a time-series representing a temperature measurement over the length of the coil at the hot-strip mill can be stretched to match the temperature measurements made in the annealing furnace, if the reduction rate applied at the cold-rolling mill is known.

The correspondence between time-series of the different process steps can be either computed when importing the data in the database, or computed “on demand”, only if required by the “data mining” algorithms.

The focus of the present work was the quality label associated to the whole metal unit.

However, it is also possible to use continuous quality information over the metal unit lengths.

Cold rolling speed (a.u)

Bad products Good products

0

500

1000 Product length (m)

Steel Times International

Arcelor Mittal.indd 5

-1

0

1

2

3

Shapelet distance to raw time-series

FUTURE STEEL FORUM

43 16/05/2018 10:31:13

DATA MINING AND MODELLING

Modelling and results Several models with different classifiers and feature selection schemes where built using the RapidMiner platform. The data sets were split into train and test with 30% for testing. From a business point of view, the performance indicators are very important in order to compare prediction results quantitatively. In the present work, the true positive rate (TPR) at 10% of false positive rate (FPR), i.e., TPR@10%FPR, was used as performance indicator. For a RandomForest 10-fold cross-validation model, when the shapelets features extracted from the timeseries are included in the model (see Ref.[5] for details) a 10% improvement is observed when compared to the benchmark model, i.e., only “global features”.

In addition to the performances improvement, one of the most interesting added values of using the shapelet on such data sets is to highlight very specific patterns on specific variables leading to more scattered temperature, especially when the phenomena is not necessarily known by process experts beforehand. For instance, in Fig. 3(a), a time-series corresponding to cold-rolling speed is shown in blue and a particular discovered shapelet is highlighted in red. It corresponds to a strong drop in the speed before the end of the product. Fig. 3 (b) (right panel) shows the histograms of two types of product population coming from the scattering data set (“scattered” and “not scattered”) and highlights the fact that this shapelet is more frequently present in “scattered” product (smaller distances, i.e., red distribution is slightly shifted to the left) suggesting the role of the cold-rolling speed in temperature scattering in annealing furnaces.

For the prediction of scattering of the temperature in the pre-heating section, the cold-rolling mill variables were found to be the most important ones (70% of the most meaningful shapelets come from the cold-rolling time-series). On the other hand, when considering the soaking temperature at the end of annealing, the most important variables come from the hot-strip mill. This tends to illustrate the fact that variation in the cold-rolling process can modify the surface of the material. Yet, cold-rolling parameters are less important on soaking

44 Arcelor Mittal.indd 6

FUTURE STEEL FORUM

temperature scattering, probably due to the temperature regulation loop which somehow compensates for variations observed at the beginning of the furnace. Furthermore, the scattering analysis of soaking temperature highlights hot-rolling parameters as key explaining parameters (60% of the most meaningful shapelets). These insights on the model obtained through shapelets are valuable for the understanding of physical processes in steel manufacturing.

Conclusion We presented a flexible and robust framework by combining in-house theoretical developments and technological solutions for the efficient data mining of time-series data and applied the methodology to the relevant industrial problems of the steel making process. In particular, the framework includes the data collection and pre-processing steps, the specification of a generic data model in MongoDB, the development of specific and interpretable algorithms for data mining based on the shapelets paradigm, the integration of these in RapidMiner, as well as model building and comparison to the benchmark models. Practical tools for capitalising the output of data mining studies were also developed based on Kasem software. This provides a very innovative way of speeding up the new analysis while taking benefit of process expert knowledge and past results/experience in statistical modelling.

Acknowledgements The authors and partners are thankful to the EU-Research Fund for Coal and Steel for funding through the RFSR-CT-2014-00031 (PRESED) project.

and Knowledge Discovery, 15, 107, 2007.

[4] S. Zhong, J. Ghosh. HMMs and coupled HMMs for multi-channel EEG classification.

In Neural Networks. IEEE. - Proceedings of the 2002 International Joint Conference, 2, 1154, 2002.

[5] X. Renard, M. Rifqi, G. Fricout, M. Detyniecki. EAST representation: fast discovery of discriminant temporal patterns from time series. ECML/PKDD Workshop on Advanced Analytics and Learning on Temporal Data, Riva Del Garda, Italy. 2016.

[6] I. Mierswa, M. wurst, R. Klinkenberg, M. Scholz, T. Eueler et. al. Yale: Rapid prototyping for complex data mining tasks.

Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining, Philadelphia, 935, 2006.

[7] D. Arnu, E. Yaqub, M. Neuer, G. Fricout, X. Renard and P. Gallinari, A Reference Architecture for Quality Improvement in Steel Production. iDSC 2017, Salzburg 2017.

[8] A. Voisin, J-B. Léger, G. Medina-Oliva, M. Monnin, B. Lung. Health monitoring and prognostic assessment in a fleet context.

Annual Conference of the Society For Machinery Failure Prevention Technology, MFPT 2014, 2014.

[9] M. Monnin, J-B. Leger, D. Morel. KASEM: eMaintenance SOA Platform. Proceedings of 24th International Congress on Condition Monitoring and Diagnostics Engineering Management, 2017, [10] K. Banker, MongoDB in action.

Manning Publications Co., 2011.

1 ArcelorMittal Global R&D, Voie Romaine, 57280 Maizières-lès-Metz, France 2 RapidMiner GmbH, Westfalendamm 87,

References [1] Tak-chung Fu, A review on time series data mining. Enginnering Applications of Artificial Intelligence, 24, 164, 2011.

[2] B. D. Fulcher, N. S. Jones. Highly comparative feature-based time-series classification. IEEE Transactions on Knowledge and Data Engineering, 26, 3026, 2014.

[3] J. Lin, E. Keogh, L. Wei, S. Lonardi.

Experiencing SAX: a novel symbolic representation of time series. Data Mining

44141 Dortmund, Germany 3 VDEH, Sohnstraße 65, 40237 Düsseldorf, Germany 4 Université Pierre et Marie Curie, UPMC LIP6, Place Jussieu, 75252 Paris, France 5 Predict, 19 Avenue de la Forêt de Haye, 54500 Vandœuvre-lès-Nanc 6 SSSUP, Piazza Martiri della Libertà, 33, 56127 Pisa PI, Italy

Steel Times International

16/05/2018 10:31:13

AIR KNIVES & BATH EQUIPMENT

FOR CONSISTENT STRIP QUALITY CMI INDUSTRY METALS provides unique expertise and know-how in steel processing. As a market leader for galvanizing technology, CMI supplies the benchmark for many of its robust, yet innovative equipment designs for processing lines.

While these lines are designed to apply different coatings, including the very latest AluSi coatings, with the highest surface uniformity, the modular design of CMI’s state-of-the-art AIR KNIVES & BATH EQUIPMENT stand for improved and homogenous strip quality and reduced zinc consumption through most effective strip stabilization and high-performance jet wiping.

www.cmigroupe.com

Re h e a t in g f u r n ace s § Pi c k l i ng § A c i d R eg enera ti o n § C o l d Rol l i ng St r ip Pr oce ssi n g (m e ch a ni c a l , c hem i c a l & ther m a l ) § A utom a ti on DESIGN

|

ENGINEERING

|

COMMISSIONING

|

TECHNICAL

A S S I S TA N C E

&

TRAINING

|

AFTER-SALES

ARTIFICIAL INTELLIGENCE

An AI-enabled future for metals?

Applying technological innovation is something metals companies must do if they are to remain competitive, argues Dr Andrew Zoryk*

I

n 2017, one of the world’s most advanced steelmaking facilities opened in Europe.

The fully automated plant has a rolling line that exceeds 500m and is controlled by an innovative monitoring system with thousands of data acquisition sensors.

The new plant underlines the potential to transform metals manufacturing processes through “Industry X.0,” which involves harnessing advanced technologies including artificial intelligence (AI) and the Industrial Internet of Things (IIoT). But the fact that the plant is also one of the first new specialist steelworks to open in western Europe for several decades underlines something else: that much of the metals industry is still operating with aging assets that make it difficult to harness new technologies.

The smart machine revolution gathers pace Yet applying technological innovation is something metals companies must do if they’re to remain competitive, attract vital skills and retain the knowledge of an aging workforce. And the importance of innovation is further underlined by the rapid change now underway across organisations of all types, as the rising usage of AI and machine learning drives a complete reinvention of the way work is done.

Across the world, organisations are using these technologies to speed up processes, reduce costs and relieve employees of repetitive tasks. But this is only part of the story.

By implementing AI solely to save time and money, organisations risk driving away the very people they need to guide and work with these machines to achieve breakthrough

46 Accenture.indd 2

results in the future.

A handful of forward-thinking organisations across all industries have spotted this risk and are acting to address it, by creating self-adapting, self-optimising “living” processes that use machine learning algorithms and real-time data to continuously improve. In Accenture’s view, this reinvention and reimagination of processes represents a massive leap forward that will unlock entirely new roles and new ways for humans and machines to work together.

Taking the temperature of AI in metals What does it mean for metals companies?

Some are already making significant use of smart technologies like AI and machine learning: alongside the new European plant already mentioned, some leading Asian steelmakers are using the IIoT and AI to run their manufacturing processes and leveraging highly-tuned wearable technologies to ensure workplace safety.

But looking across the industry globally, such bright spots of innovation currently stand out as the exception rather than the rule.

To establish whether this situation is changing, we conducted cross-industry research among

more than 1,000 process professionals that are early adopters of AI. Zeroing in on the findings from metals companies, we see that they’re using AI to varying degrees throughout the value chain, with 71% using it in at least one business process, and over 40% using it to develop and manage products and services.

Three overlapping lenses Our research also reveals that leaders and visionaries in this space across all industries – metals included – are harnessing three interrelated dimensions of AI: • Reinventing processes: applying AI to manage process change, rethinking standardised processes as continuously adaptive and applying AI across multiple processes.

FUTURE STEEL FORUM

16/05/2018 10:32:39

• Utilising data: making use of AI and data to solve previously unsolved problems and reveal hidden patterns.

• Rethinking human-machine collaboration: shifting toward an AI-enabled culture and reskilling employees to work in alliance with machines.

Currently, only a relatively select group of metals companies – just

5% of those interviewed, half of the proportion found across all industries – are doing all three of these systematically. A closer look shows metals companies are developing capabilities in the three areas at different speeds: Looking globally, there are good examples of metals companies pursuing all three dimensions. In terms of

reimagining processes, one company is working with SAP to develop solutions in areas including process analysis, machine learning, predictive analytics and production planning. In harnessing data plus AI, another organisation is using machine learning and analysis of chemical composition and production data to optimise the consumption of materials during steel production. And in rethinking the workforce and AI, yet another company has created an analytics centre of excellence to help embrace data-driven decision-making across the business.

Barriers to overcome…

However, our finding that only 5% of metals companies are pursuing breakthroughs simultaneously in all three areas reflects the scale of the barriers to progress. Two factors in particular are impeding metals companies’ advance towards greater AI enablement.

The first is that the metals industry is still widely perceived as heavy, dirty and not environmentally-friendly, making it hard to attract innovative young talent. It’s also seen as being low-tech, which is a misconception.

In reality, the metals and steel industries are highly complex, highly automated industries where precision is paramount. But this message isn’t getting across, especially to younger people.

The other– related – barrier is the industry’s aging workforce.

Many people with the most valuable knowledge, learning and experience in their heads are around 50 to 60 years old.

So, a big challenge for the industry is how to capture and retain these employees’ learning and knowledge before they retire and take it with them. …and AI can help Applying AI and machine learning can help the industry surmount both of these barriers.

Our research highlights the opportunity for companies to empower people and machines to work together in new ways in the “missing middle”– the spectrum of human/ machine alliances and collaboration where

Accenture.indd 3

FUTURE STEEL FORUM

47 16/05/2018 10:32:41

ARTIFICIAL INTELLIGENCE

are systematically applying Al to reimagine processes and process change

each enhances and augments the capabilities of the other.

By getting experienced workers to teach smart machines, companies can capture their knowledge forever, and then apply it to help improve processes. Also, as well as humans augmenting machines, the missing middle offers scope for machines to augment humans – and here AI can help make metals a safer and more fulfilling industry to work in, by taking the strain of physically hazardous and repetitive tasks. Advances such as these will enable the industry to overcome today’s barriers and accelerate AI enablement still further.

Opportunities across the business Looking forward, as adoption gathers pace, we believe the focus of AI in metals will differ from other manufacturing industries like automotive. Why? Because automotive is essentially a discreet, short-lifecycle industry based on assembling components, its use of AI is mainly targeted at customising the end-product to customer needs by applying robotics and automation.

are harnessing data plus Al to capture exponential improvements in agility and KPIs

are rethinking how humans and machines work together

In contrast, the metals industry is more process-based, using raw materials to produce batches that are then sent on to assembly-focused industries like automotive.

This means asset availability is critical, and the biggest risk is variability of production.

So, the core focus of AI in metals will be on managing continuous production processes to ensure consistently high product quality and equipment reliability. This means using AI in areas like tracking, predicting and managing quality throughout the production process, and eliminating variability by learning from historical data.

The opportunities for AI also extend into customer service. The current processes for handling cases where customers report damaged or sub-standard product are often slow and cumbersome. AI brings the potential to completely reimagine these processes, for example by videoing the defect remotely and using intelligent machines to identify the problem through a combination of human learning and technical data. AI could also transform industry R&D for new products by using data, logic and learning to understand

aspects such as chemical compositions, accelerate testing and analysis, and shorten time-to-market.

Finally, “firefighting” has become a key skill in the metals industry. AI could reduce the need for crisis management by capturing and combining data and decades of experience to help humans make better decisions more quickly, based on reliable facts and predictions on product quality, equipment reliability, upcoming maintenance and more.

The direction of travel is clear As our research underlines, the metals industry is currently lagging behind other sectors in harnessing the power of AI. But it’s clear that the momentum of industry innovation is growing and that intelligent, self-learning technologies will play an increasing role.

If they haven’t done so already, metals companies should take urgent steps to identify where and how AI can generate the greatest value for their business. Those that fail to do this will be left behind – and could face an unwinnable race to catch up in years to come.

About the author Dr Andrew Zoryk managing director, metals Accenture @ZorykAndrew Dr Andrew Zoryk is a managing director based in Vienna, Austria, and is the integrated platform team lead for the natural resources industry group. He is also the metals practice lead for Europe. Andrew’s experience includes 30 years of work within the steel, metals and mill industries in the areas of enterprise and supply chain management, manufacturing and business operations. He can be reached at andrew. zoryk@accenture.com.

48 Accenture.indd 4

FUTURE STEEL FORUM

16/05/2018 10:32:44

Taking inclusion analysis to the next level Our exclusive portfolio and capability for advanced inclusion analysis Controlling non-metallic inclusions during the steel making process is a key skill for efficient production of modern steels demanded by today’s customers. Thermo Scientific™ Explorer™ 4 Analyzer with Metals Quality Analyzer™ software provides the most exhaustive data on nonmetallic inclusions at the lowest cost per inclusion. Thermo Scientific™ ARL iSpark™ Optical Emission Spectrometer with Spark-DAT provides the fastest access to data on non-metallic inclusions. Together, the two instruments give the most powerful solution to control non-metallic inclusions, and solve or prevent cost and quality problems caused by inclusions in steel.

Discover more at thermofisher.com/ispark © 2018 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its subsidiaries unless otherwise specified.

DIGITALISATION

Digital transformation of a steel company – first steps Digital transformation is one of the main trends of industrial developments over recent years. According to the International Data Corporation, today two thirds of corporations from the Global 2000 list put a priority on digitalisation within their corporate strategy. It is a time of changes and even those who are still talking about the impossibility of breaking technological foundations understand the obvious irreversibility of change.

By Kirill Sukovykh*

50 NLMK.indd 2

FUTURE STEEL FORUM

Steel Times International

16/05/2018 10:35:12

F

or NLMK Group, digital technologies represent another tool, which helps to maintain leadership in efficiency.

This is an opportunity to speed up changes in production and business processes. In 2017, together with SAP, NLMK created the first Co-Innovation Lab in Russia. The laboratory focuses on research, ideation and prototyping of innovative solutions not only for our company, but for the entire metals and mining industry.

Digital steelmaking Today mining corporations, such as Rio Tinto, BHP Billiton and Fortescue Metals Group, have already tuned up the unmanned trucks operations at their mines. The steelmaker POSCO is close to the establishment of a smart plant. Many Russian industrial companies have also launched major projects. With the help of digital technologies Russian companies try to predict equipment failures, to determine the required amount of raw materials with the maximum accuracy in order to get the desired parameters of finished products, to improve equipment efficiency, to track transport and employees.

