Warehouse intelligent stock-up method and system based on reinforcement learning model

• Main Authors: SHAN XINNING, CHENG BOQUAN
• Format: Patent
• Language: Chinese; English
• Published: 04.06.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses an intelligent warehouse stocking method based on a reinforcement learning model, and the method comprises the following steps: S1, obtaining current inventory information, front-end demand information and supply end information, and binding the obtained current inventory information, front-end demand information and supply end information into current state information; s2, inputting the current state information obtained in the S1 into a pre-constructed and trained reinforcement learning model; and S3, based on the input current state information, outputting the suggested stock quantity corresponding to each commodity in the warehouse through the reinforcement learning model. According to the invention, the optimal stock quantity is calculated according to the current inventory condition information, the front-end demand information and the supply end information, it is ensured that each order-placing stock is carried out under the front-end demand, warehouse turnover can be reduced