An evolutionary multi-agent based search method for stackelberg solutions of bilevel facility location problems

This paper considers the bilevel facility location problem (BFLP) on a plane with the situation that a facility must be located at more than a certain distance from another facility. Since the strict solution method for finding a Stackelberg solution of the BFLP needs much computational time for large scale BFLPs, the evolutionary multiagent system is proposed for finding an approximate Stackelberg solution efficiently by utilizing characteristics of the BFLP. The efficiency of the solution method is shown by applying it to numerical examples of the BFLP.