Workshop on "ArtificiaI Intelligence Supported Decision Making in Industry"

dedicated to the findings and results of the FiDiPro project DECOMO: Decision Support for Complex Multiobjective Optimization Problems
Date: September 5th, 2017
Place: Room Alfa, Agora building, Mattilanniemi 2, University of Jyväskylä (JYU)


Prof. Yaochu Jin, University of Jyväskylä, Finland, and University of Surrey, UK

Bio: Yaochu Jin is a Professor in Computational Intelligence, Head of the Nature Inspired Computing and Engineering (NICE) group, Co-Coordinator of the Centre for Mathematical and Computational Biology (CMCB), Department of Computer Science, University of Surrey. I am also a Finland Distinguished Professor (2015-17) with the Industrial Optimization Group, Faculty of Information Technology, University of Jyväskylä, and a Changjiang Distinguished Professor, Northeastern University, China. His main research interests include evolutionary computation, machine learning, computational neuroscience, and evolutionary developmental systems, with their application to data-driven optimization and decision-making, self-organizing swarm robotic systems, and bioinformatics. He has (co)authored over 200 peer-reviewed journal and conference papers and has been granted eight patents on evolutionary optimization.

Prof. Kaisa Miettinen, University of Jyväskylä, Finland

Bio: Kaisa Miettinen is Professor of Industrial Optimization and vice-rector of the University of Jyväskylä. Her research interests include theory, methods, applications and software of nonlinear multiobjective optimization including interactive and evolutionary approaches and she heads the Research Group on Industrial Optimization. She has authored about 150 refereed journal, proceedings and collection papers, edited 13 proceedings, collections and special issues and written a monograph Nonlinear Multiobjective Optimization. She is a member of the Finnish Academy of Science and Letters, Section of Science and the Immediate-Past President of the International Society on Multiple Criteria Decision Making. She belongs to the editorial boards of six international journals and the Steering Committee of Evolutionary Multi-Criterion Optimization. She has worked at IIASA, International Institute for Applied Systems Analysis in Austria, KTH Royal Institute of Technology in Stockholm, Sweden and at Helsinki School of Economics in Finland. She just received the Georg Cantor Award of the International Society on Multiple Criteria Decision Making.


Keynote: Prof. Kalyanmoy Deb, MSU, USA

Optimizing Designs for Real-World Practicalities: Two Case Studies from Auto Industries

Abstract: Optimal design of real-world engineering products involves various practicalities which are often ignored in mainstream academic research. Some of these practicalities arise due to the need for (i) handling multiple conflicting objectives, (ii) using computationally expensive evaluation procedures, (iii) finding multiple and diverse equally good solutions, (iv) finding robust solutions, (v) optimizing with a few high-fidelity evaluations, (vi) handling discrete and mixed type of variables, etc. In this lecture, we shall present two design case studies from automobile industries and highlight the extended methodologies to existing generic multi-criterion optimization algorithms to address some of the above practicalities. From both cases it has become evident that satisfactory designs can be achieved through an appropriate use of different optimization concepts.

Bio: Kalyanmoy Deb is Koenig Endowed Chair Professor at Department of Electrical and Computer Engineering in Michigan State University, USA. Prof. Deb's research interests are in evolutionary optimization and their application in multi-criterion optimization, modeling, and machine learning. He has been a visiting professor at various universities across the world including IITs in India, Aalto University in Finland, University of Skovde in Sweden, Nanyang Technological University in Singapore. He was awarded Infosys Prize, TWAS Prize in Engineering Sciences, CajAstur Mamdani Prize, Finland Distinguished Professor, Honorary Doctorate degree from University of Jyvaskyla, Distinguished Alumni Award from IIT Kharagpur, Edgeworth-Pareto award, Bhatnagar Prize in Engineering Sciences, and Bessel Research award from Germany. He is fellow of IEEE, ASME, and three Indian science and engineering academies. He has published over 450 research papers with Google Scholar citation of over 103,000 with h-index 104. He is in the editorial board on 20 major international journals. More information about his research contribution can be found from

