Graduate Seminar Series, April 17thJoaquin Torres, a Ph.D. candidate in ISEN, Texas A&M University

Dynamic and Robust Capacitated Facility Location in Time Varying Demand Environments

Short Bio.

Joaquin Torres received his Bachelor of Science degree inIndustrial and Systems Engineering in 2001 and his Master ofEngineering degree in Quality Systems and Productivity in 2003, bothfrom Tecnológico de Monterrey Campus Chihuahua. He has worked for Tecnológico de Monterrey Campus Chihuahua as a research assistant,lecturer, consultant, and coordinator of the Research Department of theSchool of Business and Humanities. He worked as the editor of theResearch Journal “Investigación Multi-Disciplinaria”. He also served asa member of the evaluation committee of Chihuahua’s Quality Award(Premios Chihuahua a la Alta Dirección y Calidad Total) in 2003.In 2004 he received a full scholarship for graduate studies fromCONACYT (Consejo Nacional de Ciencia y Tecnología) and joined the Industrial and Systems Engineering Department at Texas A&M University for doctoral studies. He is a member of the Logistics and Networked Systems Research Laboratory (LNS). His research interestareas are discrete optimization, facility location, and logisticsystems design. He is currently working on a joint research projectwith CONACYT and Tecnológico de Monterrey related to facility location in the public sector.

In general, facility location problems are to determine the optimal location of facilitiesand the allocation of the demand of customers to facilities with the objective of minimizing the associated costs of serving the demand andfor establishing or operating the facilities.The conventional facility location models available in the literature,usually called single-period or static, assume that demand and costsare known and do not change by time. In practice, however, demand isunknown and is changing by time. Dynamic location models assume thatdemand and cost parameters are time varying. Over a given time horizon,they determine the optimal time and location of facilities in order to minimize the total costs for serving demand and for operating andrelocating the facilities.In some practical situations, the relocation of facilities may not bepossible due to budget constraints. Such situations may arise in thepublic or private sectors where facilities such as power plants,hospitals, schools, etc. are expected to be operating for a long period of time. Assuming that the total demand is time varying, a possiblestrategy is to determine a fixed configuration of facilities which willremain operational over the entire time horizon.In this talk, the speaker will present mathematical models and efficient solution algorithms for dynamic and robust capacitated facility location problems. The total demand of customers is assumed tobe time varying over a discrete and finite time horizon, and can be split or served by one or more facilities. There are fixed costsassociated with establishing, operating, and closing facilities. Also,there is a variable transportation cost for serving the customers’demand. The main objective is to find an optimal sequence for locating facilities to satisfy the time varying demand while observing the capacity restrictions.