Francisco Saldanha-da-Gama
  • recherche

LAMIH seminar "Stochastic district problems

As part of the scientific animation of the Computer Science of LAMIH, a seminar will be held by Francisco Saldanha-da-Gama, University of Lisbon, Portugal.

  • Le 10/11/2022

  • 11:00 - 12:00
  • Mont Houy Campus - Malvache Building - Room 115

Summary:

In a districting problem (DP), one aims to partition a set of basic geographic areas, called territorial units (TUs), into a set of larger groups, called districts. These problems have many applications, including the design of commercial areas for sales force assignment, school systems, police district design, waste collection, commercial area design, distribution logistics, etc. Compactness, adjacency, and balance are three major characteristics sought in district design. An aspect of practical importance in DDPs is the need to accommodate demand that cannot be accurately predicted. In this presentation, demand is assumed to be represented by a random vector with a given joint cumulative distribution function. The resulting problem can be modeled as a two-stage mixed-integer stochastic programming problem. In the first stage, a decision about the initial design of the territory is made by maximizing the compactness of the districts. In the second stage, i.e., after the demand is revealed, the balancing requirements must be satisfied. The objective function considers the total expected cost that includes the cost of the territory design in the first stage plus the expected cost of outsourcing and reallocation in the second stage. The modeling framework discussed is evaluated using real geographic data. The results obtained are presented and discussed.*

Brief Biography:

Francisco Saldanha da Gama is a professor of operations research at the Faculty of Sciences, University of Lisbon. He has extensive teaching experience, both at the undergraduate and postgraduate level, in the areas of operations research, mathematical programming, discrete optimization, stochastic optimization and logistics. He has published regularly in international scientific journals, mainly in the areas of location analysis, supply chain management, logistics and combinatorial optimization. He has co-edited both editions of the volume "Location Science" published by Springer International Publishing. He has presented more than 150 papers at scientific events and has been invited to countless scientific events as a plenary, semi-plenary or keynote speaker. He has received several awards and honors. He is a member of various international scientific organizations such as INFORMS, the European Chapter on Combinatorial Optimization, the Working Group on Stochastic Optimization and the European Working Group on Location Analysis, of which he is a former coordinator. He is currently editor-in-chief of Computers & Operations Research and a member of the editorial advisory board of Journal of the Operational Research Society (UK), Operations Research Perspectives and Algorithms. His research interests include stochastic mixed integer optimization, location theory, and project scheduling.

.

Contact: Raca Todosijevic