Keynote: Prof. Dr. Karl-Heinz Küfer, Fraunhofer ITWM

Industrial applications of multicriteria decision support

Abstract: Most decisions in life are compromises: typically several objectives arising from the families cost, quality, time, environmental impact have to be balanced. This process is not easy because one cannot have best possible values for all of these goals as they are at least partially in conflict. Many decision makers are reluctant with respect to introducing decision support tools that directly show what the possible freedom of choice or inherent restrictions are. They often do not want to defend personal preferences or biases in decision rounds which would become obvious by showing options and limitations in a fair way. The talk will demonstrate and discuss three examples of decision support tools in medical therapy planning, chemical process engineering and in the layout of renewable energy facilities, all of them in industrial practice for 5 years and more. Special attention is paid to the reception of such concepts in the companies and their impact if successfully implemented.

Bio: TBA.

Dr. Markus Olhofer, Honda Research Institute Europe GmbH

Multiobjective optimization in complex systems

Abstract: Real world optimisation tasks require the simultaneous consideration of various quality criteria of a system. The continuous advance in multi-disciplinary simulation methods and tools, the increase in computing power as well as the availability of suitable optimisation algorithms enables this simultaneous optimisation of a growing number of quality criteria. Nevertheless it is still current practice in industry to optimise single components under consideration of only a few selected criteria due to several reasons. In the presentation three selected examples of real world applications from the automotive domain are given and current obstacles in many objective optimisation are discussed. From that an outlook is given on how to support the application engineer to overcome the problem of selecting one final solution and to navigate in the high dimensional set of Pareto optimal solutions resulting from optimisation.

Bio: Markus Olhofer received the Dipl.-Ing. degree in electrical engineering and the Ph.D. degree from Ruhr University Bochum, Bochum, Germany, in 1997 and 2000, respectively. He joined the Future Technology Research Division with Honda Research and Developments Europe, Offenbach, Germany, in 1998. He has been the Chief Scientist and the Head of the Complex Systems Optimisation and Analysis Group with the Honda Research Institute Europe, Offenbach, since 2010. He is a Visiting Professor with the Department of Computer Science, University of Surrey, Guildford, U.K. His research interests include the extension of soft computing methods to meet requirements in complex engineering problems, ranging from evolutionary design optimization to engineering data mining.

Mr. Pekka Makkonen, Valtra

Shape optimization of an air intake ventilation system

Abstract: The presentation offers insight into multiobjective optimization of a tractor cabine ventilation channel geometry. The objective function values were determined using computational fluid dynamics which was considered computationally expensive. The K-RVEA algorithm was used for optimization. The final step in the process was done by a decision maker who picked the most suitable design.

Bio: Pekka Makkonen is a research engineer in Valtra Inc. that develops, manufactures, markets and services Valtra tractors.

Dr. Tinkle Chugh, JYU

Handling computationally expensive multi/many objective optimization problems

Abstract: Many real-world industrial optimization problems have several conflicting objectives to be optimized. Moreover, these objectives are usually conflicting in mature and their evaluation takes a substantial amount of time. In this talk, I will talk about a recently proposed Kriging-assisted reference vector guided evolutionary algorithm for computationally expensive optimization problems with more than three objectives. The algorithm uses Kriging models as surrogates to approximate each objective function to reduce the computational cost. The algorithm focuses on the balance of diversity and convergence by making use of the uncertainty information in the approximated objective values given by the Kriging models, the distribution of the reference vectors as well as the location of the individuals for managing the Kriging models. Moreover, a strategy is designed for choosing data for training the Kriging model to limit the computation time without impairing the approximation accuracy. Results on benchmark and a real-world shape optimization problem demonstrate the potential of the proposed algorithm.

