Iterated local search heuristic for the team orienteering problem with time windows

Tourists have the difficult task to schedule their holiday activities in such a way that they enjoy their holiday as much as possible, while keeping the opening times and exact locations of attractions in mind. This task can be formulated as the team orienteering problem with time windows (TOPTW). The TOPTW uses a list of locations as input, together with their opening times and appreciation scores. The goal of the TOPTW is to maximize the collected appreciation scores by visiting locations, while considering the opening times and the travel times within locations. This thesis is a replication of earlier research done by Vansteenwegen et al. [13]. Their iterated local search (ILS) heuristic is implemented by us and the results are compared and certain claims made by them are verified. Our contribution is to make adjustments to the original heuristic in order to improve the results. Due to possible interpretation or implementation errors, the total score of our replication was 1.52% lower than the total score of Vansteenwegen et al. However, the total score of the best variant made by us is 0.28% higher than theirs, as the score of 75 test instances (out of 224) was improved. For 32 test instances the score of the best variant is higher than the best-known result until 2009.