Product Demand Forecasting System for SMEs Using Extreme Gradient Boosting
DOI:
https://doi.org/10.70112/ajes-2025.14.2.4271Keywords:
Demand Forecasting, Extreme Gradient Boosting, Inventory Management, Machine Learning, Product PredictionAbstract
The success of Small and Medium-sized Enterprises (SMEs) is mainly dependent on inventory management; yet, many businesses struggle with the use of traditional methods of demand forecasting, which leads to overstocking or understocking that negatively influences business operations. This study aims to integrate machine learning techniques into the prediction of product demand and efficient inventory management. The dataset used was sourced from Kaggle, and feature engineering and categorical encoding were applied to prepare the data for analysis. A classification-based predictive model was implemented using Random Forest and Gradient Boosting machine learning algorithms to categorize demand into high, medium, and low levels. Metrics including accuracy, precision, recall, and F1-score were used to evaluate the system, and the results showed that Gradient Boosting performed better, with an accuracy rate of 87%, while Random Forest achieved an accuracy rate of 85%. The study concludes that machine learning techniques, particularly the Gradient Boosting classifier, can effectively forecast product demand for SMEs. The system has the potential to help minimize inventory risk while increasing operational efficiency. It is recommended that more advanced machine learning models be used to further improve demand forecasting.
References
[1] V. Dahiwale and P. B. Sangode, “A comparative study of the inventory management tools of textile manufacturing firms,” IMPACT: Int. J. Res. Humanit., Arts Lit., vol. 7, no. 4, pp. 335–344, Apr. 2019.
[2] R. Zimmermann and P. Brandtner, “From data to decisions: Optimizing supply chain management with machine learning-infused dashboards,” Procedia Comput. Sci., vol. 237, pp. 955–964, 2024, doi: 10.1016/j.procs.2024.06.096.
[3] M. C. Piñeros-Fernández, “Artificial intelligence applications in the diagnosis of neuromuscular diseases: A narrative review,” Cureus, vol. 15, no. 11, p. e48458, Nov. 2023, doi: 10.7759/cureus.48458.
[4] E. Schonfeld, N. Mordekai, A. Berg, et al., “Machine learning in neurosurgery: Toward complex inputs, actionable predictions, and generalizable translations,” Cureus, vol. 16, no. 1, p. e51963, Jan. 2024, doi: 10.7759/cureus.51963.
[5] M. Cerdas, S. Pandeti, L. Reddy, et al., “The role of artificial intelligence and machine learning in cardiovascular imaging and diagnosis: Current insights and future directions,” Cureus, vol. 16, no. 10, p. e72311, Oct. 2024, doi: 10.7759/cureus.72311.
[6] N. Nivedhaa, “A comprehensive review of AI’s dependence on data,” Int. J. Artif. Intell. Data Sci., vol. 1, no. 1, pp. 1–11, Mar. 2024, doi: 10.13140/RG.2.2.27033.63840.
[7] K. Srivastava, K. Choubey, and J. Kumar, “Implementation of inventory management system,” in Proc. Int. Conf. Innovative Computing and Communication (ICICC), 2020. [Online]. Available: https://ssrn.com/abstract=3563375
[8] R. Burtea and C. Tsay, “Constrained continuous-action reinforcement learning for supply chain inventory management,” Comput. Chem. Eng., vol. 181, p. 108518, 2024, doi: 10.1016/j.compchemeng.2023.108518.
[9] O. R. Amosu, P. Kumar, Y. M. Ogunsuji, S. Oni, and O. Faworaja, “AI-driven demand forecasting: Enhancing inventory management and customer satisfaction,” World J. Adv. Res. Rev., vol. 23, no. 2, pp. 708–719, 2024, doi: 10.30574/wjarr.2024.23.2.2394.
[10] P. Mishra et al., “Modeling and forecasting rainfall patterns in India: A time series analysis with XGBoost algorithm,” Environ. Earth Sci., vol. 83, no. 6, p. 163, 2024.
[11] O. F. Ajayi, A. A. Udosen, W. Ajayi, B. O. Ohwo, and A. I. Amusa, “Voting ensemble learning model (VELM) for harmful gas detection in environmental applications,” Asian J. Electr. Sci., vol. 13, no. 2, pp. 45–50, 2024, doi: 10.70112/ajes-2024.13.2.4252.
[12] D. Salinas, V. Flunkert, J. Gasthaus, and T. Januschowski, “DeepAR: Probabilistic forecasting with autoregressive recurrent networks,” Int. J. Forecast., vol. 36, no. 3, pp. 1181–1191, 2020, doi: 10.1016/j.ijforecast.2019.07.001.
[13] T. Adigun, A. Olosunde, and I. Eweoya, “Deep learning approaches in medical image segmentation: Implications for brain tumor detection and analysis,” Asian J. Electr. Sci., vol. 14, no. 1, pp. 1–6, 2025, doi: 10.70112/ajes-2025.14.1.4254.
[14] K. B. Praveen, “Inventory management using machine learning,” Int. J. Eng. Res. (IJERT), vol. 9, no. 6, 2020, doi: 10.17577/IJERTV9IS060661.
[15] D. Swami, A. D. Shah, and S. K. B. Ray, “Predicting future sales of retail products using machine learning,” arXiv preprint, arXiv:2008.07779, 2020. [Online]. Available: http://arxiv.org/abs/2008.07779.
[16] S. Ni, Y. Peng, K. Peng, and Z. Liu, “Supply chain demand forecast based on SSA-XGBoost model,” J. Comput. Commun., vol. 10, no. 12, pp. 71–83, 2022, doi: 10.4236/jcc.2022.1012006.
[17] L. Zeng, Y. Gao, and H. Hu, “Research on retail pricing strategy of supermarket fresh products based on XGBoost model,” Acad. J. Comput. Inf. Sci., vol. 7, no. 10, 2024, doi: 10.25236/AJCIS.2024.071012.
[18] E. E. Onuiri, T. F. Chikezie, I. E. Z. Obata, and R. C. Amanze, “High-accuracy forex trading prediction model using machine learning algorithms,” Asian J. Electr. Sci., vol. 13, no. 1, pp. 26–34, 2024, doi: 10.70112/ajes-2024.13.1.4235.
[19] R. E. Donatus, I. H. Donatus, and U. O. Chiedu, “Exploring the impact of convolutional neural networks on facial emotion detection and recognition,” Asian J. Electr. Sci., vol. 13, no. 1, pp. 35–45, 2024, doi: 10.70112/ajes-2024.13.1.4241.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Centre for Research and Innovation
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