There are a great many of those who try to fall into the world of digital technologies.

But it is important to bear in mind that steel companies have been shaping for decades, a certain culture, and technologies have been developed, both product range and business process have been automated. Therefore, of

course, it is not possible to change everything overnight, but everyone is moving this way applying different methods and approaches.

At the beginning of 2017 NLMK Group and SAP created a joint innovation laboratory. The laboratory team consists of only 10 people from NLMK and SAP but we can engage experts from various functional areas of NLMK and SAP consultants with the necessary skills set to help with our projects.

We do not invent computer technologies and do not produce computer equipment.

The Laboratory objective is to evaluate innovative technologies and trends – such as machine learning, Internet of Things, virtual reality – and see if they can be applied to the current business processes of NLMK Group. Thus, we get the opportunity to confirm or oppose the efficiency of innovative technologies in reality, understand the economic benefits and feasibility of various solutions. The Laboratory helps to consolidate technological experience.

Brainstorming the topic We started generating ideas during designthinking sessions. The first one took place in the SAP representative office in Potsdam.

There NLMK Group top managers proposed eight ideas including a smart warehouse and predictive maintenance. Four of those were taken to task: three machine learning and one IoT project.

The main feature of a design-thinking

session is not a critical analysis in contrast to analytical thinking, but a creative process, which triggers the most unexpected ideas.

In fact, creativity makes people take an unconventional perspective on things that seems to be quite usual. This allows them not only to create something new but also to hit on non-standard solutions applied to various problems. This feature is in high demand in any field – it does not matter if you are a writer or a manager who tries to build up desired relations with a customer. That is why, today such a creative way of thinking is one of the main business trends. Indeed, if you master this skill, you can become a successful “designer” of any business process.

Today employees of all levels including top managers and operational personnel, participate in design-thinking sessions.

Within one year more than100 ideas were developed during several sessions and workshops. At least 10 of them may lay down the foundation for innovations.

When an idea is approved, a prototype development starts and takes about six months. It cannot take any longer otherwise a lot of changes, which might be critical for the project, can take place.

Personnel positioning Creation of a prototype 3D-Employees positioning system based on SAP SCP is one of the first projects of the joint innovation laboratory. The prototype was created on

Objectives of NLMK-SAP Joint innovation laboratory

Implementation of innovative ideas.

Development of new competitive capabilities of NLMK

Steel Times International

NLMK.indd 3

Partnership with other processoriented partners, foreign mining and steelmaking companies and universities

Creation of a platform for development of SAP innovative solution strategy in mining and steelmaking industry

FUTURE STEEL FORUM

51 16/05/2018 10:35:13

DIGITALISATION

Results of design-sessions Beginning of 2017

Only top managers participate

a real Hot-Dip Galvanising Line at NLMK’s Lipetsk site. The system allows the plant to receive and analyse information on changes in equipment operation and all personnel movements in real time. Events, which are recorded and stored in the database, analyse employees’ actions and prevent accidents, reduce safety risks and improve operating efficiency.

At the beginning of the shift an employee puts a tracker into his chest pocket. This tracker can show where this employee is at any particular moment. The tracker is additionally equipped with accelerometer, help button and vibration sensor, which turns on to warn a person, if, for example, he enters a hazardous, gas contaminated place.

Many companies produce such systems nowadays, but a real production facility imposes a number of process restrictions.

First of all, in production conditions, due to shielding and radio interference, the signal of the device is unstable. Secondly, due to the low capacity of the battery, the tracker often needed recharging. Thirdly, in order to have the right positioning, we had to know where an employee was within a half-metre accuracy, which most technical solutions could not provide.

We had several design-sessions to elaborate possible scenarios of the system operation: crossing a hazardous area, activation of the help button on the

52 NLMK.indd 4

FUTURE STEEL FORUM

Beginning of 2018

Top managers and operational personnel participate

tracker, fall or loss of consciousness of an employee, methane or carbon monoxide emission detected. With the assistance of SAP specialists and the National Centre of Internet of Things, a process-oriented partner in the framework of this project, a required technology was selected - UWB (Ultra-Wideband). Positioning accuracy and resistance to the rough conditions of production premises, such as radio interference and shielding, characterise UWB. We developed a new solution using a SAP Cloud Platform, RTLS-UWB positioning system; LoRaWan 3D-visualisation and wireless data transfer technologies.

Today a number of positioning products are present on the market and this is far from being an innovation. However, we are the first who combined a RTLS positioning system and data collection from LoRaWAN environment sensors in the SAP Cloud platform with visualisation of all the data on a 3D-model in the Unity game engine.

Such a seemingly complex software and hardware package can be easily replicated and upgraded according to the needs of any division concerned.

Innovations in machine learning In 2017 NLMK Group was the first in Russia to switch over to the SAP S/4HANA, a most advanced IT-platform. It will not only

speed up the current business processes, but also implement cloud solution technologies, Internet of Things and machine learning into production without installation of any additional systems.

Currently, the SAP S/4HANA solution is integrated with more than 20 information and production systems of Russian and European companies within the NLMK Group. More than 6,500 users are working with it. On the basis of this new platform a number of projects were implemented within the NLMKSAP joint innovation laboratory.

We developed a model that compares invoices with bank statements and automates financial routines. Across the company’s foreign assets the operational accuracy of the model was 90%. But at the Russian sites the programme had to learn the word “advance payment”. In the group’s Russian divisions around 25% of all the payments are advance payments. The model under testing did not “know” about this and for each payment tried to find an open account. Such comparison was not possible, which led to mistakes. We changed the current model so that it corresponded to the needs and special aspects of the company. And now, based on linguistic analysis of the text in a bank transaction (purpose of payment), the model identifies the advance payment correctly in 95% of cases.

The same procedure was followed during implementation of the cash flow forecasting project.

We had a request from the financial division of the company, employees of which wanted to know when a contractor was going to pay a certain invoice over a onemonth period. The task was not easy, but we managed to implement it. Having analysed historical data regarding contractors’ payments in the previous periods, the model could forecast with a 90% probability when the contractor was going to pay his next invoices. Forecast accuracy of the model can be improved using more complicated types of machine learning model, for example, recurrent neural networks.

Another project implemented with machine learning is the creation of a predictive model, which can predict failure of one of the HSM 2000 units and thus significantly reduce costs of unscheduled

Steel Times International

16/05/2018 10:35:14

RFID

Wi-Fi +RFID tags

LoRa

UWB

BLE

GPS

GSM

Positioning accuracy

1-3m

1-3m

1-3m

30cm

1-2m

5m

50m

Speed of data transfer

1-3m

1-3m

300kb/s

6Mb/s

Mb/s

-

-

9km

200m-1,2km

50-100m

-

Effective range

50m-200m* 50m-200m

Mark operation time

several years several years several years

without recharge Resistance to radio

6 mon. and more

+

downtime. By means of the model it is possible to analyse the mill sensor information and make a forecast of operating conditions for the nearest time period. If the parameters are deviating from the standard, the operator gets a corresponding warning.

The existing forecasting timeframe – only seven minutes – is not enough to take the required measures and prevent a failure.

However, we confirmed the possibility of similar forecasting models building using SAP technologies in practice and determined possible ways of further development in this direction.

-

several years ~2 days

5 days

+

Chat bots and digital speech-controlled assistants, virtual and augmented reality, block chains, drones which will measure the quantity of bulk materials at the outside storages as well as drones for inventorytaking procedures, are the key lines along which the innovation laboratory plans to develop in 2018.

Today, the laboratory has only taken the first steps of the journey but even today, a number of Russian steel and non-ferrous metal companies have shown a real interest in our projects. The systems are developed in line with the most advanced technologies: support

of Internet of Things, machine learning and processing of big data files.

They will be able to solve business tasks, which used to be tough to implement due to a lack of technological development and high costs. In the near future the prototypes created by the laboratory will not only be able to cover the business processes of NLMK Group, but will also become a foundation for the development of future steelmaking.

* Co-Innovation Lab Lead, NLMK

OPTIMIZE SPRAY OPERATIONS FIND NEW WAYS TO IMPROVE STEEL QUALITY, MINIMIZE DEFECTS, REDUCE WATER USE AND MORE. HERE’S HOW WE CAN HELP: Request or download our new Spray Technology for Steel Mills Catalog and think about your daily operational challenges. You’ll see a wide range of unique services that can lead to new ways to optimize the performance of existing equipment such as impact testing, roll cooling/heat transfer analysis, descale header design and layout, spray characterization testing and more. You’ll also see an extensive line of spray product solutions that deliver superior performance for continuous casting, hot rolling, cold rolling, ironmaking and steelmaking.

Visit spray.com/steelcatalog OUR CATALOG IS JUST THE STARTING POINT; TALKING WITH YOUR LOCAL STEEL SPECIALIST IS WHEN OPTIMIZATION BEGINS. AD_Steel-Catalog_Steel conference Poland_80x185mm.indd 1

Steel Times International

NLMK.indd 5

07.02.2018 12:15:54

FUTURE STEEL FORUM

53 16/05/2018 10:35:15

SOCIAL PRODUCT DEVELOPMENT

Open innovation and SPD

The pressure to innovate in the current business climate is exceptionally high[1]. Competitive advantage through incremental improvement alone is no longer possible and organisations must consider new and disruptive approaches to product development in order to enhance their market position[2]. By Professor Dirk Schaefer* and Hannah L Forbes*

O

ne approach that offers the opportunity for innovation in the current climate is social product development (SPD) which is defined as a “the use of social computing technologies, tools, and media, influencing the product life cycle at any stage”[3]. Since it is a term that describes several phenomena, a thorough understanding of SPD can only be gained from understanding each of its tenets and how they are united under this new approach. In this article, each tenet of SPD will be defined and their role within SPD will then be described. Guidance for the industrial application of these phenomena will then be provided. The key tenets of SPD are mass collaboration, crowdfunding, crowdsourcing, cloud-based design and manufacture (CBDM) and Open Innovation.

54 Dirk S.indd 2

FUTURE STEEL FORUM

Mass Collaboration: A Form of SPD Mass collaboration is defined as a “form of collective action that occurs when large numbers of people work independently on a single project, often modular in its nature”[3]. An example of mass collaboration is Wikipedia, where individuals use their own knowledge and expertise to contribute to a larger online encyclopaedia. The optimal result of mass collaboration is that the overall project is completed to a higher quality because individuals are able to focus on areas of the project where they can offer the most value.

Mass collaboration can itself be described as an approach or a “way of working”. If mass collaboration is conducted, then it is integrated into all parts of the product development process and every

phase of product development is conducted collaboratively. “Any endeavour where large amounts of people come together to solve a problem or contribute to product development would be deemed social product development”[3], therefore mass collaboration is a form of social product development. It should be understood, however, that not all social product development involves mass collaboration[9].

The application of mass collaboration in industry varies according to the project.

There are, however, several principles of mass collaboration that promote effective application. These are as follows: 1) Accessibility: In order for all contributors to participate effectively, access to relevant aspects of the project must be arranged. For example, should the project

Steel Times International

16/05/2018 10:36:49

involve a centralised software system such as Wikipedia, all collaborators must have constant and easy access to the system.

2) Role and task definition: To ensure efficient collaboration, all contributors must understand their specific role, understand the tasks they need to complete and understand how their work interacts with the wider aims of the project.

3) Communication: As with all forms of collaboration, clear communication channels must be constructed and maintained. With the definition of the task and the roles, the dependent roles and tasks must also be defined. Each role should be able to clearly communicate with its dependent roles and vice versa.

When planning a mass collaboration project, these three principles should first be considered. Aspects of the individual project that influence how effective collaboration can take place, should then be considered.

The tools of SPD Unlike mass collaboration, other tenets of SPD are not necessarily integrated throughout the entire product development process.

Crowdfunding, crowdsourcing and cloudbased design and manufacture (CBDM) are applied as tools as part of SPD. As a consequence, the entire product development process does not need to be organised to

include these tenets, they can instead be employed, when needed, during relevant design phases.

Crowdfunding is defined “as the process of taking a project or business, in need of investment, and asking a large group of people to supply this investment” [6]. Four models of crowdfunding exist and they each have various corresponding crowdfunding platforms. See Table 1 below: Crowdfunding is a tool that incorporates the power of the crowd to fuel the commercialisation process. Unlike traditional investment models, crowdfunding allows organisations to gauge customer interest, gather product feedback and entice early adopters, all before launching a product to market. It is a tenet of social product development because through launching a crowdfunding campaign, organisations can gain valuable insights for the improvement of their product’s design.

Furthermore, by receiving funding, new methods for mass manufacture may be accessible, new materials may be accessible and phases of the product development process may be repeated.

Crowdsourcing is defined as “the act of taking a job, traditionally performed by a designated agent [. . . ] and outsourcing it to a [. . . ] large group of people” [11]. One of the most famous examples of crowdsourcing is Procter and Gamble’s “Connect and Develop” which allows the organisation to “partner with the world’s most innovative

Model

Definition

Example Platforms

Donation-based

Contributions are made with no expectation of any return

Just Giving, Go Fund Me

Lending-based

Peer-to-peer loans comprised of contributions from a large number of people

Rate Setter, The Funding Circle

Reward-based

Contributions are made in return for a gift or a product prototype

Kickstarter, Indiegogo, Crowdfunder UK

Equity-based

Contributions are made in return for a percentage stake in the business

Seedrs and Crowdcube

Table 1: The Four Models of Crowdfunding and Their Platforms

Crowdsourcing Initiative

Definition

Crowdsourcing contests

A contest designer poses challenge problems for the crowd. Judgement criteria and prizes available are clearly advertised E.g. Gold Corp

Open calls with direct rewards

Tasks in this class are broader. Judgement criteria not clearly advertised. E.g. Quirky

Open calls with indirect benefits

Contributors benefit indirectly from the company’s implementation of the ideas in their products E.g. Connect & Develop by P&G

Micro-tasks or Human Intelligence Tasks (HITs)

Easy for humans to accomplish but difficult to automate. E.g. Amazon Mechanical Turk

Table 2: Crowdsourcing Initiatives

Steel Times International

Dirk S.indd 3

FUTURE STEEL FORUM

55 16/05/2018 10:36:51

SOCIAL PRODUCT DEVELOPMENT

minds” by encouraging the crowd to submit product ideas and suggestions[5]. The different forms of crowdsourcing, or crowdsourcing initiatives, have been defined by Panchal[8] and are summarised in Table 2.

Crowdsourcing is most commonly used in the ideation and concept evaluation phases of the product development process. Its benefit lies in the ability to access a large number of individuals each with varying perspectives, backgrounds and cultures.

As Howe[7] states “a randomly selected collection of problem solvers outperforms a collection of the best individual problem solvers”. To apply crowdsourcing as part of a product development process, one of the initiatives outlined above should be selected based on various characteristics of the task at hand. Fig. 1 illustrates this.

In social product development, crowdsourcing is the tool that allows an organisation to incorporate the power of the crowd during development. While crowdfunding, engages the crowd after detailed design, crowdsourcing can be used in requirements analysis, ideation and concept evaluation.

Crowdsourcing ensures diverse perspectives are drawn from throughout product development and works as a driver for innovation.

Cloud-based design and manufacture (CBDM) is “a service-oriented networked product development model in which service consumers are enabled to configure, select, and utilise customised product realisation resources and services ranging from computer-aided engineering (CAE) software to reconfigurable manufacturing systems”[12].

The term, therefore, summarises the online software available to support organisations throughout the product development process.

Fig. 2 shows some examples of CBDM services[12].

By using these cloud-based services, as opposed to local software, the product development process is more easily open for collaboration. Files can be stored and accessed worldwide from a centralised location, remote collaborators can work together in real-time and manufacturing processes can be initiated remotely. Overall, the product development process becomes more accessible and more efficient.

56 Dirk S.indd 4

FUTURE STEEL FORUM

On average, how long does the task take a human to complete?

Is a tangible reward offered?

E.g. a financial reward

>1 minute

Yes

< 1 minute Initiative 1: Human intelligence tasks (HITs)

Initiative 2: Crowdsourcing contests

Initiative 3: Open Call with direct rewards

Prepare project brief with clear and advertised judgement criteria. Construct platform for uploading contributions

Construct platform for uploading contributions. Specify how contributions will be used and the associated rewards

Robots

Negotiation platform Social platform

ERP Simulation

Data base

PaaS consumer

Controller

Search Engine IaaS consumer

Knowledge Base Integration

CRM CAPP

CAE

Knowledge Negotiation Management Mechanism System Application deployment

SaaS consumer

CAD/ CAM

3D printers Machine components

Manufacturing cells Fixtures/Jigs

Ongoing

Set time

Fig1. Choosing a crowdsourcing initiative

HaaS consumer

Construct platform for uploading contributions.

