Location Allocation And Order Batch Optimization In E-commerce Warehouse Pickup System

As the scale of e-commerce transactions continues to increase,the advantages of the goods-to-man picking system are becoming more and more apparent,and more and more large e-commerce warehouses have been laid out with goods-to-man mobile shelf As the scale of e-commerce transactions continues to increase,the advantages of the goods-to-man picking system are becoming more and more apparent,and more and more large e-commerce warehouses have been laid out with goods-to-man mobile shelf picking systems.At the same time,consumers’ requirements for delivery time frames and logistics costs are getting higher and higher,and order picking costs,which account for 65% of the total picking costs of the goods-to-person system in e-commerce warehouses,are the focus and difficulty for enterprises to improve the efficiency of warehouse operations and reduce operational costs.Therefore,this paper starts from the order picking process in the goods allocation and order batching segments,and achieves order picking cost reduction through optimization models and algorithms.The order picking process in the goods-to-person picking warehouse interacts with each other.The goods allocation process determines the specific storage location of the goods to be picked,the results of goods allocation are the data input for order batching,and one of the bases for order batching is the number of shelf moves,and the number of shelf moves is determined by the results of goods allocation.The results show that the best optimization effect can be achieved by optimizing both together,and sensitivity analysis is performed on important parameters to reveal the change law of picking cost after order batching and provide relevant suggestions for e-commerce warehouse picking process optimization.The specific research contents are as follows.(1)Application of FP-growth algorithm to mine commodity association rules.In this paper,a total of 3003 association rules are mined from 15,000 historical orders in the e-commerce warehouse,and the support,confidence and elevation degrees of any association rules are output,and the elevation degrees are used to represent the association degree of different association rule commodity combinations,and the association rules are visualized and analyzed.The association rule results are used as the data input for the goods allocation,which is the basis for the solution of the goods allocation model.(2)The seed algorithm is applied to solve the space allocation model with the objective function of the same shelf merchandise with the maximum association degree.In this paper,the seed algorithm solves the space allocation model in two steps: the selection of seed goods and the insertion of remaining goods.The high frequency items with high support,i.e.,appearing in the order more often,are selected as the seed items,and the strong association rules containing the seed items are selected for insertion into the same shelf.(3)Apply the greedy algorithm idea to design the order batching algorithm process.The optimization objective of the order batching model is to minimize the order picking cost,which includes the labor cost of picking staff to pick goods from the shelf and the cost of AGV to carry the shelf.The weighted similarity measure is defined to include the similarity of the items and the shelves that need to be moved,and dividing the orders with large weighted similarity into batches can reduce the total cost of order picking.When the total number of orders to be batched is 4000,the similarity weight is0.5 to 0.5,and the maximum capacity of items contained in the batch is 200,the orders are divided into 43 batches,which effectively reduces the picking cost compared to random batching optimization.In conclusion,the combination of goods allocation and order batching strategies can effectively improve the completion efficiency of picking operations and significantly reduce picking costs.In the case of comparing random goods allocation and random order batching,the results show that the goods allocation strategy based on commodity association relationship and the order batching strategy based on weighted similarity are the optimal decision combination,which verifies that in the actual picking process,the combination of optimized goods storage and order The scientific nature of the combination of optimized goods storage and order batching in the actual picking process is verified.There are 20 figures,16 tables and 66 references in this paper.