July 11, 2022 (online)
Evolutionary Computation in Practice
In the Evolutionary Computation in Practice (ECiP) track, well-known speakers with outstanding reputation in academia and industry present background and insider information on how to establish reliable cooperation with industrial partners. They actually run companies or are involved in cooperation between academia and industry.
If you attend, you will learn multiple ways to extend EC practice beyond the approaches found in textbooks. Experts in real-world optimization with decades of experience share their approaches to creating successful projects for real-world clients. Some of what they do is based on sound project management principles, and some is specific to our type of optimization projects. If you are working in academia and are interested in managing industrial projects, you will receive valuable hints for your own research projects.
In 2022, ECiP will be a hybrid event. We will do our best to enable opportunities for establishing contacts among participants.
Session. Day: July, 11th, 2022. Time 10:40 – 14:10 Eastern Daylight Time (16:30-20:10 Central European Summer Time)
|10:30-11:00 EDT (16:30-17:00 CEST)
|Soft skills and soft computing:
when presenting the solutions
of an optimization problem
becomes harder than finding them
|University of Trento
|11:00-11:30 EDT (17:00-17:30 CEST)
|Evolutionary Algorithms Supported Decision
Making for Sustainable Cities
|University of Malaga
|11:30-11:50 EDT (17:30-17:50 CEST)
|12:50-13:20 EDT (18:50-19:20 CEST)
|Jumping in at the deep end: from university to industry
|Senior Data Scientist at OTTO FUCHS KG
|13:20-13:50 EDT (19:20-19:50 CEST)
|Eike Permin, Lina Castillo
|Consider the lathe – Why manufacturing provides an interesting playground for algorithms
|Innovation Hub Bergischen RheinLands
|13:50-14:10 EDT (19:50-20:10 CEST)
- Each presentation will be ~ 25 min with additional ~ 5 min of Q&A each.
The final 20 min of the session will be devoted to a more general discussion among the event participants.
- While GECCO 2022 is a hybrid event, ECiP will be held online since the event organizers and most of the speakers will attend remotely.
Soft skills and soft computing: when presenting the solutions of an optimization problem becomes harder than finding them
In this talk, I will briefly discuss some of the most challenging real-world applications of numerical optimization methods (including evolutionary and memetic algorithms) that I’ve been working on with my collaborators in the past few years, ranging from computer aided design to workforce management and scheduling. I will focus mainly on the issues that we have faced whenever we presented the results of an optimization process to stakeholders, the (almost unavoidable) fine-tuning that was often necessary to meet all the users’ needs, and the various kinds of reactions that we have observed (ranging from extremely enthusiastic to unexplainably disappointed). I will finally discuss what are, in my opinion, the main challenges that we need to address whenever evolutionary computation is applied in industry, and I will propose some tentative solutions that may facilitate the adoption of these methods in real-world applications.
Giovanni Iacca is an Associate Professor in Computer Engineering at the Department of Information Engineering and Computer Science of the University of Trento, Italy, where he founded the Distributed Intelligence and Optimization Lab (DIOL). Previously, he worked as postdoctoral researcher in Germany (RWTH Aachen, 2017-2018), Switzerland (University of Lausanne and EPFL, 2013-2016) and The Netherlands (INCAS3, 2012-2016), as well as in industry in the areas of software engineering and industrial automation. He was co-PI of the FET-Open project “PHOENIX” (2015-2019), and currently is co-PI of the PATHFINDER-CHALLENGE project “SUSTAIN” (2022-2026). He has received two best paper awards (EvoApps 2017 and UKCI 2012). His research focuses on computational intelligence, stochastic optimization, and distributed systems. In these fields, he co-authored more than 100 peer-reviewed publications. He is actively involved in the organization of tracks and workshops at some of the top conferences in the field of computational intelligence, and he regularly serves as reviewer for several journals and conference committees.
Evolutionary Algorithms Supported Decision Making for Sustainable Cities
In recent years, smart cities have been increasingly concerned with meeting sustainable development goals (SDGs). These goals encompass social, environmental, and economic aspects to make cities inclusive, safe, resilient, and sustainable. Pursuing these objectives requires a complex decision-making process because it requires considering simultaneously multiple, often conflicting, design criteria (such as the economical cost and the quality of service provided). This can be faced by modeling the cities and addressing the decisions as optimization problems realistically considering the constraints, the variables, and the (likely conflicting) objectives. Evolutionary Algorithms (EAs) have shown remarkable success in solving these problems, providing solutions that the decision-makers can consider. This talk aims to present the work carried out by our research group on using EAs to successfully face urban challenges related to waste management, electric vehicles, intelligent transportation systems, public transportation design, etc.
Jamal Toutouh holds a Tenure Track to Associate Professor at the University of Málaga (Spain). Previously, he was a Marie Skłodowska Curie Postdoctoral Fellow at Massachusetts Institute of Technology (MIT) in the USA, at the MIT CSAIL Lab. During his doctoral studies at the University of Málaga, Jamal analyzed and devised Evolutionary Computation methods to address Smart Mobility, one of the main challenges in modern sustainable cities. His dissertation was awarded several prizes, including the Best Spanish PhD Thesis in Smart Cities (2018). Currently, he is a Generative Machine Learning enthusiast. His research explores the combination of Nature-inspired, gradient-free approaches and gradient-based methods to address Deep Generative Modeling. All this while proposing solutions based on Evolutionary Computation to foster more sustainable societies.
Jumping in at the deep end: from university to industry
Jörg vividly reports on his experiences when changing from the university and shows how cold the water really hits you in the face as an academic when you start in industry.
Dr. Jörg Stork has been a full-time research assistant at the Technical University of Cologne. He was employed in the Faculty of Computer Science and Engineering, initially in the Institute for Computer Science (INF), in the Institute for Automation & Industrial IT (AIT), and then in the Institute for Data Science, Engineering and Analytics (IDE+A).
Furthermore, under the supervision of Prof. Bartz-Beielstein, Dr. Stork has also been involved many tasks, e.g.:
- Research and PhD in the “KOARCH” project: development of a reference architecture for cyber-physical systems in the context of Industry 4.0
- Sub-project management for the area of artificial intelligence in the “TalSich” project.
Jörg works as a Senior Data Scientist at OTTO FUCHS, KG.
Eike Permin, Lina Castillo:
Consider the lathe – Why manufacturing provides an interesting playground for algorithms
In this, we give a short intro into how manufacturing SMEs in Germany are currently tackling the digital transformation, how ecosystems and partner networks provide crucial resources for this transformation, and how manufacturing provides some interesting and serious playground for anyone interesting in algorithms and machine learning. To underline the latter, we present two example cases, one from process optimization in injection moulding and one from manufacturing scheduling considering energy and resource efficiency.
Eike Permin is an endowed professor for digitalization in manufacturing at the InnoHub. Industrial engineer and former general manager for a medium-sized, agile software and service company in the Steel industry. He possesses 10+ years of experience in international projects in manufacturing, factory organization, and process optimization. Go-to guy and problem solver with 6+ years of experience in cross-functional leadership. At the InnoHub, he primarily researches on the collection and processing of production data as well as its application, analysis, and evaluation. In addition, he is involved in cooperation in research and development projects and in the networks of the InnoHub.
Lina Castillo is a research associate at the InnoHub working with Eike Permin.
GECCO Schedule at a Glance (https://gecco-2022.sigevo.org/Program)