Fuzzy TOPSIS and Goal Programming Approaches to Multi Objective Facility Location Problem for Emergency Goods and Services Distribution and Bornova/Izmir Case Study

Abstract

Distributions of vital goods and services in emergency or post disaster situations are of paramount importance to be able to meet the requirements of those in need on time. Finding an appropriate location for facilities to distribute such goods and services efficiently and quickly is an important challenge. In such a situation location decisions for these facilities must be made quickly considering multiple objectives. This problem is a multi-objective facility location problem (MOFLP). The main focus of this study is to present two solution methodologies for a MOFLP in a post disaster situation. We specifically consider objective of minimizing maximum weighted distance traveled and minimizing total cost of facilities to be opened in order to satisfy all demand. We also provide a version of the problem when the number of facilities to be opened is limited and second objective becomes maximizing demand covered. Due to the conflicting nature of the objective functions we propose to apply Fuzzy TOPSIS and Goal Programming and compare the solutions obtained using these two techniques with respect to solution quality and computational time. We present the developed models and provide results from a real-life application using existing emergency assembly areas and current census data for Bornova/Izmir. This paper contributes in two ways to existing literature. First is the comparison between multiple (two) solution methodologies for MOPLP. Studies in the literature provide only one solution technique such as Fuzzy TOPSIS Goal Programming etc. Secondly we implement these methodologies by using real life data for emergency situations.