An ELSxPath Relinking Hybrid for the Periodic Location-Routing Problem

摘要: The well-known Vehicle Routing Problem (VRP) has been deeply studied over the last decades. Nowadays, generalizations of VRP are developed toward tactical or strategic decision levels companies. extension plans a set trips multiperiod horizon, subject to frequency constraints. related problem is called Periodic (PVRP). On other hand, motivated by interdependent depot location and routing decisions in most distribution systems. Low-quality solutions obtained if depots located first, regardless future routes. In Location-Routing (LRP), simultaneously tackled. goal here combine PVRP LRP into an even more realistic covering all levels: PLRP. A hybrid evolutionary algorithm proposed solve large size instances First, individual representing assignment customers combinations visit days randomly generated. Then, heuristic based on Randomized Extended Clarke Wright Algorithm (RECWA) creates feasible solutions. evolution operates through Evolutionary Local Search (ELS) assignments. hybridized with Path Relinking between individuals from elite list. method evaluated three sets compared literature particular cases such as one-day horizon (LRP) one This metaheuristic outperforms previous methods for