Abstract: This paper proposes an improved Bacterial Foraging Algorithm(BFA) to solve the Vehicle Routing Problem with Time Windows(VRPTW).The to-be-distributed customer points are clustered by K-means according to the geographical location,and the classification results are inserted into the optimal location of the distribution path in order within the time window,so the initial solution of the problem is constructed.Then the chemotaxis operation is combined with the removal operator in Large Neighborhood Search(LNS) for optimization,which expands the search range and improves the operation efficiency of the algorithm.Experimental results show that the algorithm can shorten the distribution path and minimize the overall distribution cost within the time window.
Vehicle Routing Problem(VRP),
Bacterial Foraging Algorithm(BFA),
large neighborhood search,
greedy insertion method