Plenary talks
Richard Hartl
|
|
Richard Hartl was born in Vienna, Austria, in 1956, and studied Business Mathematics (Master and PhD) at the Vienna University of Technology. Afterwards he was Post-Doc at the University of Toronto and after his return to Vienna he was Assistant and Associate Professor at the TU Wien, where he obtained his “Habilitation” in 1987. From 1993 to 1995 he was Full Professor and Vice Dean at the Faculty of Economics and Management, Otto-von-Guericke University of Magdeburg, Germany. Since 1995, he has been Full Professor and Chair of Production and Operations Management at the University of Vienna, where he also served as Head of Department and Dean of Studies. Since 2007 he has also been Senior Extramural Fellow of the Center for Economic Research (CentER), University of Tilburg, The Netherlands.
Prof. Hartl has covered a wide are of subjects in research and teaching, and has received several teaching awards, such as the main teaching award of the University of Vienna in 2014. He has also been active in various consulting projects with the industry.
He has also served in various functions for scientific organizations, e.g. treasurer, and president of the Austrian Society for Operations Research (ÖGOR), member of the executive board of the German Academic Association for Business Research (VHB), and treasurer of the International Federation of Operational Research Societies (IFORS).
He has been associate editor or special issue editor of several leading OR journals, e.g. Transportation Science. He has written more than 200 papers in leading international journals with more than 18.000 citations in total (h-index 62). His main research areas are quantitative methods in production, operations management, and transportation, as well as metaheuristic and hybrid optimization methods. In 2021, he received the science award for outstanding achievements in the development of Operations Research by the German Society for Operations Research (GOR).
Prof. Hartl ranks among the top 1% of researchers in business studies in Europe.
|
Horizontal Collaboration in Transportation Logistics: Recent Developments
Date: Monday, June 13, 9.30h - 10.30h
Room: Auditorium
Abstract:
Cost pressure and environmental considerations have incentivized firms to use their equipment more efficiently and to avoid idle capacities. This has led e.g. to the exchange of transportation requests between smaller carriers, when some face overload situations and others have idle resources. This talk will focus on Horizontal Collaboration in Transportation Logistics, where carriers at the same level of the supply chain exchange transportation requests. Carrier collaborations can be organized either in a centralized or in a decentralized form. The former case assumes that a fully informed coordinator redistributes the requests among carriers. This is often unrealistic, since firms are typically reluctant to reveal information about their customers. While centrally planned collaborations are commonly used for benchmarking, recent work is focusing on decentralized approaches, where collaboration partners do not have to share critical information. Decentralized systems need a smart mechanism design such that important properties can be reached. These include fair sharing of costs, in order to encourage players to maintain the collaboration and not to act strategically (incentive compatibility). Furthermore, budget-balance, efficiency, and group as well as individual rationality are seen important characteristics of valuable mechanisms. Unfortunately, all these desirable properties cannot be satisfied at the same time. The basic 5 steps of decentralized collaboration using collaborative auctions will be reviewed. Collaborative auctions are needed, since it is important to exchange bundles of requests instead of single requests, since the value of a bundle is often higher than the sum of the values of the individual requests. Furthermore, only assigning a single bundle to carriers guarantees feasibility of the redistributed assignment. While previous work has mainly focused on static collaboration problems, the ultimate goal is to solve dynamic collaboration problems, where there is a flow of customer requests coming in, and carriers have to decide on acceptance and rejection immediately. Then, at several points in time, combinatorial auctions or other exchange mechanisms are used to obtain a more efficient fulfillment of requests. In a first step, we look at a single day, where first customer request come in and have to be rejected or accepted. Then a single round of redistribution of requests is performed, and afterwards the requests are fulfilled. We investigate the possibility of overbooking. Also considered are variants with time windows where the benefit of collaboration can be different.
Kenneth Sörensen
|
|
Kenneth Sörensen obtained his PhD entitled "A framework for robust and flexible optimization using metaheuristics" from the University of Antwerp in 2003 and continued as a post-doctoral researcher at the UAntwerp. From 2006 to 2009, he worked as a postdoc researcher of the Flemish Fund for Scientific Research (FWO) at the Faculty of Engineering of the KULeuven. In 2009, Kenneth Sörensen was appointed as a Research Professor of the Faculty of Applied Economics of the University of Antwerp, a position he holds to date.
Within this Faculty, he founded the ANT/OR research group, which focuses on applications of operations research and currently consists of three professors and about ten PhD students. Under his supervision, thirteen PhD's were obtained.
Kenneth Sörensen has published a large number of articles in international refereed journals and has presented his work at numerous scientific conferences. He is considered one of the world’s leading experts in the field of metaheuristics. His main research interest lies in the application of advanced (metaheuristic) optimization methods and in the development and study of optimization methods.
Kenneth Sörensen is the founder and current coordinator of EU/ME – the EURO working group on metaheuristics, the largest online platform for researchers in metaheuristics worldwide.
|
Meta-analysis of metaheuristics - the effect of the "A" in ALNS
Date: Wednesday, June 15, 9.30h - 10.30h
Room: Auditorium
Abstract:
The development of heuristic optimization algorithms has been an active area of research for several decades. This huge body of research has resulted in a variety of metaheuristic frameworks, based on different paradigms. State-of-the-art heuristics developed according to the guidelines expressed in these frameworks can now solve large and complex real-life optimization problems ever more efficiently.
However, it could be argued that the research community's focus on algorithmic performance has not been conducive to the production of generalizable knowledge. Many of the 'facts' that are 'well-known' in the field seem to be based on folklore rather than on scientifically proven results. As a consequence, heuristic development is still primarily an art that relies almost exclusively on the developer's previous experience. Guidelines that are backed by scientifically established facts are few and far between. Faced with a new problem, a developer is generally forced to resort to a trial-and-error approach when deciding on the specifics of the algorithm.
In this talk, I discuss a methodology that can be used to obtain generalizable knowledge in the field of metaheuristics. Meta-analysis, a systematic statistical examination that combines the results of several independent studies into a single conclusion, is frequently used in e.g., the medical sciences to determine the efficacy of a new treatment. To illustrate the approach, I discuss a meta-analysis that has been carried out to gain insights into the importance of the adaptive layer in adaptive large neighborhood search (ALNS). Although ALNS has been widely used to solve a broad range of problems, it has not yet been established whether or not adaptiveness actually improves the performance of a heuristic based on ALNS. The meta-analysis discussed in this talk settles this debate, demonstrating that obtaining generalizable knowledge in the field of metaheuristics is possible.
|
