Applying the Long Short-Term Memory Technique to Sales Forecasting for an Auto Parts Vendor: Applying the Long Short-Term Memory Technique to Sales Forecasting for an Auto Parts Vendor

Napat Srinual; Sittichok Sinrat; Chanisara  Thitkrathok

Authors

Napat Srinual Department of Logistics and Transportation Management, Panyapiwat Institute of Management
Sittichok Sinrat Department of Logistics and Supply Chain Management, Huachiew Chalermprakiet University
Chanisara Thitkrathok Department of Logistics and Supply Chain Management, Huachiew Chalermprakiet University

Keywords:

Auto parts, Sales forecasting, Forecast accuracy, Long Short-Term Memory

Abstract

In the face of increasingly fluctuating car demand, sales orders for automotive parts from various vendors have highlighted the growing importance of sales forecasting and predictive analytics to prevent overstocking and stock outs. However, most of the existing research continues to rely predominantly on traditional time-series forecasting methods that do not fully capture dynamic demand. To address these issues, this study applies a deep learning time-series forecasting technique to reduce the discrepancy and find a suitable method for forecasting the orders. The objective of this study is to propose an algorithm for forecasting sales based on an LSTM network, using historical sales data from an auto parts vendor. This study was conducted in an automotive vendor located in Samut Prakan, utilising a dataset covering 130 weeks from 2023-2025 to. A total of 70 percent of the sales data were employed for training, whereas 30 percent were allocated to the testing sets. The selected item stems from high-value auto parts consisting of 77443582-441V, 77443582-521V, 55183973-060V, and 55183973-440V. The four auto parts were selected based on their high sales volume and strategic importance to the vendor’s inventory management. These items contribute to over 40% of total sales revenue and are frequently subject to stockouts, making accurate forecasting critical. Compared with actual sales orders, the results show that the MAPE forecast error from the LSTM model is lower than the 6.14% error, and it achieves an R-squared of 90.228%, outperforming the existing forecasting technique used by the purchasing department. These findings suggest that the proposed forecasting model can effectively improve the sales order accuracy of automotive part vendors.

References

W. Liao, G. Y., Y. Yanchao, Y. Weijun, Y. Yaheng, and Z. Dongzhou, “Auto Parts Sales Prediction based on Machine Learning for Small Data and a Long Replacement Cycle, in ACS/IEEE Inter- national Conference on Computer Systems and Applications, November. 02-05, 2020, pp. 1-5.

H. Abbasimehr, M. Shabani, and M. Yousefi, “An optimized model using LSTM network for demand forecasting,” Computers and Industrial Engineering, vol. 143, pp. 1-13, May, 2020.

J. Zheng, C. Xu, Z. Zhang, and X. Li, “Electric load forecasting in smart grids using Long-Short-Term-Memory based Recurrent Neural Network,” in 51st Annual Conference on Information Sciences and Systems (CISS), Mar 22-24, 2017, pp. 1–6.

C. Ghassen, H. Bakir, and H. Zaher, “E-commerce Time Series Forecasting using LSTM Neural Network and Support Vector Regression,” in Proceedings of the International Conference on Big Data and Internet of Things, December. 20, 2017, pp. 80-84.

B. Lim and S. Zohren, “Time-series forecasting with deep learning: a survey,” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, vol. 379, no. 2194, pp. 1-14, Feb, 2021.

M. Khashei and M. Bijari, “A novel hybridization of artificial neural networks and ARIMA models for time series forecasting,” Applied Soft Computing, vol. 11, no. 2, pp. 2664-2675, Mar, 2011.

K. Satashwa1, M. Nagre, and A. Bhange, “Artificial Neural network for effective demand forecasting of Product” in International Journal of Advances in Engineering and Management (IJAEM), vol. 4, no. 10, pp. 982-986, Oct, 2022.

A. R. S. Parmezan, V. M. Souza, and G. E. Batista, “Evaluation of statistical and machine learning models for time series prediction: Identifying the state-of-the-art and the best conditions for the use of each model,” Information Sciences, vol. 484, pp. 302-337, Jan, 2019.

H. A. Hurtado-Mora, A. H. García-Ruiz, R. Pichardo-Ramírez, L. J. González-del-Ángel, and L. A. Herrera-Barajas, “Sales Forecasting with LSTM, Custom Loss Function, and Hyperparameter Optimization: A Case Study,” Applied Sciences, vol. 14, no. 21, p. 9957, Oct, 2024.

A. El Filali, E. H. Lahmer, S. El Filali, M. Kasbouya, M. A. Ajouary, and S. Akantous, “Machine Learning Applications in Supply Chain Management: A Deep Learning Model Using an Optimized LSTM Network for Demand Forecasting,” International Journal of Intelligent Engineering and Systems, vol. 15, no. 2, pp. 464-478, Feb, 2022.

K. Greff, R. K. Srivastava, J. Koutník, B. R. Steunebrink, and J. Schmidhuber, “LSTM: A Search Space Odyssey,” IEEE Transactions on Neural Networks and Learning Systems, vol. 28, no. 10, pp. 2222–2232, Mar, 2015.

S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735-1780, Nov, 1997.

N. Na, “Research on Supply Chain Demand Prediction Model Based on LSTM,” Procedia Computer Science, vol. 243, no. C, pp. 313-322, Oct, 2024.

K. Chandriah and R. Naraganahalli, “RNN/LSTM with modified Adam optimizer in deep learning approach for automobile spare parts demand forecasting,” Multimedia Tools and Applications, vol. 80, no. 20, pp. 26145-26159, 2021.

A. Limbare and R. Agarwal, “Demand forecasting and budget planning for automotive supply chain,” EAI Endorsed Transactions on Internet of Things, vol. 10, no. 1, pp. 1-8, Nov. 2023.

S. Suddala, “Dynamic demand forecasting in supply chains using hybrid ARIMA-LSTM architectures,” International Journal of Advanced Research, vol. 12, no. 2, pp. 167-174, 2024.

L. Terrada, M. Khaili, and H. Ouajji, “Demand forecasting model using deep learning methods for supply chain management 4.0,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 5, pp. 517-523, 2022.

L. Cui, Y. Chen, J. Deng, and Z. Han, “A novel attLSTM framework combining the attention mechanism and bidirectional LSTM for demand forecasting,” Expert Systems with Applications, vol. 254, Art.no. 124409, 2024.

D. Pliszczuk, P. Lesiak, K. Zuk, and T. Cieplak, “Forecasting Sales in the Supply Chain Based on the LSTM Network: The Case of Furniture Industry,” European Research Studies Journal, vol. 14, Special Issue 2, pp. 627-636, Jun, 2021.

Applying the Long Short-Term Memory Technique to Sales Forecasting for an Auto Parts Vendor