Research On Joint Optimization Of Order Allocation And Path Planning In AGV Picking System

As e-commerce develops,massive customer demands and highfrequency,multi-variety,small-batch orders have brought huge challenges to the picking efficiency of distribution centers.Under the trend of intelligent logistics,a large number of e-commerce companies use the Automated Guided Vehicle(AGV)picking system for order picking.While AGV picking system realizes the transformation of "parts-to-picker" and the improvement of order processing speed,it also faces some problems such as repeated storage of shelves and detours of robot’s walking route.In order to improve the efficiency of AGV picking system and realize the reasonable scheduling of storage resources,this thesis studies the joint optimization problem of order allocation and path planning based on the characteristics of AGV picking system.The research can provide decision support for intelligent warehouse management of e-commerce enterprises,which has important theoretical value and practical significance.The main work of this thesis includes:(1)Summarize existing research results on order picking systems,order allocation,and path planning.The equipment composition,structural layout and order picking process of AGV picking system are introduced and analyzed in detail,which lays foundation for the joint optimization research of AGV picking system.(2)A joint optimization method of order allocation and path planning for AGV picking system is proposed.The connotation of order allocation and path planning in the multi-order,multi-shelf,and multi-picking station scenarios is clarified,and the decision elements of order allocation and routing planning are pointed out,as well as their necessity and relevance.Then,aiming at minimizing the robot loaded distance,a mixed integer programming model for joint optimization of order allocation and path planning is established.At the same time,combined with the characteristics of problem and model,the A* algorithm and the Adaptive Large Neighborhood Search(ALNS)are used to solve the problem,and based on the original ALNS algorithm,a set of destroy and repair operator suitable for the order allocation problem is designed.(3)The method is validated using simulation examples of different scales.Using warehouse and order simulation examples of different scales,the solution efficiency of first-come-first-service(FCFS),CPLEX and ALNS algorithm in order allocation problems are compared and analyzed.In addition,the influence rules of batch capacity,order commodity types,shelf storage diversity and warehouse commodity richness on the joint optimization scheme are excavated.The results show that the joint optimization model and algorithm of order allocation and path planning proposed in this thesis can effectively shorten the loaded distance of robots,thereby improving the overall order picking efficiency of AGV picking system.What’s more,ALNS algorithm has good adaptability to the problem of order allocation,and can obtain solutions with similar quality to CPLEX in a relatively short time,with obvious time advantages.