Krasnoyarsk, Russian Federation
employee from 01.01.1993 to 01.01.2019
Siberian Federal University (Chair of Digital Financial Technologies of Sberbank of Russia, Associate professor)
employee from 01.01.2007 until now
Krasnoyarsk, Krasnoyarsk, Russian Federation
UDK 339.133 Спрос
This article represents a practical study in the field of data analysis. In the rapidly changing retail market, accurate sales forecasting is crucial for maintaining business competitiveness. The study explores time series forecasting methods and models implemented using machine learning. We analyze sales data from the largest retail network in the Krasnoyarsk region, training the model on a two-year period. The review of scientific research allowed us to identify the most commonly used methods by analysts, assess the availability of data, and determine a preliminary list of factors influencing sales volume. Our study presents data preprocessing methods, stages of building and using forecast models. Various forecasting methods and machine learning models for building time series are described and compared, such as the Holt-Winters method, the additive linear model Prophet, and the seasonal autoregressive integrated moving average model. To evaluate forecast accuracy, we calculate the mean absolute error, mean squared error, and mean absolute percentage error. This research provides readers with an understanding of sales forecasting possibilities using machine learning models. The obtained results allowed us to justify the choice of the best time series forecast model, enabling more effective inventory management through accurate sales forecasting.
sales forecasting, time series models, trade, predictive analytics, forecast evaluation metrics, machine learning (ML)
1. Haque, M. S., Amin, M. Sh., Miah J. (2023). Retail demand forecasting: a comparative study for multivariate time series. [Electronic source] URL: https://arxiv.org/abs/2308.11939 (Date of access: 12.04.2024). DOI:https://doi.org/10.48550/arXiv.2308.11939.
2. Pivkin, K. S. (2016). Correlation analysis of factors influencing consumer demand of a retail store as a stage in the formation of a forecasting and inventory management model. Argancha University Bulletin. Economics and Law Series, Vol. 26, Is. 3, 40–50.
3. Pivkin, K. S. (2017). Forecasting key indicators of a retail network over time. Bulletin of the Perm University. Series: Economics, Vol. 12, Is. 4, 592–606. DOI:https://doi.org/10.17072/1994-9960-2017-4-592-608.
4. Korovin, A. M. (2022). Forecasting demand in the road construction equipment market using data mining tools. Bulletin of the South Ural State University. Series: Computer technology, control, radio electronics, Vol. 22, Is. 3, 117–131.
5. Zohdi, M., Rafiee, M., Kayvanfar, V., Salamiraad, A. (2022). Demand forecasting based machine learning algorithms on customer information: an applied approach. International Journal of Information Technology, 14(3). [Electronic source] URL: https://www.researchgate.net/publication/358567204_Demand_forecasting_based_machine_learning_algorithms_on_customer_information_an_applied_approach (Date of access: 12.05.2024). DOI:https://doi.org/10.1007/s41870-022-00875-3.
6. Vashishtha, R. K., Burman, V., Kumar, R., Sethuraman, S., Sekar, A. R., Ramanan, S. (2020). Product age based demand forecast model for fashion retail. [Electronic source] URL: https://arxiv.org/abs/2007.05278 (Date of access: 10.05.2024). DOI:https://doi.org/10.48550/arXiv.2007.05278.
7. Wiyanti, D. T., Kharisudin1, I., Setiawan, A. B., Nugroho, A. K. (2021). Machine-learning algorithm for demand forecasting problem. Journal of Physics: Conference Series. [Electronic source] URL: https://iopscience.iop.org/article/10.1088/1742-6596/1918/4/042012/pdf (Date of access: 11.05.2024). DOI:https://doi.org/10.1088/1742-6596/1918/4/042012.
8. Box, G. E. P., Jenkins, G. M., Reinsel, G. C., Ljung, G. M. (1976). Time series analysis: forecasting and control. [Electronic source] URL: http://repo.darmajaya.ac.id/4781/1/Time%20Series%20Analysis_%20Forecasting%20and%20Control%20%28%20PDFDrive%20%29.pdf (Date of access: 12.05.2024).
9. Ferreira, K. J., Lee, B. H. A., Simchi-Levi, D. (2015). Analytics for an online retailer: demand forecasting and price optimization. Manufacturing & Service Operations Management. [Electronic source] URL: http://hdl.handle.net/1721.1/101783 (Date of access: 12.05.2024). DOI:https://doi.org/10.1287/msom.2015.0561.
