| With the upgrading of residents’ demand for medical consumption and the continuous improvement of medical security level,the market demand for medical supplies and services has gradually increasing.The pharmaceutical supply chain plays a central role in the business ecosystem and plays an important role in ensuring the quality and timely supply of pharmaceutical commodities.At the same time,with the development of market demand and supply chain research,the intelligent pharmaceutical supply chain has become a research field that has attracted much attention,especially for the important link of the pharmaceutical logistics center,and optimizing and improving the management efficiency of the logistics center has become the key areas in this field direction.Based on the real business data of a large-scale pharmaceutical logistics center,this article focuses on the problems of inventory backlog and bullwhip effect caused by inaccurate demand forecasting,and low picking efficiency caused by unreasonable allocation strategies in warehouses.Forecasting and slot allocation.The research content of this thesis mainly includes:(1)In view of the poor accuracy of demand forecasting in pharmaceutical logistics centers,a combination model considering causality and model prediction and punishment mechanism in this scenario is constructed,which is based on differential Auto-Regressive Integrated Moving Averages model and Temporal Convolutional Network to accurately predict the demand of different types of medicines.In order to evaluate the effectiveness of the demand forecasting method for pharmaceutical logistics centers,the accuracy of demand forecasting is associated with inventory status and service satisfaction,and the evaluation objectives are proposed.The experimental results show that compared with the Long Short-Term Memory network and the Gated Recurrent Unit network,the prediction accuracy of the combined model is higher,and the evaluation target is improved by 20%.(2)In order to solve the problems of chaotic storage location assignment and low picking efficiency in pharmaceutical logistics center,combined the storage location assignment and route planning,a assignment model is constructed that considers the correlation of goods,the center of gravity of the shelf,the turnover rate of the shelf and the load balance of the roadway.In order to solve the NP-hard problem of storage location assignment in the pharmaceutical logistics center,an improved GA algorithm is designed,which uses initialization to reduce the complexity of the problem in the storage location assignment,and at the same time,load balance is considered comprehensively in the route planning,and realize parameter self-adaptation,so as to accelerate the discovery of the optimal solution of the problem,and improve the efficiency and performance of the algorithm.Experimental results show that the proposed improved algorithm is superior to other algorithms in terms of optimization efficiency,and the performance index has increased by an average of 3.142%in the same time,and picking time reduced by 23.252%.(3)Including the above two research contents,a systematic pharmaceutical warehouse management verification system is designed to meet the various actual needs of a large pharmaceutical company in Nanjing in the storage process.The system investigates the functional modules designed around the above two research points in detail,and provides users with concise and easy-to-understand system services to improve the efficiency of warehouse management. |
