Relative Performance of Certain Meta Heuristics on Vehicle Routing Problem with Time Windows

—Solving Vehicle Routing Problem (VRP) and\udits variants arise in many real life distribution systems.\udClassical VRP can be described as the problem of finding\udminimum cost routes with identical vehicles having fixed\udcapacity which starts from a depot and reaches a number\udof customers with known demands with the proviso that\udeach route starts and ends at the depot and the demand of\udeach customer does not exceed the vehicle capacity is met.\udOne of the generalizations of standard VRP is Vehicle\udRouting Problem with Time Windows (VRPTW) with\udadded complexity of serving every customer within a\udspecified time window. Since VRPTW is a NP hard meta\udheuristics have often been designed for solving it. In this\udpaper we compare the performance of Simulated\udAnnealing (SA), genetic Algorithm (GA) and Ant Colony\udOptimization (ACO) for solving VRPTW based on their\udperformance using different parameters taking total travel\uddistance as the objective to be minimized. The results\udindicate that ACO is in general slightly more efficient\udthen SA and GA