An iterated local search algorithm for School Bus Routing Problem with Bus Stop Selection

The school bus routing problem is a variant of the well-known vehicle routing problem. The goal is to pick up the students allocated to some bus stops and generate routes, which include the selected stops, to carry the students to school. In this paper, we propose simple but effective metaheuristics approach that employs two features: first, it utilizes the set of large neighbourhood structures to make exploration deeper in each iteration. Secondly, the heuristics proposed executes an efficient transition between feasible and infeasible segment of search space. The idea of exploration in infeasible area is controlled by a dynamic adjusted penalty function that applied into infeasible solution. We develop two new metahueristics, called N-ILS (a variant of the Nearest Neighbourhood with iterated local search algorithm) and I-ILS (a variant of Insertion with iterated local search algorithm), which are only different in constructive phase. Both metaheuristics combine the constructive heuristic and a set of large neighbourhood search within an iterated local search context. A comprehensive statistical analysis was executed in order to recognize the best parameter settings of the algorithm and appropriately tuned the values of these components. The experimental results reveal that the N-ILS is able to obtain better solutions in lower computing times. Additionally, the results of the computational experiments implemented denote the N-ILS is very competitive in comparison to the best metaheuristic for the SBRP.