Specify how contributions will be used.

Must the period for contributions be set or can it be ongoing?

Construct centralised platform that allows participants to access task and contribute results

Machine tools

Initiative 4: Open Call with indirect rewards

No

Network equipment

Transportation & supply chain

Manufacturing facility

Servers

Storage

Development Testing

Fig 2. Examples of CBDM Services

2. Updates model

4. Resolves conflicts before updating

Arrow represents sending file to corresponding engineer

Local CAD package

Engineer 1

Local CAD package

Engineer 2

Local CAD package

Engineer 3

Fig 3. Collaborating on a CAD model without cloud-based software

Steel Times International

16/05/2018 10:36:53

CBDM is a tenet of SPD because effective collaboration relies on CBDM services. For example, should a group of geographically dislocated design engineers look to collaborate on a CAD model, without a cloud-based CAD service, the process would look like Fig. 3.

If a cloud-based CAD service is employed, the collaborative process would look like Fig. 4.

Fig. 4 illustrates a more effective form of collaboration and a streamlined process.

Should an organisation look to involve geographically dislocated stakeholders in a collaborative process, CBDM services should be employed where possible throughout the product development process.

Open innovation The next tenet of social product development is open innovation. Open innovation is defined by Trott et al.[10] as a term “used to promote an information age mindset towards innovation”. This mindset encourages the sharing of data and knowledge with those external to the organisation. Understanding of the term is often promoted through comparison to the traditional mindset known as “closed innovation”. For example, Chesbrough[4] states that “the open innovation paradigm can be understood as the antithesis of the traditional vertical integration model in which internal innovation activities lead to internally developed products and services that are then distributed by the firm”. Open

Engineers can work simultaneously and can see changes in real time.

Engineers can communicate onteh platform to minimise and resolve conflicts as they happen

Fig 4. Collaborative process with cloud-based CAD package

Up-to-date model

Cloud-based CAD package

Cloud-based CAD package

Cloud-based CAD package

Engineer 1

Engineer 2

Engineer 3

What exactly do we need to/can we share Who do we need it from? E.g. Hobbyists? Professionals?

What do we need?

How are the contributions best communicate to us? E.g. Via email? Via a public forum?

OUTFLOW

INFLOW

Fig 5. Constructing knowledge flows for Open Innovation

Steel Times International

Dirk S.indd 5

How do we reach them?

E.g. On our website? Via social media?

In what form do we need the contributions? E.g. On paper? On a particular piece of software?

Innovation can be described, in relation to SPD, as an environment or climate that allows SPD to be fostered. For example, without adopting the mindset of open innovation, external collaborators could not be involved in mass collaboration. Furthermore, investment from crowdfunding could not be gained unless the organisation is willing to share their project with the external crowd.

When it comes to adopting open innovation in industry, Chesbrough[4] assists by describing the model of open innovation.

He states that open innovation, beyond a mindset is, “the use of purposive inflows and outflows of knowledge to accelerate internal innovation and expand the markets for external use of innovation”. Focus must, therefore, be on creating these “purposive inflows and outflows of knowledge” to foster Open Innovation. A simplified example of the process of constructing these knowledge flows is shown in Fig.5.

In order to foster open innovation, an organisation must first replicate the mindset of open innovation and adopt a culture that appreciates the value of open data and shared knowledge. The organisation must then implement the model of open innovation by constructing the required knowledge flows.

Conclusion Having described each tenet, the overall concept of SPD can now be presented by recognising the commonalities between the tenets. Firstly, SPD encourages innovation by recognising the value of ensuring product development is accessible to the masses. Mass collaboration is an approach that allows individuals, regardless of location, to employ their expertise in product development. Crowdfunding and crowdsourcing ensure that the crowd, comprised of all backgrounds, are able to participate and gain from product development. Open innovation is a movement encouraging open access to knowledge. Through CBDM, manufacturing capability has been brought to the masses by making the processes more accessible through reduced cost and increased production flexibility. Additionally, SPD appreciates the value of employing the

FUTURE STEEL FORUM

57 16/05/2018 10:36:54

SOCIAL PRODUCT DEVELOPMENT

crowd by tapping into the power of diversity and the power of numbers. Finally, each of these tenets are united by the idea that collaborative action, whether that be collaboration of the masses or collaboration of select individuals, results in better results for all. To apply social product development, organisations should aim first and foremost to adopt the open innovation culture. New projects should then be structured to allow mass collaboration and tools such as CBDM, crowdfunding and crowdsourcing to be incrementally employed. Organisations that complete this process will have the competitive advantage of obtaining innovative ideas, creating innovative products and then commercialising them in an innovative way.

References [1] Brown, S. L., & Eisenhardt, K. M. (1995). Product development: Past research, present findings, and future directions. Academy of management review, 20(2), 343-378.

[2] Bertoni, M., Eres, H., & Scanlan, J. (2014). Co-creation in complex supply chains: the benefits of a Value Driven Design approach. In Product Development in the Socio-sphere (pp. 3562). Springer, Cham.

[3] Cress, U., Moskaliuk, J., & Jeong, H. (Eds.). (2016). Mass collaboration and education (Vol. 16). New York, NY: Springer.

[4] Chesbrough, H. W. (2006). Open innovation: The new imperative for creating and profiting from technology. Harvard Business Press.

[5] Dodgson, M., Gann, D., & Salter, A. (2006). The role of technology in the shift towards open innovation: the case of Procter & Gamble. R&D Management, 36(3), 333-346.

[6] Forbes, H., & Schaefer, D. (2017). Social product development: The democratisation of design, manufacture and innovation. Procedia CIRP, 60, 404-409.

[7] Howe, J. (2008). Crowdsourcing: How the power of the crowd is driving the future of business. Random House.

[8] Panchal, J. H. (2015, July). Using Crowds in Engineering Design–Towards a Holistic Framework. In 2015 International Conference on Engineering Design, Design Society, Milan, Italy, July (pp. 27-30).

[9] Peterson, A., & Schaefer, D. (2014). Social product development: introduction, overview, and current status. In Product Development in the Socio-sphere (pp. 1-33). Springer, Cham.

[10] Trott, P., & Hartmann, D. A. P. (2009). Why’open innovation’is old wine in new bottles. International Journal of Innovation Management, 13(04), 715-736.

[11] Unterberg, B. (2012, February). Kapitel 10 Crowdsourcing (Jeff Howe). In Social Media Handbuch (pp. 134-149). Nomos Verlagsgesellschaft GmbH & Co. KG. [14] Wu, D., Thames, J. L., Rosen, D. W., & Schaefer, D. (2012, August). Towards a cloud-based design and manufacturing paradigm: looking backward, looking forward. In ASME 2012 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference (pp. 315328). American Society of Mechanical Engineers. * Division of Industrial Design, University of Liverpool, UK

58 Dirk S.indd 6

FUTURE STEEL FORUM

Steel Times International

16/05/2018 10:36:54

We drive industry.

TMEIC delivers. In fact, we’ve been a leading force in the metals industry for more than 50 years and the preferred partner for most of the recent hot mills in the world. TMEIC offers: • High-quality AC and DC variable speed drive systems • Complete automation solutions • Global services including system test, commissioning, plant service, and upgrades

www.tmeic.com +1-540-283-2100 metals@tmeic.com Roanoke, Virginia USA

MES SYSTEMS

Hi-tech steel production planning SMS group, the global leading partner for the metals industry, sees MES systems as one of the most attractive fields of application with a high improvement potential through digitalisation. By Wilfried Runde*, Michael Bruns**

D

igital transformation is making everything faster, more personalised, and more efficient. SMS group, the global leading partner for the metals industry, sees MES systems as one of the most attractive fields of application area with a high improvement potential through digitalisation. The optimisation opportunities still offer room for improvement. In particular, seamless vertical and horizontal networking, optimisation algorithms, and data-driven models are becoming crucial in providing the increasing flexibility in production planning without the negative impact on productivity and operational stability. Nonetheless, technological know-how in metallurgy continues to be indispensable, and it is now more important than ever that such domain knowledge is integrated into dynamicallygrowing MES functionalities easily and intuitively. After all, the dynamic changes in market demands, triggered by IoT, require considerably shorter cycles of automation modernisation as we head towards Industry 4.0. The growing amount of data and the increasing information exchange capability is resulting in closer co-operation between plant and automation suppliers, such as SMS group, and plant managers. The present paper shows successful digitalisation applications for greenfield – and even more challenging – brownfield projects with the SMS group’s new manufacturing execution system X-Pact® MES 4.0.

New challenges in steel production Many challenges must be successfully overcome during the production of steel. The flexible customisation of products must be

60 SMS.indd 2

FUTURE STEEL FORUM

achieved while ensuring a consistently high level of product quality. Production lead times must be minimised and high delivery punctuality levels must be maintained. At the same time, resources must be optimally utilised with reduced stock, and maximum yield must be attained to ensure profitability.

In addition, the requirements of end customers with regard to special properties and “tailor made products” are increasing, while lot sizes are tending to become smaller and smaller.

During the production process a large number of set-up values for the various process steps are created to optimise the product. In addition to these values there are a lot of measured values and signal information. Around 15,000 signals are available in a modern CSP plant. When analysing this signal information, it is easy to become confused, especially when you consider that there might be 500 or more temperature values (measured, calculated, surface, and average) from the heat, through casting and rolling, right up to cooling.

Knowing the origin of the signals and the way they are created is of vital importance during analysis and thus a crucial factor in long-term production optimisation.

Manufacturing execution system X-Pact MES 4.0 is the name of SMS group’s new integrated, modularised planning and management system, which is suitable for all types of production plants in the metallurgical process chain. By providing an extremely wide gamut of customisable functionalities for all these plants, this manufacturing execution system plays a salient role within the X-Pact

electrical and automation systems portfolio.

X-Pact MES 4.0 can be used in meltshops, casting plants, flat rolling mills, tube and pipe mills, section mills and strip processing lines.

Against the backdrop of advancing digitalisation in all areas of metallurgical plants, our clearly structured and future-proof X-Pact MES 4.0 gives plant operators an economic advantage in terms of productivity.

The cutting-edge software architecture and the modular, expandable design of X-Pact® MES 4.0 allow plant operators to put digitalisation on the right track to Industry 4.0 from the very beginning. As an industry partner, SMS group offers solutions to pressing issues that include: • How can growing product individualisation be achieved while ensuring consistently high quality?

• How can production lead times be minimised while guaranteeing timely delivery?

• How can resources be used most efficiently and inventories reduced?

A growing number of plant operators are realising the need for change towards an Industry 4.0 environment. And this is where X-Pact MES 4.0 plays a vital role.

Steel Times International

16/05/2018 10:39:50

module category will be highly influenced by new data-driven models from ongoing development activities.

Support and optimisation These modules bring together additional factors that are vital for an efficient planning process, for example the plant condition, product quality, and energy consumption.

The target is to achieve a situation in which all production equipment and plant units along the entire process chain are vertically networked in the most effective way. Additionally, these modules provide solutions for the central administration and maintenance of master and production data.

The warehouse management module keeps track of inventories and stored products along the whole production line. With the integrated barcode scanning feature, the quality of stored items can be swiftly and efficiently documented on a smart phone.

Data collected from all sensors: ~3 terabyte per month

Production volume: ~ 1.5 million tons per year

X-pact ® MES 4.0 - Level 3 Nr. of users: 150 Different use roles: 20

Nr. of CSP® sensor signals: ~15.000

Pulpit stations/HMIs: 12 System dialoges: ~100

Nr. of tables in use per L2 automation: ~350

X-pact ® Warehouse Manager Nr. of users: 110 Coil yard, 6 areas, 11 sections Stock level at hand: ~3.700 coils 5 Tablets, 10 handhelds, office 3D-HMIs

Customer orders: ~10.000 per year

X-pact ® Business intelligence Nr. of users: 21 Nr. of reports: 18 Materials processed: ~70.000 coils per year

Average order size: ~70 tons/order

Data load: ~7GB, data points: ~3.500 Planer: 40 sessions per month (~80 hours)

Fig 1. Key figures for Big River Steel in Osceola, Arkansas, USA

The functional modules are clearly divided into the following categories: • Planning • Support and optimisation • Supply and dispatch • Reporting • Basis Planning The planning modules form the centrepiece

Steel Times International

SMS.indd 3

of X-Pact MES 4.0. They generate technical orders on the basis of customer orders and set up optimised capacity plans, time schedules and production sequences. A highlight of the system is its dynamic rescheduling capability.

This means it automatically responds to any disruptions in the production process that may have an impact on the planning. The system thereby relieves the operators in the control room of this critical task. The planning

Supply and dispatch These modules provide the basis for a horizontally networked process chain along the lines of Industry 4.0 by integrating both suppliers and customers throughout the entire production process. This facilitates material supplies and requests for equipment at shorter notice, speeds up the dispatch of products, and achieves significant cost savings from these effects. Customers can obtain specific information about the progress of their production orders.

Reporting This is where the cycle ends. For reporting and shortening the time-to-decision, a series of business intelligence modules are provided – web-based and using the latest media. The wide ranging reporting options include everything from conventional reports, the display of key production parameters on smart phones, through interactive analyses of production data. In this way, data become valuable information that may be used to further improve the efficiency of planning and production processes. SMS group’s solution is called X-Pact Business Intelligence.

In addition to the creation of conventional web-based reports, it also provides an investigation of connections between data.

Interactive data evaluation is the key to business expertise and fast decision making

FUTURE STEEL FORUM

61 16/05/2018 10:39:51

MES SYSTEMS

in an organisation, and the resulting business processes are able to promote sustainable cost reductions.

Order 2

Order 3

Basis SMS group’s basis modules provide a general overview of all processes within the production facility, while guaranteeing maximum data security and data protection.

The data can be made available across various locations via a dedicated enterprise cloud. X-Pact MES 4.0 accompanies plant operators all the way from order intake to the dispatch of the product.

Cmax

X-Pact MES 4.0 at Big River Steel – the learning steel mill As a systems supplier, SMS group supplied all the plants and process know-how for the steel complex at Big River Steel, Osceola in the United States, and has supported Big River Steel during commissioning. Since operations were started, Big River Steel has achieved a steep run-up curve in hot strip production.

Covering a site of 567 hectares, Big River Steel is North America’s most modern steelworks. In the first construction stage, the plant has a designed capacity of 1.6 MT/yr.

The X-Pact electrical and automation systems solution was applied in all process stages, from steelmaking through to the strip-processing plant. It ensured the steelmaking complex went into production right on schedule in December 2016, and subsequently achieved a steep run-up curve.

Big River Steel uses the X-Pact® MES 4.0, which immediately took over production planning and control.

Big River Steel is mastering flexible production planning with varying and partly small batch sizes, while adhering to deadlines supported by this set-up, with SMS group’s approach to digitalisation and the learning steel mill. It ensures intelligent, largely autonomous steel production. This involves the interconnection and collaboration of humans and machines across the entire value chain in dynamic production processes that adjust to optimum parameters in real time.

Digitalisation is not an end in itself; rather it is a means to further increasing productivity and thus profitability, and to increasing the flexibility and resilience of the installed value chain.

Fig 2. Flexible chemical target analysis, case 1

62 SMS.indd 4

Order 1 (slab/coil)

FUTURE STEEL FORUM

Order 4

Order 5

Order 6

Ctarget Cmax

Heat 1

Ctarget

Cmin

Cmin

C represents a chemical element of heat analysis

Order 1

Displayed are maximum, minimum and target of order analysis

Order 2

Order 3

Order 4

Order 5

Order 6

Cmax Ctarget

Cmax

Ctarget

Heat 2

Heat 1

Ctarget

Cmin

Cmin

Fig 3. Flexible chemical target analysis, case 2

Material handling

DRI plant

scrap yard

EAF 1

EAF 2

EAF 3

Power/material request based on analysis and process data

Fluxes/Alloys LF1

CCM1

Charge mix calculation, material requests, based on availability and quality

CCM2

LF2

Power/material request based on analysis and process data

CCM3

Cut billet request and quality prediction based on process data

Fig 4. X-Pact® MES 4.0 at HADEED SAUDI IRON & STEEL

Steel Times International

16/05/2018 10:39:52

Highlight: PQA® (Product Quality Analyser) To ensure top quality, Big River Steel uses the SMS group’s PQA system. The PQA system collects and evaluates quality data for all products produced in all steps as well as quality-relevant process parameters along the entire process chain, from steelmaking to the finished product. Here, the process parameters may comprise measured values and results, but also complex criteria applied for quality analysis. Starting from quality order dressing, where the customer requirements are translated into detailed product quality targets, including tolerance values, the X-Pact MES 4.0 also determines the sampling required for minimum yield losses. So the “make-to-order” philosophy is supported to a significant degree by taking the capability of the plant into account. As one example of this, the so-called flexible target analysis for heat steel grades was developed.

