Hierarchical Modeling of Product Demand Forecasting in the Presence of Promotion using prophet: A Multivariate Multiple Time Series Approach

DOI

10.2991/978-94-6463-884-4_67How to use a DOI?

Keywords

Promotional Modeling; Demand Forecasting; Prophet. Multivariate Analysis; Multiple Time series

Abstract

Traditional forecasting models often struggle with multivariate and multiple time series data, especially in the presence of promotional activities that introduce nonlinear patterns and seasonal variations. More recently, statistical models encompass promotional features e.g. price cuts, display time, free products, etc. This research emphasizes a hierarchical forecasting approach using Facebook Prophet to generate series of fine-grained forecast models for location-item combinations in for a multiple time series data of the aqua products. Forecast of the sales also includes the effect of promotion features “Bonus Sales Quantity”, fallen in the promotional category of free product. This study introduces a data-driven method to engineering challenges, using AI and gives a foundation for future research tools to work with advanced forecasting techniques. The accuracy matrices were compared for two approaches: one is Facebook Prophet with Bonus sales quantity as an additional regressor, and the other one is Facebook Prophet without any regressor. The Mean Absolute Error of the multivariate analysis with bonus sales was 5, significantly outperforming the univariate approach. The result indicates that promotional feature, Bonus sales, is essential to efficiently forecasting future value. The findings have important implications for optimizing inventory management and operation strategies. The results contribute to the research question by demonstrating the effectiveness of incorporating promotional features, such as bonus sales, into demand forecasting models. Parallel processing could improve efficiency by reducing training time for multiple time series.

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  - Ahmed Abul Fazal
AU  - Farjana Khanom Sadia
AU  - Md. Muhaiminul Islam
PY  - 2025
DA  - 2025/11/18
TI  - Hierarchical Modeling of Product Demand Forecasting in the Presence of Promotion using prophet: A Multivariate Multiple Time Series Approach
BT  - Proceedings of the 8th International Conference on Engineering Research, Innovation, and Education 2025 (ICERIE 2025)
PB  - Atlantis Press
SP  - 559
EP  - 565
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-884-4_67
DO  - 10.2991/978-94-6463-884-4_67
ID  - Fazal2025
ER  -