A Two-Step Local Search Enhancement with the Novel Solution Representation for Solving Capacitated Vehicle Routing Problems

Combinatorial optimization problems like travelling salesman problem (TSP) and vehicle routing problem (VRP) are simple in statement but very much related to many practical problems in real world i.e., logistics, supply chain management (SCM), transportation problems and so on. Due to the increasing growth of businesses related to logistics and so on, solving TSP or VRP like problems with practical constraints become popular topic of research for many researchers and practitioners. Although numerous researches have been carried out yet no single algorithm or solution structure exists to solve any TSP or VRP especially when the problem becomes as complex as practical problem. However, recently particle swarm optimization gets considerable attention in handling combinatorial optimizations and a novel solution structure for CVRP has been proposed in our previous work. We have improved the proposed solution with two-step local searches in this article. As the solution representation is still very simple the program is very efficient in the terms of computational time. The proposed two-step local search enhanced technique is applied to the existing benchmark capacitated vehicle routing problems (CVRP) which shows the outstanding performance compared to the previous approaches.