Highlight: Flexible chemical target analysis The flexible chemical target analysis function is used at Big River Steel to minimise transition bars in the caster. This minimisation of transition bars offers a huge advantage in terms of yield and, therefore, an increase in production.

In the first example below, there are six orders (thin slabs/coils) which belong to one heat. So the chemical elements ‘min’ and ‘max’ must suit all orders. The target analysis of the heat is calculated while taking all single targets and limits into account.

The next example shows another idea. The heats ‘min’, ‘max’ and ‘target’ are calculated for each heat itself. The problem is that the first target is above the second maximum of the current heat. The solution to this issue is the following: for the first heat, the nearest limit of the second heat is taken. These are just two examples of solutions implemented by Big River Steel and SMS group.

The “flexible chemical target analysis” feature is realised in the SMS group production planning system together with the EAF, LMF, RH degasser and caster process models. This feature enabled the number of transition slabs to be reduced from 5 % to 2%.

According to Big River Steel, this saves them several millions of dollars each year.

Steel Times International

SMS.indd 5

X-Pact MES 4.0 at Hadeed Saudi Iron & Steel – upgraded level 3 system As part of a turnkey project, the level 3 production planning system for three electric arc furnaces, two ladle furnaces and two of the three continuous casters, as well as the maintenance section with equipment management system, were fully replaced and prepared for Industry 4.0 with the X-Pact MES 4.0 at Hadeed Saudi Iron and Steel.

The implemented solution is scalable, so further progressive digital enhancements can be accommodated.

as great flexibility in spare parts sourcing.

As a result, a comprehensive asset life cycle optimisation with attractive ROI was achieved with this revamp.

Highlight: X-Pact Business Intelligence The core elements also included a business intelligence system with interactive analysis options and a highly extensive, modern web reporting system that provides very clearly structured and comprehensive dashboard visualisation of the production processes and the use of associated inputs, in addition

Fig 5. Intelligent and up-to-date reporting with X-Pact® Business Intelligence

The key question about the modernisation project at Hadeed Saudi Iron and Steel was how to prepare a steel plant with legacy equipment that partly dated back to the 1990s for Industry 4.0, while keeping downtimes to an absolute minimum. Within a short time SMS group had already implemented a state-of-the-art automation infrastructure. The new intelligent analysis tools add leverage to Hadeed’s experience in production management, quality control, and business intelligence, including the perfect integration of pre-existing data into the new fault-tolerant database. Through collaboration between the metallurgists from Hadeed and SMS group, a wealth of empirical experience was fed into the new self-learning process models. This has resulted in significant cost optimisation.

To ensure future-proof, recurring returns on investment, SMS group supplied virtualisation techniques to give Hadeed independence from specific hardware components as well

to processing the information to the desired level of detail. The challenge was to channel the enormous amount of detailed data into relevant information packages at one central point for intuitive analysis and strategy development.

The central best practice management facilities allow Hadeed to develop its metallurgical strategies centrally and apply them to the different furnaces. This modernisation has allowed Hadeed not only to replace the old hardware and system software with new technology, but also obtain a sustainable platform that is scalable for the integration of innovative modules in connection with Industry 4.0. Such modules can be easily added, and the system’s performance is progressively increased with every new solution module. The strategic KPIs can now be deployed across all operator teams, with data dashboards allowing close monitoring and follow-up of targets.

FUTURE STEEL FORUM

63 16/05/2018 10:39:53

MES SYSTEMS

Highlight: Melt shop pacing and dynamic re-planning for caster plant (MSP) The core function of the melt shop pacing and dynamic re-planning system is to keep the original planned caster sequence as long as possible, even when some deviations or unforeseen variations occur during production. The system is used by the caster operators and/or dispatchers in the pulpits for the initial selection and timing of used equipment at the steel making plant based on the planned casting sequences. It enables an optimised routing of heats and determination of planned treatment times, continuous monitoring of the actual and planned treatment times as well as a rerouting of heats in case of considerable delays or unforeseen plant unit stoppages. There are also unplanned production events that will require changes to the production schedule.

Dynamic re-planning and the availability of an order list mean that the operator can: • increase the production to order (reduce production to stock); • increase the production in quality (operator may change the billet orders if heat

Fig 6. Self-learning production planning

grade has not been achieved at steel plant); and • increase the production efficiency of the plant.

The selection of equipment and accurate time planning for all units in the steel plant are done automatically by melt shop pacing, used to support the steel plant supervisor (dispatcher). The system checks for free production units and generates a time schedule for the production steps of steel making plant (EAF, LF, CCM). In re-planning the supervisor has the option of manually moving the treatment of heats to another plant unit (e.g. from LF1 to LF2).

Furthermore, the supervisor can change the treatment start time and duration. The Gantt chart is automatically updated at regular intervals in all pulpits, taking the latest changes to the schedule by the supervisor into consideration. Mathematical methods are utilised to calculate the start and end times of the work operations of the heat orders for the various plants on the process route and also to calculate the routing alternatives within the steel making plant. Last but not

Process

Production

Process

SMS.indd 6

Production

optimisation

Process

optimisation

FUTURE STEEL FORUM

Production

optimisation

optimisation

Process

optimisation

Slab

64

Real-time production planning Today’s generally centralised and rigid medium to long-term production planning will be replaced in the learning steel mill in future with real-time, self-optimising production planning. At Big River Steel, the X-Pact MES 4.0 system reviews current planning whenever a change takes place (new order, altered maintenance plan, quality deviation). It does so in real time and applies clearly defined key performance indicators while searching for a better planning result. That involves taking into account real time data and, therefore, factbased views on the product quality and plant condition, as well as a structural production planning model. In short, X-Pact MES 4.0 optimises production planning dynamically in respect of four main key performance indicators: output, delivery performance, product in stock, and cost efficiency.

Machine learning methods with artificial intelligence

Production optimisation

optimisation

least, it is possible to manage the scheduled maintenance operations by indicating an approximate start time for a maintenance operation. Optimisation algorithms consider this flexibility in their calculations and provide the (fixed) planned start time.

Process

optimisation

HRC

optimisation

CRC

Processed coil

Steel Times International

16/05/2018 10:39:53

We understand how important it is to find the right expertise for your industry application needs.

knowledge +â&#x20AC;&#x2030;know-how You are assured to get the best-fit products, solutions and services for your specific requirements.

Customers around the world trust us when it comes to process automation. Our shared goal is plant safety, availability and efficiency. We are with you every day, everywhere.

People for Process Automation

Do you want to learn more? www.endress.com/en

Untitled-1 1

02.05.2018 08:48:35

MES SYSTEMS

As part of its digitalisation offensive, SMS group has launched a research and development project on intelligent production planning in co-operation with Jacobs University of Bremen, Germany. The project will cover aspects such as dynamic reactions to specific production situations, the use of artificial intelligence, and autonomous learning of automation systems.

The dynamic planning and optimisation processes to be developed during the project will be integrated into the automation environment in place at Big River Steel. Improved adherence to production schedules and increased yield by reducing downgrading and scrap will have positive effects on the economic efficiency of the customer’s production facilities.

Machine learning and pattern recognition techniques are to be introduced to predict the timeliness of orders.

A further objective of the project is the development of a planning module based on artificial intelligence - X-Pact MES 4.0 performance enrichment analysis. This module will be used to detect relationships between production parameters and performance indicators on the basis of historic production data.

These capabilities are intended to be used, for example, to perform scalable, self-learning order analyses and generate plans that take order schedules into account.

Safetyfittings for oxygen lances � Lance-holders � Slag return safety devices � Quick shut-valves � Oxygen safety hoses � (glas fibre, metal braiding) � Carbon injection lances EAF � Safety hose reels � Argon Bubbling Equipment � 600 °C heat resist. on ladles � Safety devices for � Oxy-fuel burners EAF

Conclusion SMS group has been dealing with digitalisation not only under technological aspects, but also it impacts the integral processes of a company and has fundamental effects on existing processes and business models. To SMS group it is more an evolution than a revolution.

According to SMS group, digitalisation is set to offer entirely new opportunities for both steel producers and plant manufacturers. A key feature in this area is X-Pact MES 4.0. In times such as these where an exchange of information can be very easily achieved, product added value can be generated. Much more intensive co-operation between the plant suppliers and their customers is possible, and so the implementation of digital solutions is commercially rewarding for both sides.

To put it briefly, the clearly structured and future-proof X-Pact MES 4.0 gives plant operators an economic advantage in terms of productivity. The MES system is linked with the automation systems of all units in the plant and obtains status information as well as production and process data. * Production planning systems at SMS group, Germany. ** EA direct business and EA services, SMS group, Germany. Contact: wilfried.runde@sms-group.com

highest quality highest safety BEDA

SMS.indd 8

®

Oxygentechnik

Armaturen

GmbH

Germany � An der Pönt 59 � D-40885 Ratingen Tel. + 49 (0) 21 02-910 920 � Fax + 49 (0) 21 02-1 77 42 E-Mail: info@beda.com � Internet:http://www.BEDA.com

Steel Times International

16/05/2018 10:39:54

INNOVATIVE TECHNOLOGIES

FOR THE METALS INDUSTRY CMI INDUSTRY METALS designs, installs, modernises and services installations and machinery for the steel and non-ferrous industries throughout their whole life cycle in order to improve the economic, technological and environmental performance of this equipment. CMI’s clients benefit from: a unique combination of expertise in engineering, maintenance and the management of international technical projects, a vast geographic and technological scope, and an ability to innovate in accordance with the concrete needs of its customers.

CMI Industry is established in North America, China, Europe, India and Russia. It can also rely on the worldwide network of the CMI Group.

www.cmigroupe.com

Reheating furnaces § Pickling § Acid Regeneration § Cold Rolling Strip Processing (mechanical, chemical & thermal) § Automation D E S I G N | E N G I N E E R I N G | C O M M I S S I O N I N G | T E C H N I C A L A S S I S TA N C E & T R A I N I N G | A F T E R - S A L E S

VIEWPOINT

How tech catalyses disruptive change An abundance of new technology in the R+D pipeline doesn’t always point to a dynamic industry, and the global steel industry is a case in point. The capital intensity of our industry is high by any standards, and since the time of Bessemer there has been a perpetual challenge of a new generation of technology arriving before the last generation’s assets have paid for themselves. The industry’s adoption of Bessemer’s convertor itself was famously slow, and contemporary alternative high-volume steelmaking processes were cautiously adopted too. By Mick Steeper*

68 M steeper.indd 2

FUTURE STEEL FORUM

Steel Times International

16/05/2018 10:41:50

T

he volume boom of the mid-19th Century eventually ensued, but steel has been a fraught sector for investors ever since. Overcapacity became intrinsic, and thus low margins were persistent. As a result, the return on expensive new technology has long been weak compared with most other sectors.

None of this has ever stopped the development of the technology. It has slowed its uptake, but the pipeline has always built up steadily nonetheless. The pattern of adoption of the most significant new technology has itself become foreseeable, because a point is always reached where the potential competitive advantage of a technology shift is seen to outweigh the barriers to its implementation. A disruptor then acts, and his success (or anticipation of it) compels others in the industry to follow and re-equip.

In the most complete form of this process, the entire industry responds and all steelmakers are either driven to re-equip or else exit.

The technology base of the global steel industry, therefore, proceeds in bursts, with infrequent “sea-changes” taking the place of the continuous incremental development that prevails in smaller-scale and more highly differentiated industries.

The sea-change of Bessemer’s time was in operating volume. Before it, steel companies were viable as ten-tonnes-a-day enterprises.

As a result of it, the baseline became something like a hundred tonnes an hour. The technology had been waiting, and almost as soon as the Bessemer converter appeared a rival emerged in what became the open hearth process, since reverbatory furnaces were already in use for copper smelting.

Steel companies could now become large and powerful by adopting one technology or the other, investors became persuaded of the opportunity and a new order was eventually established across the global industry.

The most recent sea-change concerns process simplification. The bulk materials industries all migrate naturally towards continuous flow, as far as their technology permits it. Energy and time efficiency are maximised when it is possible, put in simple terms, to pour raw materials into one end of a production line and take finished goods from the other. Throughout the 1960s, the steel industry shifted from ingot-teeming and

Steel Times International

M steeper.indd 3

breakdown rolling to a continuously-cast mill feedstock. The next level of a flow process required a disruptor, however, and in the 1990s Iverson at Nucor obliged (Fig. 1).

The success of the EAF-based mini-mill as an alternative to the traditional integrated steelplant provides a well-documented case study in steel technology evolution1.

Today’s technology platforms The enabling technology for the next seachange in steel might well be the tools and methods of Industry4.0. The remainder of this paper looks at foreseeable sea-changes, and how Industry4.0 might accelerate them.

The most likely sea-changes share common characteristics, in that: i. They are already technically feasible ii. They tend to have either an economic or an environmental context (in some cases both) that is broader, and potentially more compelling, than the interests of the steel industry iii. They offer a business opportunity through early exploitation and hence a potential trigger to disruptive adoption.

Any potential sea-change that exhibits all of these characteristics is very likely to happen to some degree and at some point in time.

The completeness and rate of adoption, as

well as the location of the disruptor, are less predictable.

The core technology of Industry4.0 itself (meaning the automation and data exchange technologies that integrate modern manufacturing systems) is not especially well-suited to steelmaking, nor indeed to any of the bulk materials industries. The benefits of Industry4.0 are maximised when process routes and supply chains are more complex than those of steel, and the development focus of Industry4.0 centres on product integration and assembly, in industrial sectors which include steel’s major customers. The automotive industry is an obvious example, and there are some instructive steel company successes based on co-operative development with car manufacturers2.

The anticipation of more fundamental Industry4.0-driven change in the steel industry, leading to potential sea-change shifts in process technology, requires a different and deeper analysis, however. A starting-point for this is the identification of current economic and/or environmental contexts as described in point ii. of the list. To illustrate the approach, the table of Fig. 3 gives three such contexts.

To identify ways in which Industry4.0 might transform the steel industry, we can

Fig. 1 - Bessemer and Iverson

FUTURE STEEL FORUM

69 16/05/2018 10:41:51

VIEWPOINT

Fig. 2 voestalpine blanking

now look for ways in which today’s industry and its value chain are constrained by the limitations of present-day automation and data access. For example, the third point of Fig. 3 introduces the possibility of a declining strip market. It will be instructive to remind ourselves of why we make strip in the first place.

A strip paradox?

Steel is in one way a poor material for surface applications: its corrosion behaviour is adverse. In order to alleviate this, the bodyin-white of a modern steel car is invariably coated with another metal (zinc). Steel’s ascendancy in the application is historically founded in the ease and low cost of fabrication, and a strength-to-weight ratio that has always been advantageous and which is progressively consolidated by R+D effort.

There is an unstated attribute here too, however. Steel is a high quality material, meaning that it is very consistent in its properties and thus very stable in both its manufacturing and service performance.

The pursuit of this attribute further defines the performance criteria of strip production, in terms of the precise control of gauge and profile, but also of metallurgical processing control

70 M steeper.indd 4

FUTURE STEEL FORUM

conditions such as strain-temperature path.

We make strip this way because it assures the maximum yield of a homogeneous product.

The idea is intrinsic to the notion of rolled steel as a “semi”. Highly uniform material properties across the width and along the length of the coil (as well as from coil to coil) provide a foundation for manufacturing consistency. 3D structures can be built by fabrication, and in many applications the homogeneity requirements have rendered this a more usual method than incorporating a rolled profile.

Industry4.0 could change this, however.

The heterogeneity of profiles is repeatable and predictable, and the nature of cooling in the hot mill gives the property variation a generally helpful bias, towards harder edges and surfaces relative to a more ductile bulk metal. The power of modern computing, moreover, makes design for heterogeneity a realistic proposition. The very beginnings of the possibilities underpin the tailored blank examples from Ref. 2 (also Fig. 2), but in future this could progress a lot further.

The other contexts of Fig. 3 also play into this idea. Why should we learn to manufacture in steel with a systematic property variation, when doing so will still be

more difficult than fabrication from suitablytrimmed strip? Because the climate change imperative means that we need to use less steel, and to recycle more of it from scrap (which in turn may mean compensating in manufacture for the effect of residuals).

The most complete study of these principles to date is due to the University of Cambridge3. An example of the potential of heterogeneous design collaboration in a different supply chain is illustrated in Fig. 4, illustrating the principle of the Ideal Beam.

