No abstract

The vehicle routing problem (VRP) with split delivery is an important branch of the classic VRP. The objective is to reduce the transportation distance and number of vehicles used. This study proposes a two-stage algorithm based on an improved sweep heuristic approach to solve this problem. We cluster customer points into a minimum number of groups through a multi-restart-iteration sweeping algorithm (MRISA). The load demands and split point in each group are fine-adjusted using the load rate and threshold coefficient. To reduce the travel distance in each group, an optimal route is produced by a tabu search algorithm (TSA). The computational simulation results demonstrate that the proposed method in this study is feasible and efficient, and the near-optimal performance of the proposed method is achieved regarding the transportation distance and computation time to the instances with a scattered distribution geographical characteristic; however, the strategy of "the maximum-minimum distance clustering first, and MRISA + TSA executing later" is much more useful for instances with a clustered-distribution geographical characteristic.

No abstract

No abstract

No abstract