Hybrid CW-GA Metaheuristic for the Traveling Salesman Problem

The Traveling Salesman Problem (TSP) is one of the most challenging problems in Logistics and Supply Chain Management. Its relevance in routing planning and distribution has significant impact on reduction of operative costs for all enterprises. However, due to its NP-hard complexity, it is difficult to obtain optimal solutions for TSP instances. This paper describes a hybrid approach based on Clarke and Wright (CW) and Genetic Algorithms (GA) to provide near optimal solutions for the TSP. Performance of this meta-heuristic was assessed by comparing it with other well-known methods such as CW, GA and TabuSearch (TS). Results obtained from experiments with TSP instances corroborated the suitability of the hybrid approach for the TSP.