Enhancing Demand Prediction Accuracy in E-commerce With Ensemble Machine Learning

DOI

10.2991/978-94-6463-866-0_44How to use a DOI?

Keywords

Demand Forecasting; Ensemble Learning; XGBoost; LightGBM; Random Forest; Stacking; Inventory Optimization; Sales Prediction

Abstract

Efficient inventory management is critical in e-commerce, where inaccurate demand forecasts can lead to costly stock outs or excess inventory. This study presents a scalable, machine learning-based framework to optimize warehouse operations. Historical sales data—including product/store identifiers, pricing, holidays, promotions, and weather—were used to train three tree-based models: XGBoost, Random Forest, and Light GBM. To enhance prediction performance, a stacked ensemble method was employed, integrating the outputs of the base models through Ridge Regression as the meta-learner. This technique harnesses the complementary strengths of each model to deliver more reliable and consistent forecasts. The resulting ensemble attained a Mean Absolute Error (MAE) of 7.177, surpassing the accuracy of any individual model. Using aligned features from future data, the model generates forecasts that inform critical inventory metrics such as average daily demand, safety stock, and reorder points. The proposed system not only enhances forecasting accuracy but also integrates ensemble learning with practical inventory planning metrics, enabling proactive and interpretable decision-making in real-world warehouse environments.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - S. M. Bhavishya
AU  - T. Anusha
PY  - 2025
DA  - 2025/10/31
TI  - Enhancing Demand Prediction Accuracy in E-commerce With Ensemble Machine Learning
BT  - Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)
PB  - Atlantis Press
SP  - 527
EP  - 538
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-866-0_44
DO  - 10.2991/978-94-6463-866-0_44
ID  - Bhavishya2025
ER  -