Solving the orienteering problem using attractive and repulsive particle swarm optimization

The initial study of this research applied the particle swarm optimization (PSO) heuristic to the orienteering problem (OP). PSO is a fairly new evolutionary heuristic-type algorithm created by Drs. Eberhart and Kennedy in 1995. Similar to ant colony optimization, motivation for PSO is nature-based on fish schooling and bees swarming. The OP is a variation of the well-known traveling salesmen problem (TSP) and is an NP-hard benchmark problem. Given a set of nodes with associated scores, the objective of the OP is to find a path that maximizes the total score subject to a given time (or distance) constraint. This paper presents an attractive and repulsive particle swarm optimization (ARPSO), which prevents PSO's weakness of premature convergence by maintaining solution diversity while retaining a rapid convergence. The ARPSO solves the OP with significant improvement in results when compared to PSO and is more competitive to known best published results.