EURO Doctoral Dissertation Award 2018 - Finalists

The three finalists of the 2018 EURO Doctoral Dissertation Award are:

 

 

Margarida Carvalho, Department of Mathematics and Industrial Engineering, École Polytechnique de Montréal

Computation of equilibria on integer programming games

 

The formulation of optimization problems as mathematical programming models is the standard way for describing and communicating real-world situations unambiguously. Very often, a decision process depends on external parameters controlled by other agents (players). This establishes the connection between mathematical programming and the game theory framework. The goal of this theses was to extend the algorithmic tools for single decision-maker combinatorial problems to games. Namely, we concentrate on non-cooperative games, where each player's set of feasible strategies is characterized through linear inequalities and integer requirements on player's decision variables. These are called integer programming games (IPGs). This work is fleshed out in two parts: we first considered two-round games, called Bilevel Programming problems, and then we focused on games where players select their strategies simultaneously. Although in both types of games each player goal is to maximize his/her utility, the approach to a solution significantly varies. We studied specific games that generalize classical optimization problems with applicability in robust optimization, homeland security, competitive lot-sizing and multi-agent kidney exchange programs. Besides classifying their computational complexity, a strong emphasis was given to algorithmic design which resulted in new methods and frameworks for solution computation. The thesis concludes by analyzing general simultaneous IPGs.

 

 

Martin Luipersbeck, Department of Statistics and Operations Research, University of Vienna 

Large-scale network optimization: Steiner tree problems

 

This work studies exact solution methods for the Steiner tree problem in graphs (STP) and other related NP-hard network design problems. The main focus is put on the development of practical computational techniques that are suitable for finding optimal solutions on large-scale problem instances. These efforts are motivated by recent applications of network design models in bioinformatics and pattern recognition. The work's core contributions involve the STP, the prize-collecting STP (PCSTP) and the stochastic STP (SSTP). For the STP, a node-based integer-linear programming (ILP) formulation is introduced which is shown to be particularly effective on notoriously difficult benchmark instances. An algorithmic framework integrating this formulation with other state-of-the-art techniques won the most categories among all participants during the 11th DIMACS Implementation Challenge. For the PCSTP, a branch-and-bound procedure based on dual ascent and reduction tests is designed by which several previously unsolved large-scale instances are solved to optimality. For the SSTP, an ILP formulation is presented which is shown to dominate known formulations and whose convenient structure is exploited in the design of a successful exact algorithm combining dual ascent, Lagrangian relaxation, and Benders decomposition. Additional contributions involve four related network design problems for which ILP-based algorithmic frameworks are introduced and computationally evaluated.

 

Maximilian Schiffer, School of Business and Economics, Chair of Operations Management, RWTH Aachen University

Logistics networks with intermediate stops - designing innovative and green solutions

 

Current challenges and future requirements for logistics networks like tremendous growth in small package shipping and extremely ambitious ecological targets at global and local level call for more flexibility, efficiency, and sustainability in logistics. Herein, logistics networks with intermediate stops constitute a promising concept to realize these goals, as they allow for freight replenishment in-between service stops to use smaller vehicles, for multimodal transportation in city logistics, and for recharging electric commercial vehicles (ECVs) en- route. Against this background, this dissertation focuses on the strategic design and operation of logistics networks with intermediate stops. Regarding its methodological contribution, it introduces a new class of location-routing problems (LRPs), the LRP with intra-route facilities (LRPIF), for the design and operation of such networks. It further presents a generic and competitive algorithmic framework for LRPIFs, but also for 14 different classes of vehicle routing problems with intermediate stops. This algorithmic framework represents a new state of the art in terms of solution quality, computational performance, and usability as it can be applied to a large variety of LRPIFs and VRPISs without additional tailoring. Moreover, these approaches are applied to a real-world case on the deployment of ECVs in mid-haul transportation. Accordingly, this dissertation derives deep managerial insights for fleet operators to foster the adoption of ECVs in logistics networks and thus helps to realize sustainable modes of transportation.



Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 International License and the GNU Free Documentation License (unversioned, with no invariant sections, front-cover texts, or back-cover texts).

Privacy Policy.

EURO-Online login

 

Sign Up for e-Newsletter

 

EURO Publications