Steel I-beams should not be uniform in section for maximum efficiency in their conventional mode of use as a beam spanning two columns. The bending moment is maximum at the centre of the span, so the flange thickness should be greater there than at the ends. If all construction beams could be optimised on this basis, the annual reduction in carbon emissions would run to hundreds of millions of tonnes. In fact it is entirely possible to roll Ideal Beams, using a modern variant of an established design (the original Grey Mill), although to do so with the desirable range of thickness variation would also require a variable dogbone caster. The fact that no such mill/caster combinations exist is a consequence of the obvious drawback – that steel companies are unlikely to buy new and complex equipment in order to sell less steel.

Possible future business models Selling less steel (in tonnage terms) is implicit in a sustainable future, however, because using steel more efficiently is an imperative.

Fig. 5 (after Ref. 3) identifies the four main ways in which this might be achieved. The tools and principles of Industry4.0 make progress in all four directions realistically possible, especially by sharing the steelmakers’ expertise in the form of their material models through the entire value chain to include end-product design and manufacture. Industry4.0 can also realise mass customisation goals, and in many if not all cases, the flexible and reconfigurable variants of steel processing plant that would also be needed are known already.

The missing elements are, therefore, the source of ROI on new capital plant and the remuneration of the steelmaker for a broadened role of knowledge provider. Some of the historical conventions, such as selling

Steel Times International

16/05/2018 10:41:52

Global context

Steel industry context

Climate change Sustainable materials sourcing Autonomous vehicles

Opportunity for

Threat to

CO2 emissions from BF ironmaking

EAF operators

Integrated works operators

Increasing scrap supply (EoL of noughties output)

EAF operators, scrap processors

Less material in automotive BIW (diminished crash resistance

Strip producers

requirement, extreme lightweighting opportunity)

Fig 3. Table of Contexts

Variable Flange Thickness

Variable Web Depth

Fig 4. Principles of Ideal Beam (also showing “Dogbone” casting of Beam Mill feedstock)

Recycling without re-melting, re-use Reduced demand: light-weighting (using less material by design) Reduced demand: Extending service life of steel products Radical process efficiencies: Near net shape processing, minimised heating cycles

Fig 5. Table of 4 ways to use steel efficiently

steel by weight, will have to change too.

The steel industry’s key skill is the alteration of steel properties through controlled adjustments in the manufacturing process. Of course this is not an original idea, and neither is the extension of the skill to cover steel applications as well as production. What would be new is an acceptance that the skill is what the steel customer pays for, rather than the commodity of the metal.

Early examples of knowledge-based steelmakers are already recognisable (see Refs. 1 and 2 for cases). The next step towards a sea-change to knowledgebased steel might be backward integration, of which there will be raw materials and engineering variants within the supply chain.

The reclaiming of scrap prospecting by steelmakers is predictable. Less obvious but just as likely is the reclaiming of process equipment development – western steel has long since spun out its design expertise, but China in particular might avoid making the same mistake.

The trend to electric arc steelmaking will accelerate as zero-carbon electricity

Steel Times International

M steeper.indd 5

generation takes hold, and might presage a major role for original steelmakers in power generation as a core business. Collaborative operations with other materials producers may also evolve: the co-refining of titanium and direct-reduced iron from rutile sands is one variant predicted in a future world of electricity from fusion, in which the cost of electricity will consist almost solely of the investment burden of the generating assets.

Mass-customised profile products can be expected to displace strip as the principal “traditional” product of the global steel industry.

When our successors are in a position to view the first half of the 21st Century in hindsight, they will probably conclude that the influence of Industry4.0 on the transformation of the steel industry was indirect. Steel’s customers can and will exploit Industry4.0 more than steel itself does, but it will be their demands that draw steelmakers into the knowledge-based value chain. Once there, new opportunities to add value ought to kindle a (long forgotten?) spirit of enterprise.

As any metallurgist will confirm, there is no

material more receptive to knowledge-based manipulation than steel. A knowledge-based steel industry will moreover enjoy some built-in resistance to low-cost entrants and hence to the investment-stifling bane of overcapacity.

The future of steel is destined to be very different from the present, and better too.

References 1. R. Ranieri and J. Aylen, “The Steel Industry in the New Millennium, Vol. 1 – Technology and the Market”, IOM Communications, London, 1998. ISBN 978-1-8612-5019-3 2. D. Madar, “Big Steel: Technology, Trade and Survival in A Global Market”, UBCPress, Vancouver, 2009. ISBN 978-0-7748-1667-0 3. J Allwood, J Cullen et al, “Sustainable Materials With Both Eyes Open”, UIT, Cambridge, 2012. ISBN 978-1-90686005-9 * Mick Steeper recently retired from the Chair of the Iron and Steel Society (the Steel Division of the Institute of Materials, Minerals and Mining), following a career in R+D in the plant building industry.

FUTURE STEEL FORUM

71 16/05/2018 10:41:53

IIoT APPLICATIONS

Asset health The Industrial Internet of Things (IIoT) offers innovative ways of doing things better and utilising existing assets. In a nutshell, it’s all about optimising installed assets with a view to increasing efficiency and availability and, ultimately, predicting future asset behaviour based on historical data. By Jens Hundrieser* and Steffen Ochsenreither**

M

anufacturing and production, the typical domain of IIoT applications, are to a great extent about optimising installed assets to increase efficiency and availability. The ultimate goal is to predict asset behaviour in the future based on historical data – often described as predictive maintenance or health. The majority of today’s assets in process automation plants already deliver a lot more data and information than just one single process value.

This additional data can range from simply additional process values, to self-diagnosis about the asset’s health or even the prediction on potential occurring problems in the near future based on internally diagnosed device parameters.

Unfortunately, this kind of information is most often locked into the asset itself and can only be retrieved locally at the asset: Process

72 Endress.indd 2

FUTURE STEEL FORUM

automation plants around today are 5, 10 or even 20 years old and during the planning phase of these plants, asset diagnostics was often not considered. Over the years of operation, assets get replaced and new technology finds its way into existing plants – however, the existing integration of these assets into a PLC/DCS is rarely touched. All the new features and functionalities are not accessible without interaction on the asset itself. The digitisation and interconnection of all operational assets offer enormous potential for cost savings and optimisation in the process industry.

How to get the asset information While the philosophy of IIoT is to unlock exactly this hidden potential of connected devices, existing plants are often quite the opposite: locked down systems, with no

means of connecting to the installed assets.

Further, if one wants to make use of the features and functionalities of an asset, he/ she would need to have an overview about what is actually installed in the plant, where it is installed and what the asset actually can offer – not all data provided is of use sometimes.

When talking about IIoT, it is always stipulated that we require data to create valuable insights. Although this is true, it is often disregarded that there are some necessary steps that must be taken in between. Before one can analyse gathered data, it is crucial to know who the data provider is and what kind of data one can expect to receive.

Without knowing what is installed, what data will be provided, it is a hard to analyse this data. Especially in plants that have been around for a few years, it is often not clear

Steel Times International

16/05/2018 10:44:56

what the actual installed base looks like. Who are the manufacturers I have installed? How many different asset types do I have? Are the assets still available or are there any obsolete ones?

So the first step into IIoT includes manual work: creating a list of all installed assets in a plant, with at least some basic information such as manufacturer, asset type, location, and a unique identifier (usually the serial number).

Traditionally, one was sent into the plant with pen and paper, gathering the serial numbers, the manufacturer, the asset type and other relevant information such as location.

After this data has been put into a list then the real work starts: are the assets still available?

Where are the documents, manuals and calibration certificates stored?

Relying on existing documentation is not usually recommended: documentation about installed assets is often outdated or incomplete, so that this can hardly serve as a basis. There is usually no other way than to go through the system physically, to manually identify, capture, and create the database. Obviously, this method of creating the database presents a great deal of time and personnel. In addition, neither the data consistency nor the actuality of the data is guaranteed.

Steel Times International

Endress.indd 3

Even though this is a crucial step towards the world of data analytics/IIoT, the entry hurdle is very high as the time and money spent on creating a simple list of installed assets is outweighing the benefit.

Manual and automatic asset database creation and digital twin With today’s available technologies, this entry hurdle can be lowered: mobile devices are capable of creating a device database easily utilising a smartphone app (see Fig. 1: Optimised asset list creation); in just a few steps and in a matter of seconds. For the unambiguous identification of the devices usually a combination of the serial number and manufacturer is used. These can be found on the respective markings of the devices, ranging from metallic identification plates to identification methods such as QR codes to digital tag labels such as RFID. After the identification of the assets, further information can be attached to this now generated “digital twin”, such as geolocation (using the GPS functionality of the mobile device, for example), the tag of the asset, information such as criticality, other comments or even photos and drawings of the instrument and its location. In tests with participants of different ages and education levels, the average time it took to gather all this information and create the database entry was less than a minute.

Obviously, in a plant with hundreds of assets, this still might become a tedious task.

In addition to manual capture, there are nowadays also methods for

automatically creating this database (see Fig. 2). Since the development of digital communication protocols such as HART, PROFIBUS or FOUNDATION fieldbus, the goal was always to provide the user with more information out of the field and unlock the features and information that the manufacturers built into their devices.

As these protocols are standardised by the corresponding authorities, there are means to read out the electronic name plate of connected assets automatically. This will reduce massively the effort in creating an installed base overview with the digital twins of a process automation plant.

In field trials with selected partners Endress+Hauser was able to automatically generate a database including more than 800 assets in a single plant in less than four hours – from the installation of the edge device to the last digital twin being created.

But what happens next? Regardless of whether the database is created manually or via a so-called edge device, it can now connect in the background with the asset information database of the manufacturer and be filled with the device-specific documents such as manuals and certifications. With a database that is second to none, the Endress+Hauser asset management system contains data records of 47 million installed field devices. This is important as in today's process automation plants, a large number of field devices and sensors from various manufacturers are installed and used. To operate and maintain these plants economically, a successful plant asset management (PAM) system is necessary.

Studies have shown that up to 70% of the time

Fig 1. Optimised asset list creation

FUTURE STEEL FORUM

73 16/05/2018 10:45:00

IIoT APPLICATIONS

benefits: • No programming of the PLC/DCS is needed to unlock the asset features.

• Existing plants can be easily retrofitted without the fear of interfering with the existing process.

• A bypass establishes another level of security, as asset management data is clearly separated from process data.

Network topology - from sensors to digital services

Fig 2. Automatic creation of asset database

spent on a maintenance job is in search of information – not with the actual maintenance job. Taking into account that currently there are around 30% of obsolete assets installed, this bears a big risk for running a plant smoothly.

Just simply making the user aware of the obsolescence situation would be a big step forward into the direction of asset management. Luckily, asset management can easily provide this information as shown in Fig. 3.

By having all this information at hand, this not only increases the efficiency of the maintenance technicians, but also reduces the risk of faulty/inefficient maintenance, as the correct information is provided to the right person. However, this requires a well-maintained and comprehensive device information database, which we just created in the last step.

So what about the device features and functionalities, such as asset health?

factors can then ultimately end up in a predictive maintenance application.

This is the logical step from static to dynamic asset information. Collecting and trending asset health of periods of time and storing it in a database can ultimately lead to a collection of data, which can then be used to predict the asset health.

Asset obsolescence Below you find the availability of the assets at the provider 3%

32% 32% 32%

75 order stop

Asset health – from static to dynamic asset information Now that the connection to the field is established (via the edge device) and a comprehensive overview on the installed assets is available, the next step can be performed: the visualisation of asset health.

Thanks to Heartbeat Technology for example, the field devices are also able to output diagnostic values and device-specific trend parameters.

This asset data can be visualised to give the user an indication on the availability of his assets. Gathering this information over a longer period of time and cross-referencing it with other process variables or external

74 Endress.indd 4

FUTURE STEEL FORUM

11

8

140

phase out available

undefined

Product status for 8 assets is undefined Set the product status and improve your insight

Fig 3. Visualisation of asset obsolescence automatically generated in the database by cross-referencing the gathered asset data with data from the manufacturer

Of course, all these additional features and functionalities related to asset management should never compromise on the security and integrity of the actual process. As shown in Fig.2, by adding a bypass channel (through the Edge Device and the Ethernet to PROFIBUS Gateway) to the asset management database, the PLC/ DCS remains untouched. This offers multiple

Today’s field devices often have the necessary connectivity already built-in to transmit data directly to the database. This can be done by connecting through Wi-Fi, Ethernet technologies or even cellular connection.

Security aspects In order to understand the relevant security aspects, it is necessary to take a look at the architecture. This will give the entry points for the security discussion and show critical points of interest: The data flow starts in the field at the instruments. Via interfacing devices like gateways these data are then transmitted into the cloud, where they are transformed into information. There, additional data sources may be injected to create even more information. These can be other Endress+Hauser systems or customer environments like engineering tools or ERP systems.

The connection to the asset management database has to be established in a secure manner. As shown in Fig. 2, the Edge Device is located behind the company’s firewall. As an additional security measure, the connectivity between the Edge Device and the Asset Management Database is a one-way street: In this example there is no direct connection possible between the asset management database and the field network.

As security, trust and compliance are sensitive topics, a quality audit is essential.

When the decision is to go for any IIoT offering, an accountable quality assessment of cloud services through a transparent and reliable certification process should be part of the process. Any quality audit needs to consider different frameworks and law regulations, these should include at least: • ISO 27001: Information security management.

Steel Times International

16/05/2018 10:45:01

OAuth (Open Authorisation): To support safe user identification during the usage of the software, there should be a tokenised process to identify users against the cloud service.

User passwords should be transmitted only for token generation. This compli-cates scamming attempts and guarantees a safe authorisation.

Encrypted communication channels only: The communication channel to the cloud service should always be established via a secure and encrypted https connection.

Thereby all payload data should be encrypted according to industry standards and the cloud computers should be trustfully authenticated by a certificate issued by a world-wide renowned certificate authority.

User information: When accessing his or her account the user should be able to see past activities. The same mechanisms are used for online banking to detect possible fraud usage or failed login attempts.

Processes: In the event of serious security incidents, which may occur even in the safest environment, the provider should have established internal processes to react as quickly as possible and to inform all affected parties to keep users safe from harm.

Steel Times International

Endress.indd 5

Fig 4. Digital service program for IIoT applications Cloud solutions

Summary: IIoT offers innovative ways in doing things better, utilising assets that are already existing. With Analytics, a digital service for installed base analysis is offered to customers already today (see picture 4). The installed base of a system can be simply captured and analysed using real-time and historic data.

Asset informa¬tion in the field is recorded with a mobile smart device using the new Endress+Hauser Scanner app that reads an RFID chip, QR code or tag – alternatively, the asset informa¬tion can be captured automatically by an edge device. All the data are saved to the cloud, visu-alised on a dashboard, and asset management recommendations are issued, e.g. product availability or suggestions for a suitable replacement device.

� Application solutions

Apps & services rity

Encryption of passwords: To ensure user confidentiality of passwords they should not be stored in plain text. In the case of the aforementioned asset database, for example, passwords at the user side are encrypted with ‘bcrypt + salt + pepper’ and just the hash is stored in the database.

Gateway security: The gateway is a critical point in the architecture because it represents the access point from and to the user’s plant and should record only data from the field and transmit these into the cloud. Vice versa, i.e. from cloud to the gateway, no communication should be initiated. Thus all

ecu

To comply with all previously mentioned requirements, it is necessary to have proper functions and features implemented in the software. The following outlines some of the security measures that we undertake:

Server location: A trustful and reliable cloud hosting partner should be used, where the servers are located in a safe location. For example, the servers in the here shown example are located and operated under European law and jurisdiction, which is among the most stringent in the world.

Customers should be made aware that their data is subject to one of the highest data security standards worldwide.

IT s

• IEC 62443 Security for industrial automation and control systems.

• Contract & compliance.

• Data privacy.

• Operational processes.

• Software as a service.

• ISO 20000: Service Management System.

Big data platform

Connectivity

Processors 4.0

� Cloud-based services for data analysis and generation of new insights � Endress+Hauser IoT platform � Integration into customer cloud platforms (SAP, Microsoft, GE etc) � Edge gateways for IT/OT connectivity � Advanced process data � IT-based connectivity � Security by design

incoming ports to the gateway need to be blocked. The only exceptions are software updates for the gateway: Software updates need to be installed in parallel to the running system. When everything is complete the gateway would switch to the updated runtime and disconnected for the period of the reboot. To guarantee safe downloads, these updates should be certified and checked against the original file to prevent manipulation.

In the near future we expect to see even more possibilities of connecting and how the data can be used. Of course, data security and privacy are a big concern and should be considered carefully when deciding for any IIoT solution.

During the selection process, the IIoT provider should be critically reviewed and checked. An independent certifica-tion organisation can help here.