Data driven multiobjective optimization

Abstract: Many real-world optimization problems rely on the data available e.g. obtained through some physical experiments. In such cases, one does not have the simulators, codes or analytical form of the objective functions. Such problems where only data is available to do optimization are called as data-driven optimization problems. In this work, we do several numerical experiments on benchmark problems and propose a data-driven multiobjective evolutionary algorithm. We use different kinds of data generated e.g. randomly, with some design of experiment, mixture of optimal and random. The proposed algorithm uses Kriging models to approximate the objective functions based on the data available. In addition, we demonstrate solutions obtained with the algorithm in terms of their uncertainty, which may be useful for an expert or a decision maker when selecting final solution(s) to be implemented.

Bio: Tinkle Chugh is a Postdoctoral researcher with the Faculty of Information Technology, University of Jyvaskyla, Finland. Tinkle obtained his Bachelor and Master degree in Chemical Engineering and PhD in Mathematical Information Technology. He won the best student paper award at IEEE Congress on Evolutionary Computation (CEC) 2017. His research interests include evolutionary computation, surrogate-assisted single, multi and many-objective optimization and preference incorporation in evolutionary algorithms.

Dr. Jussi Hakanen, JYU

Interactive surrogate assisted multiobjective optimization

Abstract: To enable decision making for computationally expensive multiobjective optimization problems, surrogate-based approaches must be combined with decision maker (DM) preferences in order to steer the solution process towards practically relevant solutions. The overall idea is to include human DM into the solution process interactively, that is, the preferences of the DM are asked several times and incorporated in to the search process. This is especially usefull for problems with more than three objectives where the visualization of the Pareto front becomes challenging. In this talk, we show how decision making strategies from the multiple criteria decision making (MCDM) field can be incorporated into the evolutionary multiobjective optimization (EMO) approaches in different ways.

Bio: Jussi Hakanen is a senior researcher with the Faculty of Information Technology at the University of Jyvaskyla, Finland. He received MSc degree in mathematics and PhD degree in mathematical information technology, both from the University of Jyvaskyla. His research is focused on multiobjective optimization with an emphasis on interactive multiobjective optimization methods and computationally expensive problems. He has participated in several industrial projects involving different applications of multiobjective optimization, e.g. in chemical engineering. He has been a visiting researcher in Cornell University, Carnegie Mellon, University of Surrey, University of Wuppertal, University of Malaga and the VTT Technical Research Center of Finland. He also has a title of docent in industrial optimization at the University of Jyvaskyla.

Dr. Karthik Sindhya, JYU

Software prototype – Demo

Abstract: We demonstrate a software prototype that is based on the methods developed in the DeCoMo project. It can be used to solve optimization problems where the evaluations of the objective and constraint functions are time consuming and have a high number of objective functions. In other words, the optimization methods combine multiobjective evolutionary algorithms with surrogates. Finally, the methods included can interact with human decision maker's and take into account their preferences in identifying the most preferred compromise between the conflicting objectives.

Bio: Karthik Sindhya is a researcher at the Department of Mathematical Information Technology, University of Jyvaskyla Finland. He has a bachelors and masters degree in chemical engineering and a PhD in multi-criteria optimisation. His PhD thesis on hybrid algorithms was supervised by Professor Kalyanmoy Deb and Professor Kaisa Miettinen and was nominated as one of the top three dissertations in multiple criteria decision making during the period: 2010-2013. He was also the recipient of the "Best algorithm award" at IEEE CEC 2007 and best poster award at MCDM 2011 conferences. His research interests include evolutionary algorithms especially evolutionary multi-objective optimisation (EMO) algorithms, handling computationally expensive objective functions that arise in industry and multiple criteria decision making approaches. He currently co-supervises three PhD students in the area of multi-criteria optimisation and is actively collaborating with Professor Yaochu Jin on a Finnish Distinguished Professor project called DeCoMo. In addition to his academic duties, he is an executive board member of the Finnish Operations Research Society and also a co-founder of FINNOPT Ltd. a consultancy services company based in Finland providing optimisation services to several global companies.