Task Scheduling In Intelligent Warehouse System

With the rapid development of the electronic information,the e-commerce has ushered in rapid growth.The comprehensive popularity of e-commerce makes shopping convenient for people and also poses a huge challenge to traditional warehouse picking.In the e-commerce trading environment,users not only pursue product quality,but also put forward higher requirements for delivery efficiency.Therefore,it is important to accurately and quickly process massive orders and improve order sorting efficiency.At present,smart logistics are actively advocated at home and abroad,and large-scale smart warehousing,as its most important support,has attracted the attention of researchers.Among them,Amazon's "Kiva" intelligent warehousing system took the lead in using AGV mobile robots to replace pickers,freeing pickers from the time-consuming task of picking goods,and setting off a wave of reforms in the warehousing industry.This "goods-to-person" picking mode using AGV can greatly improve the efficiency of product picking and has been widely used.However,the intelligent warehousing system under this sorting mode still has some new problems,which cannot meet the picking needs of massive orders.The overall efficiency still needs to be improved.The scheduling of order tasks is one of the key issues.In the process of order task processing,picking time occupies more than 60% of the proportion,and in the process of picking goods,AGVs often have repeated driving paths and go wrong,which seriously affects the efficiency of picking.Therefore,the order picking process is optimized and researched.An effective task scheduling strategy is of great significance to the improvement of the efficiency of intelligent warehousing.The research work of this thesis starts from the following two aspects:Firstly,pre-processing the orders that arrive in the system,so as to reduce the influence of the order arrival sequence and batch plan in the warehousing system on the AGV driving path,and improve the sorting efficiency of the warehousing system.Starting from the current intelligent warehousing system structure and picking mode,this thesis analyzes the problems in common order batching methods by studying the order processing process,and designs a two-stage order batching plan based on actual warehousing,and creates an order similarity matrix through order information.An initial batching scheme based on order similarity is designed,and then the sorting station is considered as a whole to calculate the degree of similarity with the remaining orders,so as to realize the adaptive batching of the remaining orders in the order set.Finally,it is verified on the simulation platform that the algorithm can effectively reduce the number of AGV handling racks and improve storage efficiency.Secondly,select the appropriate execution AGV and task picking order for the completed batches of orders,that is,to allocate tasks reasonably.In order to realize the distribution of completed orders in batches and further reduce the total driving distance of the AGV,this thesis starts from the actual storage environment and summarizes the current order distribution problems that need to be solved according to the driving path of the AGV handling rack,and discretizes the particle swarm optimization algorithm to solve the task allocation problem in the system.Considering the limitation of premature of discrete particle swarm optimization algorithm,the differential evolution algorithm is combined to improve it,so as to improve the convergence speed and optimization ability of the algorithm.Finally,the simulation results show that the algorithm has a good optimization ability,and has a great gain to the system picking efficiency.