Customer data: All customer data used by the provider should solely be owned by the customer and should not be shared with 3rd party service providers.

* Regional industry manager Europe metal IIoT, Endress + Hauser Messtechnik GmbH ** Business development manager, Endress + Hauser Process Solutions AG

FUTURE STEEL FORUM

75 16/05/2018 10:45:02

PROCESS CONTROL

Through-process optimisation This article presents an overview of the newly developed through-process optimisation solution. The holistic approach and the combination of metallurgical and operational know-how with an intelligent IT-system allows steel producers to improve their overall efficiency, achieving higher levels of quality and developing and maintaining their knowhow basis. By Jan Friedemann Plaul*, Wolfgang Oberaigner*, Yuyou Zhai**, Thomas Pfatschbacher*, Manfred Kuegel*

T

he progress of digitalisation in many industries is creating new opportunities for the improvement of overall efficiency and quality in steel production. At the same time, end-customers are demanding steel products at higher quality levels. They look for tailor-made steel-grade solutions, short development times for new steel grades, and the manufacturerâ&#x20AC;&#x2122;s ability to swiftly respond to quality deviations that have resulted in rejections. The number of customers requesting zero-defect products is constantly growing. As a consequence of this trend, steel

76 primetals.indd 2

FUTURE STEEL FORUM

producers need to target even higher quality levels, increased process stability, greater process flexibility, and high production efficiency, in order to be successful at manufacturing challenging products for demanding customers.

This means that steel producers need to become even more capable and to use a system that ensures accurate and stable control of all process parameters. Producers require quick and complete access to quality- and process-relevant data, a deep understanding of how a change in process

parameters will affect the properties of their products, and the know-how to develop products quickly and successfully. To meet these requirements, Primetals Technologies has developed and introduced the ThroughProcess Optimisation (TPO) solution, which targets the smart, digital interconnection of various process units and the accumulation of know-how along the entire steel-production chain.

Introduction Todayâ&#x20AC;&#x2122;s steel producers have to perform in a

Steel Times International

16/05/2018 10:47:30

Fig 1. Overview of through-process optimisation

THROUGH-PROCESS QUALITY CONTROL

THROUGH-PROCESS KNOW-HOW

• Comprehensive data recording with full product genealogy • Quality control system functionality • Deviation & root cause analysis THROUGH• Corrective & compensational PROCESS actions OPTIMISATION • Automatic product grading • KPI evaluation and visualisation • Know-how rules editor • Statistical process control and intelligent analysis • Interface to data mining platforms

Hot rolling/plate rolling

Steel melt shop

• Know-how expert service • Know-how rules generation • Definition and improvement of key performance indicators • Product development • Quality management • Know-how modules, e.g. metallurgy,operations, quality • Data based know-how generation & analysis • Trainings • Audits & consulting

Cold rolling

Cold band processing Annealing line

converter Ladle furnace

RH plant

Slab caster

Hot strip mill

Picking line tandem cold mill

Galvanizing

EAF (EBT)

Chemical composition Cleanliness

Internal soundness

Homogeneity/segregation

Austenite formation Recrystallazation Precipitation formation Ferrite grain structure Control of r-, n-values Fig 2. Metallurgical know-how along the entire steel production route

Metallurgical know-how best technological properties

• Chemistry • Process parameters • Metallurgical modelling • Metallurgical know-how base • Defects/defect classification (“defect catalogue”)

• Thickness, flatness, profile, roughness • Root cause and remedial actions

Operational know-how stable production

• Operational practice • Campaigning • Preventive-maintenance practices • Consumables, energy efficiency

Material-development know-how

Sustainable high quality products

• Material-grade & process development • Simulators & systematic simulator use • Know-how base, and partnership with suppliers and steel companies

Quality management & process know-how Sustainable strategy & culture

• Quality-management consulting • Feasibility & improvement studies • Process & organisation consulting

Fig. 3 Through-process know-how modules

Steel Times International

primetals.indd 3

THROUGH-PROCESS KNOW-HOW

• Surface quality

Quality know-how stable quality properties

Properties of final product

Batch annealing

highly challenging market environment that is marked by declining raw material quality along with price volatility, increasing pressure from stricter environmental legislation, and fierce market competition due to excess capacities all around the world.

These challenges force many steel producers to thrive into the market of highvalue and high-margin steel products like automotive grades. When moving from commodity grades and mass products to high-quality and advanced high-strength products, steel producers have to address the following issues: • Stable production with highest quality standard, minimisation of rejection rate: because of the tighter production tolerances and lower process reliability under such conditions, a low rejection rate becomes a highly important prerequisite for a profitable production.

• Extension and intensification of quality and research activities: in order to reduce the rejection rate while maintaining cost effective production, it becomes favourable to focus R&D and product development efforts and to specialise in certain product areas.

• Comprehensive know-how build-up, documentation and protection measures, which are difficult to deal with without adequate IT-system support.

• Become a learning company that is among the fastest to learn and improve, in all terms of production, automation, technological and internal processes.

In order to reduce the rejection rate and to provide support for product and production development, Primetals Technologies introduced its new know-how based Through-Process Optimisation Solution (Fig. 1) [1,2,3], which consists of two parts: 1. A new, intelligent through-process quality control (TPQC) system, and 2. The through-process know-how (TPKH) packages 1. Through-process know-how 1.1 Metallurgical know-how for quality control, rules generation and KPI definition For production of high value added steel products many hundreds of properties, parameters and conditions have to fit together and be aligned to achieve good product quality and a cost efficient and

FUTURE STEEL FORUM

77 16/05/2018 10:47:32

PROCESS CONTROL

Level 4 (Entreprise Resource Planning System) Level 3 (Manufacturing Execution System, APS, ODS) Through-Process Quality Control (TPQC-System) L2

L2

L2

L2

L2

L2

L2

L2

L1

L1

L1

L1

L1

L1

L1

L1

HM De-S

LD (BOF) converter

Ladle furnace

RH-OB plant

Slab caster

Hot strip mill

Pickling and tandem mill

Galvanizing

Product yard management

Coil storage

Smart Sensors/Robotics/Tech. Packages/Quality Inspection Systems/CMS Maintenance Asset Technology Fig 4. Integration of TPQC with automation and IT-systems

stable production (Fig. 2). To develop rules which are controlling the process and quality parameters of the product, a deep understanding of all metallurgical reactions and complex physical processes is

mandatory. In a first step metallurgical experts are defining for each process unit a â&#x20AC;&#x2DC;basic set of rules,â&#x20AC;&#x2122; which will be implemented in the TPQC system by using a special rule editor.

Beside the development of formulas for the

Slab 1

Slab 2

Finished coil

Hot coils

Fig 5. Genealogy based data tracking across multiple processing units

HM De-S

LD (BOF) converter

Ladle furnace

Process parameter(s) threshold upper limit

Process parameter(s) threshold lower limit

RH-OB plant

HM De-S Process parameter(s) collection

Process parameter(s) evaluation

Fig 6. Quality control function along the production route

78 primetals.indd 4

FUTURE STEEL FORUM

Slab caster

Hot strip mill

Galvanizing

rules, the experts are also implementing a set of possible route causes in case of deviation as well a set of corrective actions for the individual process unit.

In a second step a set of compensational actions is defined for each type of deviation, which can be applied for downstream process units. The intelligent design of the rules enables an automatic evaluation of possible route causes by the TPQC system. It is also possible to define a different set of rules for individual steel grades or groups.

To measure the efficiency and the process stability for each process unit, key performance indicators (KPIs) are developed by experts and implemented in TPQC. Together with the plant management the target setting for each KPI is agreed. During regular review meetings the performance of the individual KPIs are verified and action plans are developed and agreed to improve the performance. The KPI tracking can, therefore, be an important part of the continual improvement process (CIP) of a steel plant.

1.2 Expert service For the implementation of TPO and in order to resolve specific problems, Primetals Technologies supports its customers with its own experts and with external consultants.

These specialists are highly experienced and cover various disciplines, acting as consultants for a wide range of topics; for example, plant operation, quality management, maintenance, or end-customer

Steel Times International

16/05/2018 10:47:33

List of possible root-causes Most likely root-cause on top

time TPQC keeps track of most frequent root-causes and allows for trend analysis with respect to continual improvement

Manual root-cause confirmation increased root-cause awareness root-cause documentation

Documentation about most likely root-cause to sort possible root-causes by their documented frequency

Fig 7. Semi-automatic root-cause analysis support

qualification (Fig. 3). Primetals Technologies also provides training sessions for steel producers who decide to implement TPO. In close co-operation with the respective customer, the team of experts devises and incorporates an optimal set of rules into the customer’s TPQC system. These rules will ensure better control of the steel-production process, optimise product-quality consistency, and improve general plant operation.

2. Through-process quality control system The basis of the TPO solution is the throughprocess quality control (TPQC) system, which creates a central database by receiving quality- and process-relevant production data from all production units via the Level 1 or Level 2 automation systems. Additionally, laboratory measurements and data from all types of sensors and equipment are centrally stored. The TPQC creates an information-rich genealogy of each individual product that is processed, and makes it possible to retrieve process data of all production steps for every part of the product. This allows users of TPQC to track quality issues in very little time and analyse them by reviewing process data for all relevant production steps, which is key for fast troubleshooting and claim management.

Fig. 4 shows the TPQC System embedded into the automation and IT environment.

The essential functionality of TPQC is to ensure desired product properties and increase quality levels by monitoring all quality-relevant process parameters along

Steel Times International

primetals.indd 5

the full production route at defined quality gates. Therefore, communication between the production-management system (PMS) and individual automation systems of each production unit is required to evaluate distinct steps of the production process.

2.1 Product genealogy The TPQC automation system is collecting all relevant process and production data provided by the various plant units of the whole process chain from ironmaking through hot rolling, cold rolling, annealing and galvanising (Fig. 5). In addition to pure process data, which is collected from various processing unit, the genealogy information

Target temp 1620° C

1320° C

Blast furnace

Hot metal

DeS

Target temp 1680° C

LD (BOF) converter

HD de-S

interconnects this data across all involved processing units. For example, the genealogy keeps track of elongation factors, head/ tail changes, upside/down-changes as well as cutting and welding operations that are possible between different processing units of the production chain.

The genealogy information supports claim management, since no important information would be lost any more. Even after years it will be possible to deal with a customer claim since heat number, slab number and coil numbers, even daughter coils and sheets, can be identified and any relevant deviation or defect before delivery may have been logged in the long-term archive of TPQC.

DeC BOF #2/3

1620° C

1585°C

1560°C

Ladle furnace

RH-OB plant

Slab caster

RH-OB

CCM reduction of casting speed within limits to avoid a strand-break

LF

Immediate compensational Temperature action: reblow at BOF too low TPQC identifies possible root-causes Corrective action Check/change material related of new material in use

Input material (scrap, hot metal)

analysis, temp, weight)

Input of additions, cooling, heating agents (weight) as calculated L2/offgas model working/in use New material quality in use

Fig 8. Root-cause analysis within the melt-shop

FUTURE STEEL FORUM

79 16/05/2018 10:47:34

PROCESS CONTROL

root-cause, the responsible person places a check-mark and thereby confirms the rootcause for the given quality deviation.

4.

The system keeps track of the rootcauses for any detected quality deviation and calculates a root-cause statistic.

general, RCA is often addressed by several methods and has to be applied iteratively as a tool of continuous improvement (Fig. 7). In order to support quality engineers and operators in respect of root-cause analysis across multiple production processes, TPQC provides a semi-automatic approach:

The documentation of root-causes for identified quality deviations is mandatory in case of some quality standards like ISO/TS 16949 [2]. In addition, a root-cause statistic can be calculated for an arbitrary time period and thereby enables the calculation of trends over time. Hence, the root-cause statistic feature is a valuable aid to provide convincing evidence on the effectiveness of the quality management with respect to continual improvements. In particular, it is a useful tool for the plant manager, enabling a strict monitoring of deviation frequencies, in order to identify the most frequent and most costly deviations. These identified deviations may then be addressed by additional problem solving methods and corrective actions (see below), if adequate.

1.

In case of a detected deviation, a list of highly probable root-causes is shown to the operators or quality engineers depending on the plant location and organisational responsibility.

2.

Each root-cause comes with a specific description for root-cause verification, in order to eventually remove any doubt in cases in which more than just one root-cause might be possible.

3.

After identification of the actual

2.4 Corrective and compensational actions A corrective action can be defined as a set of actions to eliminate the cause of a quality deviation under specific conditions.

Unfortunately, the production environment is subject to a large number of influences like any kind of external disturbances, raw material unknowns as well as the human factor and other non-deterministic events.

Hence, it is seldom possible to eliminate the cause of a quality issue permanently. For this

Fig 9. Quality assistance

2.2 Quality Control System Functionality Quality conformance checks are carried out by means of a specific rule system on the production data, and the results will be shown to operators and quality engineers, respectively, depending on the kind of quality issue and location in the plant. By means of the rule editor TPQC offers a flexible way to make quality checks, which paves the way for future adaptions and extensions without having to change any part of the systemâ&#x20AC;&#x2122;s source code. The rule system of TPQC is also an important pillar for Primetals Technologies Through-Process Know-How described above.

Fig. 6 shows the principle of this process/ quality checking inside TPQC. A green traffic light indicates that all values (of a production unit) are inside the specified bounds. A yellow should express the detection of minor problems, requiring some further checks. A red light should indicate more serious issues, which in most cases need a more complex remedial action and final decision plan to resolve the problem, which can even include a dismissal of the product in the worst case.

What can be defined as minor and major has to be developed from the process experience and must be finally specified as expert rules.

2.3 Deviation and Root-cause analysis Root-cause analysis (RCA) [4] is a method of problem solving used for identifying the root-causes of faults or problems, well known from literature and required by the ISO/TS standards for automotive production [5]. In

80 primetals.indd 6

FUTURE STEEL FORUM

Fig 10. Various surface defects scattered across a hot rolled coil

Steel Times International

16/05/2018 10:47:35

reason, the definition of corrective action has to be broadened in so far as a certain corrective action for a given root-cause may eliminate a quality deviation absolutely and permanently only under the specific condition that applied previously, although not all of them can be determined or measured and archived.

A compensational action is defined as an action to repair an already affected semi-finished product, e.g. by cutting out defective sections from a strip or by means of surface scarfing of a slab as indicated in the example below. Because of the fact that the system provides guidance to operators and quality engineers, the tool can also be seen as a learning tool. It is a fact that people who are actively involved in root-cause analysis for some time will become more aware regarding quality related influences, which allows them to take preventive measures even before a quality related incident happens, that is without having to wait for TPQC rootcause suggestions at any rate and for any incident, respectively.

Thus, TPQC can be used as a pure conformance tool for quality checks and rootcause analysis support, but the real benefit of this system is realised when it is also used as a continual learning tool in order to improve operator and quality engineer skill levels.

The next easy example in Fig. 8 deals with quality issues in the melt-shop.

This Fig. 8 illustrates the example of a produced heat, which is planned to arrive subsequently at a continuous casting machine.

At the end of the BOF process, the L2 system indicates a temperature of the liquid steel that violates the grade specific requirements. In this case, TPQC evaluates the temperature by means of process specific rules and records a quality issue. In addition, the TPQC system immediately suggests a reblow at the BOF and if the operator confirms this, a message is sent to the operator at the continuous casting machine to reduce the casting speed. By reducing the casting speed, TPQC tries to compensate for the additional processing time caused by the re-blow at the BOF, which otherwise would increase the risk of a sequence break or slab quality issues.

In addition, the system indicates again a

Steel Times International

primetals.indd 7

Fig 11. KPI visualisation

Fig 12. SPC standard diagram

list of likely root causes for this problem. In the example case, a new material had been used for the material additions. The operators, for example, had not managed to update the L2 data base with the new material data in time, which led to an inaccurate calculation of the L2 model for the BOF process and finally caused the temperature deviation.

Fig. 9 shows a screenshot of how TPQC indicates root-causes for recorded quality issues to quality engineers. The list on the left hand side of the screen shows all semi-finished products for a given day (in this case filtered for heats). The selection of a heat with recorded quality issues (indicated by the call sign) shows all recorded quality alarms/issues. In this case there are two quality alarms. By selecting a quality alarm, the instructions for verification as well as root causes are shown in the text window at the bottom of the screen.

Even though the given example shows a root-cause analysis for a single process unit,

this kind of root-cause analysis is not restricted to single processes or plant areas (meltshop, hot strip mill, cold mill etc.). By means of carefully specified rules for root-cause analysis, TPQC can also indicate root-causes across process or plant boundaries. In this way, the system supports through-process root-cause analysis, which is especially useful in case of certain surface defects that can be tracked back to the liquid phase. Surface defect rules as well as surface grading will be treated in section 2.5 below.

