Scientific Programme > Plenary Sessions

Solution Methods for Optimization Problems under Uncertainty in
Logistics and Transportation

Abstract: Many  decision problems in logistics and transportation are dynamic and affected by uncertainty. Stochastic Programming provides a suitable modeling framework able to manage uncertain data in a multi-period decision making process. However, such kinds of problems are usually hard to solve since their size grows exponentially with the number of stages. For this reason approximation techniques which replace the original problem by a simpler one and provide lower and upper bounds to the optimal value are very useful in practice. In this talk solution  methods for  decision problems in logistics and transportation affected by uncertainty are discussed. We first present monotonic bounds based on the concepts of scenario grouping and we apply them to an inventory management problem whether the underlying probability distribution is known or not. Secondly, we discuss  the rolling horizon approach, obtained by solving the problem on a reduced time horizon, providing its worst-case analysis on a multistage waste-collection problem. Finally partial Benders decomposition methodologies are presented and applied to a stochastic multicommodity network design model showing the advantage of the proposed approach versus the classical one.



Short Bio:

Francesca Maggioni is Professor of Operations Research at the Department of Management, Information and Production Engineering (DIGIP) of the University of Bergamo (Italy). She graduated summa con laude in 2003 in Mathematics at the “Università Cattolica del Sacro Cuore” of Brescia (Italy) and completed her PhD in Pure and Applied Mathematics at University of Milano Bicocca (Italy), in 2006. Her research interests concern both methodological and applicative aspects for optimization under uncertainty. From a methodological point of view, she has developed different types of bounds and approximation for stochastic, robust and distributionally robust multistage optimization problems. She applies these methods to solve problems in logistics and transportation. On these topics she has published more than 60 scientific articles featured in peer-reviewed operations research journals. In 2021 her research has been supported as principal investigator by a grant from the Italian Ministry of Education for the project “ULTRA OPTYMAL Urban Logistics and sustainable TRAnsportation: OPtimization under uncertainTY and MAchine Learning”. She currently serves the EURO Working Group on Stochastic Optimization and the AIRO Thematic Section of Stochastic Programming as chair. She has been secretary and treasurer of the Stochastic Programming Society. She is Associate Editor of the journals Computational Management Science (CMS), EURO Journal on Computational Optimization (EJCO), TOP, International Transactions in Operational Research (ITOR), Networks and guest editor of several special issues in Operations Research and Applied Mathematics journals.

Questions!

(What questions should we be asking as a transportation science and logistics society?)

Abstract: I am frequently impressed by the great skill that our transportation science and logistics society has in solving problems and answering questions. We may wonder, however, whether we are always asking the right questions to begin with. I want to share my personal experiences on this topic. The goal of the talk is not to answer but to pose the query of what the right questions are, in the hope of instigating interesting and fruitful discussions. The scope of the talk will be on exact routing algorithms. Some fundamental questions will be highlighted, which have only recently been asked and answered in the scientific literature, of which we may wonder whether this could not have been done decades ago. Then, I will discuss some questions that I feel may be important in the pursuit of exact algorithms for routing problems with uncertainty. The content of the session includes recent work in which predominant modeling choices are evaluated for the vehicle routing problem with stochastic demands. Here it is common to enforce (1) that the expected cumulative demand of the customers on a route does not exceed the vehicle capacity, (2) that the number of routes is fixed, and (3) we even may allow negative realizations of demand when this sometimes has no real-world interpretation. The effect of these modeling choices on solution quality and computation time of exact algorithms are discussed.

Short Bio: Remy Spliet is an associate professor at the Econometric Institute of the Erasmus School of Economics at Erasmus University Rotterdam. He is high-performance member of the Erasmus Research Institute of Management (ERIM). He obtained his MSc. in Econometrics and Management Science at Erasmus University Rotterdam in 2009, in the track Quantitative Logistics and Operations Research. He obtained his Ph.D. at Erasmus University in 2013, with a dissertation on Vehicle Routing with Uncertain Demand. His field is Operations Research, specifically in the domain of transportation. His research interests include vehicle routing with and without uncertainty, exact algorithms, and last-mile logistics.