A DEA-based approach for finding a favorable multi-objective shortest path

Abstract

Multi-objective shortest path problems deal with networks in which each arc is associated with multiple types of costs and profits. The conditions can increase number of non-dominated paths and compounds difficulty of selecting a favorable one for Decision Maker (DM). In this paper we suggest a hybrid method to generate a non-dominated path which fulfills decision maker's demands. It uses DEA in a more logical way than some previous DEA-based approaches. We take advantage of common weights methodology. DM can give relative importance of some cost and profit indicators. Eventually, we have a shortest path problem that can be solved with Dijkstra's algorithm. Two numerical examples are presented and the results are analyzed.