Two-Stage Hybrid Scheduling Models And Opitimization Algorithms On Cross Docking Logistics

This paper studies two-stage hybrid scheduling models and optimization algorithms on cross docking logistics. Cross docking is kind of operation mode which moves goods directly from receiving point to shipping point. The objective is to eliminate the inventory-holding and order picking functions of a warehouse or distribution center while still allowing it to keep the receiving and shipping functions. Cross docking can increase the agility and decrease the inventory, more and more logistics companies are turning to cross docking-the logistics operation management which brings significant benefits including little or no inventory, low handling costs, low space requirement, centralized processing and low transportation costs.By analyzing the definition, history, types, conditions of success and advantages and disadvantages of cross docking, and combining of the applications and researches of cross docking, two-stage hybrid scheduling models on cross docking logistics has been proposed based on the characteristics of cross docking. Although the effect of cross docking is very good, it is still a big challenge for the researchers to find effective solutions within limited times in theory. At present, the classical cross docking scheduling model mainly focuses on classical two-stage scheduling model, in which only one vehicle exists in each stage. Based on the investigation on the third party logistics company, two-stage hybrid scheduling models on cross docking logistics has been proposed for the first time, in which multiple vehicles exist in each stage. Obviously, this model is much more complicated and the challenge to find good solutions is much bigger. However, the value of its real application is much bigger.Based on the different modes of vehicle transportation in cross docking logistics, two problems will be concerned:1) In the condition that suppliers or clients have large number of goods or large goods, direct delivery based two-stage hybrid scheduling model on cross docking logistics is proposed. Firstly, the effect of direct delivery in real logistics operation is described. Secondly, a mixed integer programming (MIP) is showed and solved by optimization software CPLEX for small scale problems. Further more, based on the characteristics of cross docking and hybrid scheduling, four heuristics are described and one lower bound is proposed to value the performance of heuristics. Finally, computational experiments are carefully designed to test the performances of heuristics and lower bounds based on several factors of the problem. The result shows that both the heuristics and lower bounds perform quite well, and the best heuristic differs within 6% from optimal for small scale problems.2) When suppliers or clients have small number of goods or small goods, in order to save the transportation cost, the integration of transportation for suppliers or clients is urgent. In such a condition, milk run based two-stage hybrid scheduling model on cross docking logistics is proposed. Firstly, the effect of milk run in real logistics operation is described. Secondly, a MIP is showed. Further more, based on the characteristics of cross docking and hybrid scheduling, two heuristics and one tabu search algorithm are described. Finally, computational experiments are carefully designed to test the performances of algorithms based on several factors of the problem. Result shows that the performance of tabu search can acheieve great improvement based on heuristics.This paper aims to provide decision support for the real cross docking operation of logistics companies. Also, the models and algorithms in this paper can be used to provide theory support for the cross docking scheduling management in logistics company.