Adapted Ant Colony Optimization for Large-Scale Orienteering Problem

Abstract

he orienteering problem (OP) is a challenging route optimization problem with the objective of finding an optimal subset of nodes and the optimal path to visit these nodes so that the total profit under the constraint of the cost is maximized. Since ant colony optimization (ACO) algorithms have shown remarkable performance in solving path planning problems, they are likely promising for solving OP. To verify this, this paper takes the first try to adapt five classical ACO algorithms, namely Ant System (AS), Elite Ant System (EAS), Rank based Ant System (ASrank), Min-Max Ant System (MMAS), and Ant Colony System (ACS), to solve OP. To this end, we determine the candidate nodes in the path construction by considering the constraint. To verify the optimization effectiveness of the five ACO algorithms in solving OP, we also take the first attempt to conduct experiments on large-scale OP instances. Experimental results have shown that the five ACO algorithms are very promising for OP and ASrank obtains the best performance on the large-scale OP instances.