Green Weber Problem and its Extensions
Asst. Prof. Dr. Mustafa Kemal Tural, Department of Industrial Engineering, Middle East Technical University
May 4, Friday 13:30
EA-409
Abstract: The Weber problem aims to locate a single facility in the plane so that the sum of weighted Euclidean distances between the facility and the customers is minimized. Sending vehicles from the facility to the customers for delivering/picking goods results in a significant amount of CO2 emissions affecting citizens’ quality of life and the climate. We consider an extension of the Weber problem, named as the Green Weber Problem (GWP), which decides on the location of the single facility in the plane and the speeds of the vehicles serving the customers within the one-sided time windows so as to minimize the total amount of carbon dioxide emitted in the distribution system. We also introduce time-dependent congestion which limits the vehicle speeds in different time periods and call the resulting problem as the time-dependent green Weber problem (TD-GWP). In the TD-GWP, the vehicles are allowed to wait during more congested time periods. We formulate the GWP and TD-GWP as second order cone programming problems both of which can be efficiently solved to optimality.
Moreover, we consider a multi-facility version of the GWP and formulate it as a mixed-integer second order cone programming formulation. As this formulation is only able to solve small size instances to optimality, well-known heuristics developed for the multi-facility Weber problem such as the alternate location-allocation heuristic, transfer follow-up heuristic, and decomposition heuristic are utilized in addition to a newly developed local search approach.
(Joint work with Arsham Atashi Khoei and Haldun Süral.)
Bio: Mustafa Kemal Tural received his B.S. degrees in Industrial Engineering and Mathematics from Bogazici University, Istanbul, Turkey in 2004 and his Ph.D. degree in Operations Research from the University of North Carolina at Chapel Hill, NC, US in 2009. Following his Ph.D., through the Institute for Mathematics and its Applications (IMA) in Minneapolis, he spent one year as an NSF industrial postdoc at Telcordia Technologies and one year as an NSF postdoctoral researcher in the Industrial Engineering and Operations Research Department at Columbia University. He joined the Department of Industrial Engineering at the Middle East Technical University, Ankara, Turkey in November 2011, where he is currently an assistant professor. His main research interests are in theory and applications of (integer) linear and nonlinear optimization and graph theory.