2.5 Automatic product grading TPQC incorporates a rule-based module for coil-grading. The coil-grading module preprocesses surface defect maps. Afterwards, the grading-module performs an in-depth evaluation of the defect map, in order to determine instantly whether a coil surface quality matches the pre-defined requirements of the end-customer, and what action to take if this is not the case.

FUTURE STEEL FORUM

81 16/05/2018 10:47:36

PROCESS CONTROL

1. Problem Definition Why do some grades have a high number of surface defects per strip

2. Data Understanding Are these missing values, constant values, outliers

3. Models Applying machine learning methods to isolate root causes

4. Derive actions Create tasks and implement new rules. KPIs and SPC charts

rolling time RHF furnace gas pickling speed

Fig 13. Example for data mining project and definition of actions

2.6 Key Performance Indicator evaluation and visualisation The centralised collection of data enables the generation of key performance indicators (KPIs), which convey information about technical and business-related achievements and illustrate what progress has been made.

TPQC implements various types of graphical human-machine interfaces to support staff members from the quality and production departments, as well as top-management executives, in monitoring and benchmarking production conditions with respect to specific targets that are in alignment with the KPIs. 2.7 Statistic-process control (SPC) SPC [6] is a reliable and proven tool to provide statistical evidence that a production process stays within its predefined operational range and, therefore, behaves in a controlled way. There are significant advantages of having this statistical process control applied to quality assurance processes, since sooner or later statistical significant deviations of process values will have an impact on product quality. In general, SPC aims on sampling data which is meant to be stable over a long period of time in order to allow for a reliable detection of undesired process dynamic. TPQC can follow up single process criteria but also combined calculated performance indicators and KPIs, and offers SPC charts for selectable material/product related measurements or process data.

Fig. 12 shows a typical SPC control chart with a list of subgroups. Each subgroup which

82 primetals.indd 8

FUTURE STEEL FORUM

is exceeding the control limit is indicated by red or blue according to a violation of the upper or lower control limit.

2.8 Machine learning capabilities The extraordinary large amount of structured data in the TPQC system is perfectly suited for analysis with data mining and machine learning algorithms. TPQC offers a direct interface to transfer data effortlessly to leading data mining platforms. All acquired data is automatically assigned to the correct products along the process chain (steel making heat, slab, hot rolled strip, cold rolled strip, galvanised strip) by the genealogy function in the TPQC. Data mining and machine learning in combination with TPQC offers the following benefits: • Fast analysis for data quality and potential problems of measurements, manual inputs, etc.

• Quick analysis of multi-variate problems such as variation of mechanical properties by visualising the raw data from TPQC • Root cause analysis of problems using through process data and applying advanced data mining methods • Predictions using machine learning methods for end-of-line properties of products or process stability Closing the loop by creating new rules, KPIs or SPC charts in TPQC based on the data mining results. The data mining functionality of TPQC supports quality and process engineers in optimising the product

quality and stabilising the production process.

It is an extremely useful tool to increase the productivity of technologists and engineers throughout the process chain.

References [1] Kurka, G.; Hohenbichler, G.: TPQC – Through-Process Quality Control, CISA 2016, Beijing [2] Kurka, G.; Hohenbichler, G.; Pfatschbacher, T.; Pichler, L.: Know-how based root cause analyses tool to ensure high product quality and process; ESTAD 2017, Vienna [3] Ringhofer, M.; Wimmer, G.; Plaul, J.F.; Tatschl-Unterberger, E.; Herzog, K.: Digitalsierung in der Stahlindustrie, Stahl und Eisen 137, 2017, Nr. 5 [4] Wilson, Paul F.; Dell, Larry D.; Anderson, Gaylord F. (1993). Root Cause Analysis: A Tool for Total Quality Management. Milwaukee, Wisconsin: ASQ Quality Press. pp. 8–17. ISBN 0-87389-1635 [5] ISO/TS 16949(E), “Technical Specification: Quality Management Systems – Particular Requirements for the Application of ISO [6] Statistical process control – Reference manual, second edition, DaimlerChrysler Corporation, Ford Motor Company, and General Motors Corporation, 2005

* Primetals Technologies Austria GmbH **Primetals Technologies China Ltd.

Steel Times International

16/05/2018 10:47:37

ROSS CONTROLS A Global Leader in Pneumatic Safety Valves

Energy Isolation

Load Holding

Safety Exhaust

Energy Isolation Lockout Valves for LOTO with Pressure Verification

PO Check Valves for External Monitoring

Control Reliable Double Valves with Internal Dynamic Monitoring or for External Monitoring

Cat. 1 PLb

Cat. 2 PLc

Cat. 3 PLd

Cat. 4 PLe

Is Your SAFETY Application SMART & Connected?

Safe Cylinder Control and Stop

Safe Cylinder Return

Double Valves for External Monitoring

Double Valves for External Monitoring or with Dedicated Reset

www.rosscontrols.com

DIGITALISATION

Is your plant digitally mature?

With Industry 4.0 the new buzzphrase across global manufacturing industries, the big challenge for plant builders is how to prepare steelmakers for the digital revolution.By Marco Ometto1, Andrea Merluzzi1, Costanzo Pietrosanti2

I

ndustry 4.0 or the fourth industrial revolution (Fig. 1), or ‘smart manufacturing’ as it is known in the USA, are terms that became popular in 2013.

The digital transformation (or digitalisation) of manufacturing companies has been viewed as a mandatory path for increasing competitiveness and prosperity, affecting also the steel sector. The economic resources needed to support such a transformation are huge: forecasts indicate a spending growth from $66.67 billion in 2016 up to $150 billion by 2022 and a compound annual growth rate (CAGR) of 14.7%. There is growing evidence that industrial robotics spending growth will hit $81.47 billion by 2022 from a current level of around $41.75 billion (CAGR of 11.7% over the period 2017-2022).

Thanks to huge developments in terms of connectivity, we also see an incredible and fast diffusion of innovative technologies such as cloud computing, big data, artificial intelligence (AI), Industrial Internet of Things (IIoT) and the smart network of intelligent machines. Forecasts indicate worldwide growth from 15.4 billion connected devices in 2015 up to 30.7 billion in 2020 and 75.4 billion in 2025 with machine-to-machine (M2M) connections reaching 18 billion by 2022.

The big question is: How to support an organisation in getting ready for such disruptive technologies and in selecting the best opportunities? In fact, considering the outstanding number of I4.0 enabling factors and the relevant offer of technologies and solutions, the risk to invest in directions with lower returns could be very high.

This scenario clearly highlights the need for methods and tools for a systematic

84 danieli.indd 2

FUTURE STEEL FORUM

Fig. 1 The industrial revolutions (Aberdeen Essentials - 2017)

INDUSTRY 4.0 INDUSTRY 3.0 Cyber physical INDUSTRY 2.0 Automation, INDUSTRY 1.0 Mass produc- computers and systems, internet of things, Mechanisation, tion, assembly electronics networks steam power, line, electrical weaving loom energy

1784

1870

quantitative estimation of the company in question’s digital maturity and its readiness for the implementation of smart factory concepts.

Digital maturity assessment Danieli Automation has tailored the concepts of Digital Readiness and Digital Maturity[1], [2] to the metals industry offering its customers a digital service named Qs-Digital Scan, an assessment study focusing on technology and operations dimensions. Digital Maturity is defined as the state of adequacy of the digital landscape (architectures, systems, networking and sensors) to make possible the successful implementation of Industry 4.0 paradigms[3]. Such definitions are coined combining the term ‘digital’, referring to the IT and automation landscape, and the terms ‘readiness’ and ‘maturity’, representing the state of being complete and ready.

The approach consists of introducing the so-called Digital Matrix defined by the cross-correlation between the digitalisation technological domains and the steel manufacturing areas obtained by the logical

1969

TODAY

splitting of the production chain, auxiliary plants included.

The digitalisation technological domains, taken in consideration are the following: • Connectivity and Data Sharing: It represents the capability of collecting, storing and sharing data for its further transformation into knowledge; this capability is measured considering the presence of sensors, instrumentation, data repositories and network infrastructure.

• Data Processing: It represents the capability of manipulating data in order to produce valuable information through validation, aggregation and analysis steps.

• Human in the Loop: Human-in-theLoop is the concept based on systems and models that require human interaction inside the computation chain for accomplishing tasks. It includes advanced human-machine interaction solutions as well as simulation systems.

• Advanced Modelling: It considers the availability of advanced process control and optimisation models, either of a deterministic

Steel Times International

16/05/2018 11:21:22

Connectivity & data sharing

Digitalisation matrix

Layer 1

Data processing Human in the loop

Automomous & robotic

Advanced modelling

Layer 2 Transversal digital techniques Digital techniques

Manufacturing coordination, execution & optimisation

or statistical nature, and technological packages.

• Autonomous & Robotic Systems: It considers the presence of autonomous and context-aware self-adapting systems as well as robotic solutions.

Each domain is then composed, applying the proper weights, by a given number of so-called digitalisation factors, which can assume a value between 0 and 10; such values are defined by the Danieli experts through benchmarking with a set of plants that are identified on the basis of products, grades, production route and overall productivity.

The following figure provides a graphical representation of the digital matrix, where the above-mentioned domains are grouped under the Layer 1: Layer 2 includes the so-called smart factory enabling solutions, which support the most critical plant-wide business processes, bringing a significant contribution towards digitalization too, but also represent, together with an advanced business and predictive analytics platform, the most proper places where to implement the virtuous circle ‘Measure-Gain Insights-Improve’.

In an integrated digital-ready plant, it is

Aluminium Inter-

danieli.indd 3

national Today

Fig. 2 Digital matrix – graphical representation

Fig. 3 Digital Readiness - Radar Chart Representation

almost mandatory to support the entire customer order lifecycle from its acquisition up to material dispatch, through planning, execution and material certification steps.

Since this support can be provided either by an extended MES or by an integrated solution combining ERP, APS and MES, its contribution to digital maturity is split at function level (order acceptance, operational planning, material certification) and each function has its proper weight in the overall contribution calculation.

The Digital Matrix is used to quantitatively assess the aforementioned digital readiness and digital maturity, which can be defined as follows: • The digital readiness (DR) represents the current digitalisation degree of the factory.

• The digital maturity (DM) represents the level of compliance of the factory’s digital landscape (architectures, systems, networking, sensors, etc.) with the smart factory and Industry 4.0 paradigms.

be summarised by overall plant efficiency, product quality, worker’s health and safety, environmental sustainability.

Conclusions The first implementations of the present assessment study already demonstrated its effectiveness as well as its capability to respond to today’s customer needs.

In fact they are not looking any more for product suppliers, but for tough business partners capable of driving them along their evolutionary path from the ‘running the business’ to the ‘improving the business’ approach.

References 1.

Boston Consulting Group, “Digital Acceleration

Index”, https://www.bcg.com/capabilities/technologydigital/digital-acceleration-index.aspx 2. Kane, G. C., Palmer D., Phillips A. N., Kiron D., Buckley N., “Achieving Digital Maturity - Adapting Your Company to a Changing World”, MIT Sloan Management Review in co-operation with Deloitte Digital,

An effective representation of the calculated digital indexes as well as its theoretical evolution considering the implementation of a given set of proposed improvements is given through a Radar chart, as represented in Fig.3.

The AS-IS situation given by the current values of the digital indexes allows Danieli to identify possible areas of improvement but, in order to define the related implementation roadmaps, it is mandatory for customers to share their objectives and the relevant priorities. In fact, each so-called smart factory enabling solution has a different impact on the possible objectives that can

July 13, 2017 3. Schumacher A., Erol S., Sihnab W., “A Maturity Model for Assessing Industry 4.0 Readiness and Maturity of Manufacturing Enterprises”, The Sixth International Conference on Changeable, Agile, Reconfigurable and Virtual Production, Procedia CIRP Volume 52, 2016, Pages 161-166 , 2016.

1. Danieli Automation S.p.A. Via B. Stringher 42, Buttrio (UD), 33042, Italy 2. Senior Consultant Danieli Automation Via Adua 34/C, Latina (LT) 04100, Italy Email: m.ometto@dca.it a.merluzzi@dca. itcostanzopietrosanti@gmail.com

FUTURE ALUMINIUM FORUM

3 16/05/2018 11:21:24

DIGITALISATION

New economics of steel The steel industry has an historic opportunity to change the economics of steel-making by adopting digital transformation across the life-cycle of its operations leading to significantly elevated sustained EBITDA, EBITDA/ton and return on capital employed. By Atanu Mukherjee1 and Kinnor Chattopadhyay2

W

orldwide the bane of excess capacity, commodity price swings, demand stagnation and operational flexibility and efficiency have led to depressed EBITDAs and global steel companies have rarely exceeded profitability levels above 10% on average in the last 10 years. Similarly, the EBITDA/ ton of profitable integrated steel companies worldwide have ranged from $50-80/ton, well below a sustainable level of $100120/ton. In developing countries like India, barring a few integrated steel companies, the EBITDA levels after discounting for the elevated cost of debt have struggled to breach the sustainable level of about 16%.

With worldwide excess capacity expected to recede slowly, flexible design, operational flexibility and operational efficiency enabled through digital technologies hold the key

to economic transformation and sustained competitive advantage for steel firms. Our research shows that applying specific integrative digital techniques can improve EBITDA margins by 5-7% points and EBITDA/ ton by $25-60/ton over baseline performing levels. Focused and realistic adoption of digital technologies in progressive steel firms could well push the productivity envelope to over $180/ton in terms of EBITDA/ton. In the life-cycle of an integrated steel plant, we think that the digitally enabled capital-investment and design phase, and digitally integrated operations execution provides the deepest impact in the steel industry.

1. Digital enablement of the capital investment cycle Engineering intelligent, flexible and productive capacities is a critical enabler for sustained

100% 90%

Cumulative probability

80% 70%

ENPV benchmark design

ENPV of flexible design concept 1&2 ENPV of flexible design concept 7

60% 50% 40%

ENPV of flexible design concept 3

ENPV of flexible design concept 4,5 &6

30% 20%

Improved downsides (e.g. P5)

Better upsides (e.g.

P95)

Better expected value

10% 0% -20

0

20

40

60

80 100 120 140 160 180 200 220

Net present value (S$ million)

Flexible design concept 1 Flexible design concept 4 Flexible design concept 7

Flexible design concept 2 Flexible design concept 5 Benchmark design

Flexible design concept 3 Flexible design concept 6

Fig 1. An indicative shift of the NPV curve with flexible design options for an integrated steel plant

86 digitalisation.indd 2

FUTURE STEEL FORUM

EBITDA trajectory transformation in steel plants. The capital investment cycle in steel plants is fraught with uncertainties both in the design and the project execution phase.

These uncertainties elevate the risk premium in terms of project finance and reduce life cycle cash-flows due to delays. This increases capital cost substantially, raises sustainable EBITDA barriers and reduces the return on capital employed. Evaluating options for flexible design of plants in the face of market and raw material uncertainties is likely to result in more flexible plants which can alter product-mix, substitute raw materials and trade-off between economies of scope and scale in iron-making, steel-making and downstream units. Market uncertainty modeling coupled with design options analysis using real-options models can yield realistic scenarios and shift the NPV likelihood curve with higher expected value and lower risks. An indicative shift of the NPV curve with flexible design options for an integrated steel plant is shown in Fig. 1 below Outcomes of such stochastic and complexity models of capital investment and flexible design options require largescale digital simulation to explore the solution possibilities. Options then need to be evaluated for operational integrity and optimal outcomes. This requires virtual simulation of the plant design options based on a digital operational plant model.

Once the design option is selected based on simulations, project execution needs to ensure that the design, engineering and construction is completed with minimal delays and cost escalations. Unfortunately, our current project management models are actually time and resource accounting models with linear dependence and do not capture or reflect the dynamics of project

Steel Times International

16/05/2018 10:50:04

execution and control. Large projects are inter-coupled, non-linear and have interacting feedback loops that exhibit what is known as dynamic complexity rather than combinatorial complexity. This makes large project executions inherently unstable. This means a change or an intermediate delay can have an exponential ripple effect on the end outcome in terms of time and cost. The impact of these changes and uncertainties across the project lifecycle can be significant, at times leading to cost and time overruns of well over 100% and beyond, leading to suspensions or eventual abandonments. Therefore, projects need to be structured, organised, financed and managed in a way that accommodates change, characterises risk while minimising late cost, functionality and schedule impacts. A digital project dynamics model captures the non-linear structure and form of such large projects, which lays the basis for characterising and understanding its behaviour. Fig. 2 below outlines a framework for modeling the dynamic behaviour of large engineering projects.

