Swarm intelligence approaches for multidepot salesmen problems with load balancing

Abstract

Given a set of n cities and m salesmen stationed at d depots, the fixed destination multidepot salesmen problem consists in finding tours for all the salesmen which start and end at their corresponding fixed depots such that each city is visited exactly once by one salesman only, the workload among salesmen is balanced and the total distance traveled by all the salesmen is minimized. In this paper, we have proposed two swarm intelligence approaches for this problem. The first approach is based on artificial bee colony algorithm, whereas the latter approach is based on invasive weed optimization algorithm. Computational results on instances derived from TSPLIB show the effectiveness of our proposed approaches over other state-of-the-art approaches for this problem.