Optimization And Simulation Of Dynamic Allocation Of Automated Warehouse

As a new type of storage method, automatic storage and retrieval system has been more and more widely used in enterprises. In the existing facilities conditions, storage system capacity often becomes the bottleneck of production capacity of enterprises. How to improve the system capacity of the automated warehouse has became the object of attention of various scholars and enterprises. Through the relevant research of domestic and foreign scholars, this paper finds that storage location assignment optimization mainly based on cargo turnover rate, goods correlation and stacker work path. The warehouse management research, which according to users’dynamic needs of products, is less. Location assignment policy, the enterprise generally uses, is the nearest or random storage. This can easily cause longer running time of the stacker, low efficiency of warehouse storage operations. Frequency of In-Out warehouse in different periods is changing. Storage partitioning can make adjustments to achieve dynamic location allocation. At the same time, storage partitioning can combine the goods storage frequency changes. Dynamic allocation optimization is more in line with the actual situation of enterprises to meet the automated warehouse, storage requirements. In order to optimize the storage location assignment, this paper according to the different periods of the goods out of storage frequency changes, focus on location dynamic allocation. This can reduce the time of the crane and improve the ability of the storage system and the efficiency of storage.Taking a tobacco distribution center as the study object, the existing storage problems in the distribution center are analysised in this paper. And on the basis of the turnover rates of goods and goods correlation, this paper takes ABC classification method to achieve product classification. The classification is on the basis of the goods out of storage frequency. Through the above steps, this paper builds basic expert knowledge base. And through the knowledge base constantly updated, the division of goods dynamic classification and location dynamic are realized.This paper constructs an improved location allocation model in order to find the shortest time of the crane and balance the work. The self-learning particle swarm optimization algorithm is used to achieve the best results and compare the results with the knowledge base and select the best scheme. After that, the location dynamic allocation is realized.This paper uses the Flexsim simulation software to establish 1:1 entity model of tobacco distribution center. The obtained results are applied to the optimization of storage location assignment model. And this paper analyzes the stochastic storage strategy, nearby storage method and partition storage strategy of system capacity in different storage capacity.The results of the study show that:(1) dynamic classification, based on self learning and dynamic storage location assignment optimization research, improved the automation warehouse system handling ability. (2) The dynamic partition storage strategy and the random storage strategy of the carrying capacity are not affected by the library capacity. Nearby storage strategy of automated warehouse system carrying capacity decreases gradually with the increase in the storage capacity, when the storage capacity is higher than 75%, transportation system capacity rate of decline began to decrease. When the storage capacity is more than 85%, the handling ability of the system tends to a constant. (3) Compared with the random storage strategy, the automated warehouse system carrying capacity of the dynamic partitioning storage strategy is improved by 8.94%. (4) When the storage capacity is less than 70%, the automated warehouse system carrying capacity of the nearest storage strategy is higher than the dynamic partition storage strategy. When the storage capacity is higher than 70%, the system capability of the nearest storage strategy is lower than that of the dynamic partitioning storage strategy. When the storage capacity is higher than 85%, the system capacity of the dynamic partition storage strategy is 6.12% higher than that of the nearest storage strategy and the system capacity of the nearest storage strategy is about 2% higher than that of the random storage strategy.The simulation by 1:1 entity model of the actual business data and distribution centers verifies that automated warehouse cargo dynamic allocation optimization is effective and feasible, which provides guidance and reference for the enterprise automated warehouse storage allocation scheduling.