Warehouse Prediction and Planning Based on K-MEANS Algorithm and GA-LSTM

Abstract

The purpose of this paper is to study warehouse prediction and planning based on K-means algorithm and GA-LSTM through data analysis and deep learning methods. Firstly, this paper utilizes the inventory and daily sales data and uses K-means clustering analysis to classify the data and verifies the validity of the clustering results by evaluating the indicators. Then, the daily sales volume is predicted based on the LSTM model, and the GA-LSTM is further used for multiple fitting, which yields more stable and accurate prediction results. The dual-objective function model for minimizing the daily rental cost and maximizing the upper capacity limit is established by introducing piecewise constraints and correlation constraints through optimization algorithms (SSA, AROA). The results show that the daily rental cost of warehouse capacity and the capacity cap fluctuate within a reasonable interval in the selected optimization solution, respectively, which verifies the effectiveness and superiority of the proposed method in warehouse planning. The research in this paper provides a new idea of warehousing prediction and planning for e-commerce enterprises, which can effectively improve the accuracy and efficiency of warehousing management and provides a theoretical basis and practical guidance for subsequent inventory optimization and supply chain management.