Research On Order Processing Optimization In Picking Station Based On Parts-to-picker Mode

Under the background of trade globalization,e-commerce consumption is developing rapidly,driving the explosive growth of e-commerce orders.Due to the characteristics of e-commerce orders,such as multi-variety,small batch and highfrequency,and customers are highly sensitive to the order response time,increasing the difficulty of order processing operations in distribution centers,and the efficiency of warehouse order picking urgently needs to be improved.With the increasingly mature exploration of intelligent algorithm in the logistics industry,the intelligent parts-to-picker logistics system begins to apply to warehouse operations,greatly improving the safety and accuracy of picking and storage,and reducing human labor.Compared with traditional picker-to-parts warehouses,the parts-to-picker system is more intelligent and warehouse management is more complicated,so many research problems have arisen.Due to the limited number of orders and racks that can be processed simultaneously by the picking station,the storage assignment of different types of goods are diverse,so the operational efficiency of different warehouse management strategy combinations is quite different.The order processing problem in the picking station is to analyze how to arrange the order and rack sequence in the parts-to-picker mode,so as to minimize the total handling times of racks when all orders are finished picking,thus reducing the total expenditure and improving systematic order picking efficiency.Based on the movable rack warehouse,this paper establishes an integer programming model to optimize the order processing activities of a single picking station.Since this problem is NP-hard,this paper designs VNS-OS,VNS-RS an AH algorithms to solve larger scale examples.In the case of small-scale example,comparing the known optimal solutions obtained by CPLEX 12.8,the average optimization degree of the three algorithms is between 6.4% and 16.6%,which proves the feasibility and effectiveness of the algorithm designed in this paper.Based on the analysis of the results of small,medium and large scale calculations,it is verified that the optimization effect of the two VNS algorithms on the initial solution is obvious,and the average optimization degree is higher than 10%.It is concluded that VNS-OS has the best overall solution performance,VNS-RS is suitable for the case of small scale or the large scale case with large processing number of orders,and in other cases choosing AH saves time.In addition,this paper also analyzes the sensitivity of the total rack handling times and rack utilization,both of which are negatively correlated with the maximum number of orders that can be processed simultaneously.In terms of storage location allocation,the storage strategy allocated according to the association rules is much better than the random storage,the total rack handling times are reduced by 13.2% on average,and the rack utilization rate is greatly improved,which provides a basis for the actual operation of the parts-to-picker storage system.