Optimization Research Of E-commerce Order- Picking Based On Goods-to-man Mode

With the development of computer science and technology, as well as the attention of theE-commerce industry from our country, the proportion of the E-commerce enterprises in thenational economy occupies larger and larger. In the commodity trading of E-commerce, thecustomers not only pay attention to the quality of the goods, but also pursuit the arrival speed ofthe goods. To improve the efficiency of the order processing is very important to improvecustomer satisfaction. During the whole process of the order processing, order-picking occupiesmore than 50% of the time. So to optimize of order-picking process becomes a powerfulguarantee for improving the efficiency of the order processing.At present, the distribution center is developing in the direction of automation and unmanned. The method of order-picking is gradually changed from "man to goods" to "goods to man". And use AGV cars in the "goods-to-man" system in the distribution center has become a promising way of order-picking. This paper is in the background of using the AGV cars in "goods to man" picking system. There are some problems in the system, such as the long driving path of the AGVs, route duplication, the imbalance of the task allocation of AGVs and the long time waiting at the picking station. According to the problems and the order-picking process, the paper mainly does some optimization of the order-picking process from three stages: order batching, the path of order-picking and the queuing system of the AGVs at the picking station.(1) Analyze the current orders of distribution center, and compare it with the order-picking way of "man to goods" mode. And establish mathematical model with the goal of the least total number of AGVs carrying shelves. Then solve the problem using the saving algorithm and find a reasonable way of order batching. Finally compare the result with the situation before, proving the validity of the algorithm and the model.(2) For each batch of orders and according to the location of their shelves in the distribution center, do a plan of the driving path for the AGVs. The paper establishes a mathematical model with the objective function of the shortest driving path of the AGVs. And set some constraints to make the path of each AGV more balanced. Then use the genetic algorithm to solve the problem and find a optimized driving path for the AGVs. In the end, do a comparison of now and before to prove the superiority of the research. Finally compare the results by changing the parameter of the algorithm and provide a guidance of the setting of parameter in the algorithm.(3) Through the statistics of AGVs arrival time at the picking station and the time accepting service of the AGVs, the paper simulates the queuing process of the AGVs using Flexsim and find out the bottleneck problems of the system. Then changing the way of AGVs entering into the queue, it makes the way change from entering the fixed queue to choosing the queue dynamically, which increases the flexibility of the AGVs queuing process. At the same time, do some simulations by changing the number of the picking stations, find the suitable number of the picking stations, and ultimately improve the efficiency of the order-picking system.