CLASH: a heuristic to solve vehicle routing problems with delivery, pick-up and time windows

This paper presents a multistage heuristic to solve a vehicle routing problem with delivery, pick-up and time windows that is applicable to the public healthcare system. Blood banks constitute a vital part of public healthcare services. Their logistics include blood procurement, processing, cross-matching, storage, distribution, recycling, pricing, quality control and outdating. This paper seeks to study the operational decision-making in the distribution of blood. The problem is considered with the condition that pick-up should occur last. We propose a methodology that exploits the inherent features of the problem. Prior to a meta-heuristic search, we form clusters of nodes, determine the route to be followed within a cluster and assign vehicles to routes in a feasible manner. Using this as input, we develop a genetic algorithm with innovative operators for crossover and mutation and conduct a rapid search to find a near-optimal solution. Computational results show that the new heuristic can solve problems effectively. Some of the results are better than the best solutions that have appeared in the literature.