An enhanced iterative three-component heuristic for the team orienteering problem with time windows and multiple deliverymen

Abstract

In this paper, we extend the team orienteering problem with time windows by considering multiple deliverymen. This variant arises from piratical situations where firms deliver goods to customers in congested urban areas, and particularly, the service time is a function of the size of crews who are assigned to the vehicles. In addition to the routing and scheduling decisions, the crew size for each route has to be decided. The resulting problem is called the team orienteering problem with time windows and multiple deliverymen (TOPTWMD). We propose an enhanced iterative three-component heuristic to solve this problem. The three components include (i) an iterated local search in feasible region, (ii) a tabu search in infeasible region, and (iii) a route recombination method for a high-quality solution. We conduct computational experiments on a set of test instances, and the results demonstrate the effectiveness of the proposed algorithm in terms of both solution quality and speed. Moreover, we substantiate the contributions of the joint use of the three components and the alteration between feasible and infeasible regions to the performance of the proposed algorithm.