10. Teng, S. (2021). Route planning method for cross-border e-commerce logistics of agricultural products based on recurrent neural network. Soft Computing. [Electronic source] URL: https://www.researchgate.net/publication/351508064_Route_planning_method_for_ cross-border_e-commerce_logistics_of_agricultural_products_based_on_recurrent_neural _network (Date of access: 10.05.2024). DOIhttps://doi.org/10.1007/s00500-021-05861-8.
11. Starovoitov, V. V, Golub, Yu. I. (2021). Data normalization in machine learning. Computer science, 18(3), 83–96. DOI:https://doi.org/10.37661/1816-0301-2021-18-3-83-96.
12. Anafiev, A. S., Karyuk, A. S. (2022). Review of approaches to solving the problem of optimization of hyperparameters for machine learning algorithms. Tauride Bulletin of Computer Science and Mathematics, 2(55). 30–37.
13. Tofallis, C. (2015). A better measure of relative prediction accuracy for model selection and model estimation. Journal of the Operational Research Society, 66(8), 1352–1362. DOI:https://doi.org/10.1057/jors.2014.103.
14. Hodson, T. O. (2022). Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not Geoscientific Model Development, Vol. 15, Is. 14, 5481–5487. [Electronic source] URL: https://gmd.copernicus.org/articles/15/5481/2022/ (Date of access: 10.05.2024). DOI:https://doi.org/10.5194/gmd-15-5481-2022.
15. Bickel, P., Doksum, K. (2007). Mathematical Statistics: Basic Ideas and Selected Topics. Journal of the American Statistical Association, 56(4). [Electronic source] URL: https://www.researchgate.net/publication/236736826_Mathematical_Statistics_Basic_Ideas_and_Selected_Topics (Date of access: 10.05.2024). DOI:https://doi.org/10.2307/2286373.
16. Mirolyubova, A. A., Ermolaev, A. D. (2021). ARIMA – forecasting the creativity of a manufacturing enterprise. Modern high technology. Regional application, 2(66), 50–55.
17. Kolassa, S. (2011). Combining exponential smoothing forecasts using Akaike weights. International Journal of Forecasting, Vol. 27, Is. 2, 238–251. [Electronic resource] URL: https://www.sciencedirect.com/science/article/abs/pii/S0169207010001032 (Date of access: 13.05.2024).
18. VanBerlo, B, Ross, M. A. S., Hsia, D. (2021). Univariate Long-Term Municipal Water Demand Forecasting. [Electronic source] URL: https://arxiv.org/abs/2105.08486 (Date of access 13.05.2024). DOI:https://doi.org/10.48550/arXiv.2105.08486.
19. Hoffmann, D. (2023). Impact of HPO on AutoML Forecasting Ensembles. [Electronic source] URL: https://arxiv.org/abs/2311.04034 (Date of access: 13.05.2024). DOI:https://doi.org/10.48550/arXiv.2311.04034.
20. Taylor, S. J., Letham, B. (2018). Forecasting at scale. [Electronic source] URL: https://www.researchgate.net/publication/344989540_Forecasting_at_scale (Date of access: 13.05.2024). DOI:https://doi.org/10.7287/peerj.preprints.3190v2.
21. Kurilin, B. L., Kiselevskaya-Babinina, V. Ya., Karasev, N. A., Kiselevskaya-Babinina, I. V., Kislukhina, E. V., Vasiliev, V. A. (2019). Selection of a method for predicting the main statistical indicators of the work of the State Budgetary Institution “Research Institute of SP named after N. V. Sklifosovsky Department of Health of the City of Moscow”. Journal named after N. V. Sklifosovsky “Emergency medical care”, 8(3), 246–256.
22. Milushenko, O. A., Pupkova, A. G., Kovalev, A. I. (2020). Forecasting revenue from sales of a trading enterprise. Bulletin of the Siberian Institute of Business and Information Technology, 4(36), 58–64.
23. Vakhrushev, I. A. (2020). Forecasting the annual revenue of Russian companies of large and medium-sized businesses in the trade industry. Scientific journal of the ITMO Research Institute. Economics and Environmental Management Series, 3, 45–51.
24. Murray, P. W., Agard, B., Barajas, M. A. (2015). Forecasting Supply Chain Demand by Clustering Customers. IFAC-PapersOnLine, 48(3), 1834–1839. [Electronic resource] URL: https://www.sciencedirect.com/science/article/pii/S2405896315005923 (Date of access: 05.10.2024).
25. Das, A., Kong, W., Sen, R., Zhou, Y. (2023). A decoder-only foundation model for time-series forecasting. [Electronic resource] URL: https://arxiv.org/abs/2310.10688 (Date of access: 05.10.2024). DOI:https://doi.org/10.48550/arXiv.2310.10688.