Dynamic project models are simulated in real time for predicting scenario outcomes and for dynamically assessing risks, which then aid rapid decision making in the face of changes and uncertainty during a steel plantâ&#x20AC;&#x2122;s project execution.

2. Digital transformation of plant operations Even todayâ&#x20AC;&#x2122;s best-in-class steel plant operations have significant headroom for improvement from both an operational efficiency and flexibility perspective. Digital technologies and artificial intelligence (AI) perhaps have the biggest immediate contribution to make in upstream iron making and steel-making, maintenance and supply-chain operations. High temperature metallurgical processes although tractable have several opportunity areas for improvement in productivity, quality and consistency of output. BOF steel-making for auto-grade steels in India, for example, can improve end-blow phosphorous strike rates by several percentage points using a combination of computational fluid dynamics, chemical reactions and deeplearning AI models. Similarly, integrating kinetics and thermodynamic models for

Steel Times International

digitalisation.indd 3

Feedstock

Market

Investment

Organise

Structure

Technology

Finance

Land

Labour

Insure

Manage

Dynamic Behaviour, Risk and Uncertainty Characterisation work to be done

Work being done

Really done

Customer changes

Causal models

Design Structure Matrix

Project Databases

Hybrid Simulations Unfertainty Models

Project Supply Chains

Flexibility and Options

Fig 2. Framework for modeling the dynamic behavior of large engineering projects

0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01

Strike rate improvement A B C

PHOS Levels

0-0.015% 0.016% >0.30%

B to A

6.6% C to B 2.2%

0

Incremental EBITDA Total incremental EBITDA per year

Incremental EBITDA

2.5 Million USD 7.5 Million USD 20 Million USD

Fig 3. An indicative shift of the phosphorous level strike rates in a BOF and its EBITDA impact based on a dynamic temporal ML model

blast furnaces with support vector machines can aid in improving blast furnace process parameters and productivity. Maintenance is one of the single largest cost line items in capital-intensive steel plants. A reduction of several percentage points is possible using a well-architected AI-based prescriptive and predictive maintenance system in upstream, like sinter plants and downstream facilities such as rolling mills. While condition-based monitoring systems have been around for a

while, distributed computational intelligence methods embedded in AI chipsets, advanced signal processing and accurate sensors for vibrations can make practical conditionbased monitoring systems possible.

However, deep learning and AI-based systems for plant operations have three distinctive challenges. Unlike traditional Machine Learning (ML) systems in areas like retail and credit card processing, understanding the nature of causality â&#x20AC;&#x201C;

FUTURE STEEL FORUM

87 16/05/2018 10:50:05

DIGITALISATION

or what causes what – for steel plant operating processes is crucial for its success.

Unfortunately, most machine learning approaches frequently use correlation as a proxy for causality. This may work some times, but is unlikely to work consistently in steel plant machine learning/AI applications.

Understanding of metallurgical processes is an essential first step in understanding causality in deep-learning models in steel plant operations. This may take different forms of metallurgical modelling, but we find computational fluid dynamics and thermo-chemical modeling are essential complements to machine learning algorithm design.

The second challenge has to do with the ongoing capture and conditioning of the deluge of near real-time data acquired for processing. Real-time signal streams from high-temperature processes have significant noise embedded in them. Unlike data cleansing in static AI systems, dynamic AI requires advanced signal processing to extract the signal from the noise on an ongoing basis. This requires intimate understanding of the applicability of the appropriate signal processing algorithms in the context of the process operations.

The third challenge in applying machine learning to steel plant operations is the time dimension. Unlike, finding a pattern or a behaviour at “a” point of time, most steel plant unit operations demand that the pattern of the variables be determined over the temporal dimension as it evolves. In BOF steel-making end-blow phosphorous level strike rate maximisation, for example, it is not only important for the AI algorithm to accurately estimate how much oxygen flow is optimal or how much fluxes should be added, but is equally important to estimate the determining difference in hitting the end point – which is the lance blow sequence and the flow pattern and the sequence of flux additions.

The latter requires advanced deep-learning algorithms, and much of the approach is not standard ML and forms the later stage of exploring for minimums in the feature space.

Some of the choices for such modelling are deep neural nets or random forests – possibly even recurrent nets (such as long short term memory).

However, operationally stable recurrent

88 digitalisation.indd 4

FUTURE STEEL FORUM

neural networks in dynamical systems require careful consideration to rules for tuning the learning weights. An indicative shift of the phosphorous level strike rates in a BOF and its EBITDA impact based on one of our BOF dynamic temporal ML models is shown in the Fig. 3.

Over time, operationally stable learning models adjust their parameters to push the strike rate envelope outward for better endpoint phosphorous control in the ML models.

Better supply chain operations – inbound, outbound and in-plant – have a significant EBITDA impact on operations. In some of our supply chain digital enablement and optimisation analyses we found that between raw materials, ladle movement tracking and optimisation, in-shop logistics and outbound dispatch, the effective improvement in EBITDA could range from $5-15/ton.

Digital simulation of supply chains yields important insights on the tracking and routing possibilities leading to improved cycle times, significantly lower demurrage, route optimisations, better crane sequencing, better resource utilisation and lower capital outlays.

All such digital enablement of the steel plants is closely tied to the capability and efficacy of sensor technologies. Smart sensors directly influence the availability and the quality of the signals and data required for process modelling, simulations and MLbased algorithms. Apart from low power smart sensors for physical parameters, vision is becoming an important dimension of smart sensing. High temperature and noisy processes in steel plants can take advantage of both optical vision and thermal imaging along with digital image processing to extract both equipment and material data in realtime. Tracking of ladles in noisy environments using a combination of thermal imaging, optical imaging, digital signal processing and image reconstruction can increase ladle lifetimes, reduce refractory wear and decrease number of ladles in circulation. Emissivity analysis based on optical imaging of liquid steel and slag can allow ML algorithms to estimate the state of an operating process, decreasing the cycle time and improving quality. Similarly, automated visual inspection of surface defects on rolled steel, by using computer vision and Artificial Neural Networks will increase throughput of flats as

visual inspection has proven to be a critical production bottleneck. Distributed intelligent sensors are, therefore, the essential enablers for digital transformation of operations in the steel plant.

3. Digital bridge to new EBITDA Comprehensive digital enablement of a steel plant from the concept stage to operations can thus transform how we conceive, design, build and operate steel plants. This can change the economics of steel making by changing the EBITDA trajectory, thus creating a durable competitive advantage for the steel firm.

Digital enablement and transformation spanning across the capital investment cycle and the on-going plant operations can serve as the bridge to elevating EBITDA, productivity and the return on capital substantially. In fact, contribution to EBITDA across the life-cycle can be over $60/ ton, which elevates the Return on Invested Capital to 20% levels. Apart from operations, flexibility and opportunity costs make a substantial contribution to the improved EBITDA. 4. Conclusion It has been conjectured that the steel industry in general has been conservative in the adoption of new technologies. It is true that compared to other fast moving industries like software and semiconductors, the “clock speed” – or the rate of change of technology – in the steel industry has been measured in decades. However, pervasive digital technologies and AI provide the steel industry with a unique opportunity to engender the new economics of steel making, as there is a substantial EBITDA and productivity motive.

It is likely that the early embracers of digital enablement and transformation will establish leadership positions in the industry with superior EBITDAs, higher return on capital and much higher levels of productivity.

1 President, M. N. Dastur & Co. (P) Ltd.P-17 Mission Row Extension Kolkata 700013, India Phone: +913322250500 2 Principal Consultant, Dastur Innovation Labs, 250 Yonge Street, Suite 2201, Toronto, ON M5B 2L7, Canada Phone: +(1) - 647-6-DASTUR, +(1) - 647-632 7887

Steel Times International

16/05/2018 10:50:05

Fives, digital solutions for operational efficiency PRACTICES AND METHODOLOGIES

Today’s plants are becoming “smart” and more agile.

Digital technologies are powerful tools to improve operational efficiency. Fives’ range of digital solutions improves manufacturing practices and methodologies: — Digital performance management and advanced process control — Improved operational efficiency — Ultimate quality process and reliability — Elimination of equipment failure — Enhanced environmental and operational safety

EYERON™, A QUALITY AND PROCESS WATCHDOG

Eyeron™ is a quality management tool that captures data from different steel processing lines in order to analyse it and give operators a clear view of the process. This data is then used to suggest smart solutions to metallurgical problems, resulting in increased productivity, greater efficiency and improved quality.

Eyeron™ replaces the need for five separate software tools and enables the comparison of data from different lines. It generates a potential yield increase from 92% to 98% in a year, with six times faster troubleshooting.

DIGITAL MAINTENANCE SYSTEM

Fives has developed a new, web-based system for on-line digital maintenance to solve maintenance problems, identify corrective actions and to easily order relevant spare parts.

The interactive system allows for on-line access of general and detailed 3D images in the areas where maintenance is required. At the same time, the system checks pertinent technical documentation and proposes relevant spare parts.

An operator can send an e-mail to Fives’ service department with just one click. The interactive system can be used by any operator with minimal training.

www.fivesgroup.com steel@fivesgroup.com

INNOVATIONS

Ma Steel orders continuous caster from Danieli Ma Steel (Maanshan Iron and Steel) has also ordered a two-strand, 12m radius continuous casting machine from Italian plant builder Danieli. It is claimed that the equipment will produce the largest beam blanks in the world.

The largest among the three beam-blank sections will measure 1300x510x140 mm and boast a linear weight of almost 2,700 kg/m.

The caster will be supplied with two twin-moulds, which will enable the casting of 550x280 mm blooms on four strands.

Ma Steel’s new continuous caster will be equipped with advanced Danieli technological packages, such as the Q-INMO oscillator with twin cylinders to support the big sections and provide full flexibility in guaranteeing top surface quality.

Full Danieli Automation L1 and L2 systems will include Danieli Liquid Pool Control System for ‘accurate and balanced secondary cooling’ according to Danieli.

As a result of this investment MaSteel will consolidate its leading position in the Chinese market for sections, with H-beams in sizes never reached before.

This project, part of Ma Steel’s 35-year strategy, will be the third long-product

conticaster project between Maanshan and Danieli in the last 15 years.

For further information, log on to www.danieli.com

Steel Dynamics revamps Columbus pickling line BTU Bridle Technology has received an order from SES Engineering LLC covering the design, construction and supply of three Umlauf bridles for a revamp of a push-pull pickle line operated by US-based Steel Dynamics Flat Roll Group (SDI). After the revamp, at the plant in Columbus, Mississippi, it is claimed that the pickle line will be the first in the world to use exclusively Umlauf Bridles to transport the strip through the line.

BTU claims that its Umlauf bridles build up strip tension high enough to ensure that thick-gauge and high-strength strip will leave the pickling line levelled. Once revamped, the pickle line will be able to operate at much higher speeds than before, as scale breaking will be much more intensive.

According to BTU, the latest 3.0 generation of Umlauf bridle technology is capable of applying much higher forces onto strip in processing lines than conventional units. Furthermore, it claims that Umlauf bridles distribute strip tension ‘extremely uniformly’ across the complete width of the strip.

The first of the three Umlauf bridles will be arranged directly behind the pay-off reel.

90

FUTURE STEEL FORUM

iNNOVATIONS PAGE 91.indd 2

There it will immediately bite the very first centimetres of the head of the strips, which can range from up to 13 mm thick and up to 1,880 mm wide, guiding them into the stretchleveller. The stretch-leveller will be supplied

by SES. The second Umlauf Bridle will be arranged behind the leveller, where strip tensions of up to 1,250 kN can be achieved.

An intensive scale breaking effect will result from elongation rates of 0.5% to 1%, making it possible to operate the line at speeds of up to 150 m/min.

As the strip is pulled through the leveller, no roller drive equipment is required in the leveller, reducing investment and maintenance costs and preventing the roller from slipping.

The second Umlauf bridle pushes the levelled strip into the pickling tank, and the third will be arranged at the run-out to bite the head of the pickled strip and guide it into the recoiler. At the same time, it creates the strip tension needed to produce exactly wound coils.

SES Engineering’s senior sales manager, Daniel Cullen comments: “The most important aspect for us was to find a technology that

would be able to reach very high strip tension and allow us to control that strip tension in a very precise way. Apart from that, the simplicity of the Umlauf principle convinced us: we will be able to set the right strip tension exactly where in the line we need it – without any conventional bridle rolls, driven rollers in the leveller or an additional braking unit.

Moreover, in future, stretch-levelling will require less strain energy as there is no bending of strip in the Umlauf Bridles.”

BTU Bridle Technology’s Michael Umlauf said that by moving the strip exclusively in a linear way the Umlauf bridles are superior to bridle rolls in several respects: “Using Umlauf bridles in connection with stretch-levellers means that levelling is performed primarily by stretching and only to a minor degree by bending. Actually, up to 90% of the levelling work comes from the stretching effect.”

The line is scheduled to come back on stream after the revamp in January 2019.

For further information, log on to www.btu-bridle.de Steel Times International

18/05/2018 12:20:09

OVER 120 YEARS IN STRIP FINISHING EQUIPEMENT

UNGERER STRIP FINISHING LINES From Coil to Coil | From Coil to Sheet | From Coil to Slit Strip

UNGERER Technology GmbH, specialist in plant, machinery and systems engineering, enables experienced, eco-designed, customized one-stop solutions and comprehensive services from a single source.

> Stretch-Bend-Levelling Lines > Cut-to-Length Lines > Slitting Lines > Roller Levellers > Unflatness Measuring System UMS > Electric & Automation > World Service

Find out more about our equipment portfolio

www.ungerer.com

UNGERER Technology GmbH | 75179 Pforzheim | Germany Ph. +49 7231 942-0 | E. info@ungerer.com

CONNECTING THE INTERNATIONAL STEEL INDUSTRY CTIVES

MAGAZINE

FORUM STEEL FUTURE try – 2018 steel indus the 4.0 and Industry 50 now! go to page

ORT ENCE REP

CONFER

scope the micro n under rence Digitalisatio Innovaction confe li’s at Danie

TUBE, PIPE

AND WIRE ROD

tube, ts in the developmen markets The latest pipe and steel wire rod

PERSPE

etaltech r of LTI-M Edgar Rayne questions our answers

DIRECTORY

mesint.com

www.steelti

mesint.com www.steelti - Vol.42 No2 March 2018

Order your copy priced from £78 or FREE for paid subscribers

17 ORY 20

– Vol.42 No.2

dd 1

march.in

WEB

DIRECT STI BUYERS

GUIDE.indd

17 ORY 20

17 14:15:13

06/09/20

1

WEEKLY E-NEWSLETTER

Packed with information on the steel industry and continually updated with news for steel professionals, steeltimesint.com also features special articles and interviews with leading industry figures.

@SteelTimesInt

STI_Full_page_Portfolio_A4_Ad.indd 1

18 11:02:10

15/03/20

T L DIREC

March 2018

STI Cover

@S

ATIONA

IONAL –

INTERNAT

CS 4.0

S ETHI

BUSINES

RY

DUST TEEL IN

TERN IMES IN STEEL T

The Steel Times International Directory is the essential guide to steel manufacturers, producers, suppliers of plant equipment and services to the steel industry.

STEEL TIMES

Steel Times International is the key publication for the steel market, reporting on iron and steel making issues from all corners of the globe.

A round-up of the top news stories is also sent to more than 11,000 industry professionals each week. You can register online to receive the weekly newsletter and keep up-to-date with the latest news from across the steel manufacturing industry.

www.steeltimesint.com/e-newsletter

WWW.STEELTIMESINT.COM 17/05/2018 13:05

www.harald-wehrle.de

Tools & Equipment

Consulting & Qualification

Engineering & Projects

Services & Spare Parts Since 1983, the Badische Stahl-Engineering GmbH (BSE) has been acting as a service provider for increasing the efficiency and productivity in the electric steel industry worldwide.

We are Steelmakers!

BSE is a sister company of the Badische Stahlwerke GmbH (BSW), one of the world’s most efficient Electric Arc Furnace steel plants.

BSW and BSE – a unique partnership that will help you to reach even ambitious goals.

Badische Stahl-Engineering GmbH Robert-Koch-Straße 13 D-77694 Kehl/Germany Phone (+49) 78 51/877- 0 Fax (+49) 78 51/877-133 eMail info@bse-kehl.de www.bse-kehl.de

BSE0372_Ad_NewPic_2018_210x297.indd 1

This unique partnership between BSW and BSE ensures that all products and services provided by BSE are not just based on mere theory, but on more than 4 decades of own proven operational experience.

24.01.18 10:21