A Hybrid ARIMA-LSTM Approach with Dual Optimization Modes for E-Commerce Warehouse Inventory-Sales Prediction and Allocation_Vol. 3 No. 2 (JES 2025)_Journal of Engineering System (ISSN: 2959-0604)

Home > Journal of Engineering System (ISSN: 2959-0604) > Vol. 3 No. 2 (JES 2025) >

A Hybrid ARIMA-LSTM Approach with Dual Optimization Modes for E-Commerce Warehouse Inventory-Sales Prediction and Allocation

Download PDF

DOI: https://doi.org/10.62517/jes.202502202

Author(s)

Jianheng Lu, Shan He, Yurong Wu, Jiaxi Xing, Ning Wang, Yuan Long

Affiliation(s)

School of Artificial Intelligence, Guangzhou Huashang College, Guangzhou, Guangdong, China

Abstract

This paper tackles e-commerce warehousing challenges through a data-driven approach. We preprocess interrupted data using interpolation techniques, then predict inventory via Auto ARIMA and sales via LSTM. Two optimization models are developed: a 0-1 integer programming model for cost-effective "one-product-one-warehouse" allocation, and a PSO algorithm for multi-objective "one-product-multiple-warehouses" planning. The proposed methods outperform baselines in cost reduction and operational efficiency, offering practical solutions for warehouse management.

Keywords

Multi-Objective Optimization; Time Series Analysis; Auto-Regressive Integrated Moving Average Model; Long Short-Term Memory Network; Particle Swarm Optimization

References

[1]Statista Digital Market Outlook. Global E-commerce Market Size 2023. Statista GmbH. Retrieved from [2]Y. Zhang, Research on inventory optimization of L cross-border e-commerce company's overseas warehouses based on inventory classification and demand forecasting, Master's thesis, Zhejiang Univ., Hangzhou, China, 2024. [3]J. Jiang, J. Han, and Y. Zhang, Research on optimization of storage location assignment in e-commerce warehouses, Mechanical Design and Manufacturing, 2025, 1-6. [4]Li J, Liu R, Zhang X, et al. Constrained multi-objective evolutionary algorithm based on the correlation between objectives and constraints. Swarm and Evolutionary Computation, 2025, 94101903-101903. [5]Hyndman, R. J., & Athanasopoulos, G. Forecasting: Principles and Practice (3rd ed.). OTexts. 2021. [6]D. Salinas, V. Flunkert, J. Gasthaus, and T. Januschowski, DeepAR: Probabilistic forecasting with autoregressive recurrent networks, Int. J. Forecast. 2020, 36(3): 1181-1191. [7]S. Mayadunne, H. K. Rajagopalan, and E. Sharer, A multi-step mixed integer programming heuristic for warehouse layout optimization, Supply Chain Analytics, 2024, 8, p. 100088. [8]R. T. Marler and J. S. Arora, The weighted sum method for multi-objective optimization: New insights, Struct. Multidiscip. Optim, 2010, 41(6): 853-862. [9]Z. Tan, H. Li, and X. He, Optimizing parcel sorting process of vertical sorting system in e-commerce warehouse, Advanced Engineering Informatics, 2021, 48, 101279. [10]B. L. Shankar, S. Basavarajappa, J. C. H. Chen, and R. S. Kadadevaramath, Location and allocation decisions for multi-echelon supply chain network-A multi-objective evolutionary approach, Expert Syst. Appl, 2013, 40(2): 551-562.