Enterprise Inventory Demand Forecasting Based on K-Means Clustering and a BP Neural Network

Chengyang Zhang

doi:10.62051/ijgem.v10n8.10

Authors

Chengyang Zhang School of Management, Southwest Petroleum University, Chengdu 610500, China

DOI:

https://doi.org/10.62051/ijgem.v10n8.10

Keywords:

K-means clustering, BP neural network, e-commerce retailing, Inventory-demand forecasting

Abstract

Against the backdrop of rapid e-commerce growth and increasingly personalized consumer demand, fast-moving consumer goods (FMCG) enterprises face the dual challenges of excess inventory and stockouts, which impose higher requirements on the accuracy of inventory-demand forecasting. Intensifying market competition, frequent promotional campaigns, and rapidly changing consumer preferences have made the forecasting task substantially more complex. This study uses daily sales data for the core products of Company Z, an FMCG enterprise, from 2023 to 2025. A multidimensional indicator system is constructed from product-demand characteristics, and K-means clustering is applied to classify the products. BP neural-network models are then used to forecast daily inventory demand over the following four weeks according to the demand patterns of each product cluster. The results show that, compared with the exponential-smoothing method currently used by the company, the proposed hybrid model achieves a closer fit to actual sales and reduces the root mean square error by an average of 42.3%, thereby demonstrating stronger accuracy and adaptability. The combination of K-means clustering and a BP neural network can effectively capture the nonlinear and volatile characteristics of FMCG demand and provide more accurate support for inventory planning, with considerable practical value and potential for wider application.

Downloads

Download data is not yet available.

References

[1] Li, J., & Zhang, Y. (2022). Demand forecasting for household chemical FMCG products based on exponential smoothing. Journal of Commercial Economics, (15), 132–134. [In Chinese]

[2] Zhang, M., & Li, Q. (2021). Application of the ARIMA model to demand forecasting for food products in e commerce. Statistics & Decision, (8), 178–181. [In Chinese]

[3] Wang, L. (2024). Inventory demand forecasting for e commerce enterprises based on a BP neural network. China Computer & Communication, 36(6), 20–24. [In Chinese]

[4] Chen, M., & Liu, J. (2023). A fresh FMCG demand forecasting model based on a Grey BP neural network. Logistics Technology, 42(3), 98–102. [In Chinese]

[5] MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability (pp. 281–297). University of California Press.

[6] Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back propagating errors. Nature, 323(6088), 533–536.

[7] Tan, T., & Wang, J. (2025). Inventory demand forecasting based on K means and a BP neural network: A case study of an in vehicle intelligent terminal enterprise. Logistics Engineering and Management, 47(3), 17–22. [In Chinese]

[8] Ou, H., Yao, Y., Feng, H., et al. (2023). A ship spare parts demand forecasting model based on a GA Grey BP neural network. China Shiprepair, 36(2), 41–46. [In Chinese]

[9] Yan, Q., Li, J., Zhang, Z., et al. (2014). A spare parts demand forecasting model based on grey relational analysis and a BP neural network. Logistics Technology, 33(2), 90–92, 104. [In Chinese]

[10] Cheng, F., & Sun, R. (2017). Inventory demand forecast based on grey correlation analysis and a time series neural network hybrid model. In 2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC FSKD) (pp. 2491–2496). IEEE.

[11] Ma, X., Duan, G., Wang, J., et al. (2021). Airline customer segmentation based on an improved silhouette coefficient method. Operations Research and Management Science, 30(1), 140–146. [In Chinese]

[12] Long, W., Zhang, X., & Zhang, L. (2020). A business process clustering method based on K means and the elbow rule. Journal of Jianghan University (Natural Science Edition), 48(1), 81–90. [In Chinese]

[13] Zhang, S., & Li, X. (2017). An estimation algorithm for hidden layer nodes in a BP network based on simulated annealing. Journal of Hefei University of Technology (Natural Science), 40(11), 1489–1491, 1506. [In Chinese]

[14] Xu, J., Zhou, J., Guo, Y., et al. (2021). Material management status and demand forecasting in a pump manufacturing enterprise: A case study of JS Pump Manufacturing. Logistics Engineering and Management, 43(2), 32–34. [In Chinese]

[15] Bai, C., & Song, L. (2018). A combined material demand forecasting model based on EMD PSO LSSVR. Statistics & Decision, 34(18), 185–188. [In Chinese]