Home / Current Issue / Paper 1715847
AI-Based Demand Forecasting and Inventory Optimization for Seasonal Retail Cracker Stores
Subject area: Science,Engineering and Technology · Area of research: AI-Based Demand Forecasting and Inventory
DOI: https://doi.org/10.64388/IREV9I10-1715847
Abstract
Seasonal retail businesses, particularly firecracker stores, operate within high-stakes, compressed selling windows characterized by extreme demand volatility and significant inventory risks. Traditional manual forecasting methods often fail to capture non-linear surges, leading to prohibitive overstocking costs or lost revenue due to stockouts during peak festival periods. This paper presents an AI-driven framework for demand forecasting and inventory optimization using the XGBoost (Extreme Gradient Boosting) algorithm. By integrating historical sales data with high-impact seasonal features-including festival proximity, holiday indicators, and external environmental factors the proposed model identifies complex patterns that traditional statistical methods overlook. The system further employs dynamic safety stock calculations to optimize inventory levels and maximize serviceability. Experimental results demonstrate that the XGBoost-based approach significantly outperforms traditional moving-average and linear regression models, achieving a reduction in Mean Absolute Percentage Error (MAPE). The findings suggest that AI-based optimization not only enhances profitability but also improves operational safety by minimizing the risks associated with the prolonged storage of hazardous seasonal goods.
References
[1] J. Brownlee, Machine Learning Mastery with Python, Machine Learning Mastery, 2016.
[2] G. E. P. Box, G. M. Jenkins, and G. C. Reinsel, Time Series Analysis: Forecasting and Control, 5th ed., Wiley, 2015.
[3] T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794.
[4] S. Chopra and P. Meindl, Supply Chain Management: Strategy, Planning, and Operation, 6th ed., Pearson, 2016.
[5] H. Heizer, B. Render, and C. Munson, Operations Management: Sustainability and Supply Chain Management, 12th ed., Pearson, 2017.
[6] R.S.SuttonandA.G.Barto,ReinforcementLearning:An Introduction, MIT Press, 2018.
[7] A. Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, O’Reilly Media, 2019.
[8] K. P. Murphy, Machine Learning: A Probabilistic Perspective, MIT Press, 2012.
[9] D. Silver et al., “Mastering the Game of Go with Deep Neural Networks and Tree Search,” Nature, vol. 529, no. 7587, pp. 484–489, 2016.
[10] M. Christopher, Logistics and Supply Chain Management, 5th ed., Pearson, 2016.
[11] R. Hyndman and G. Athanasopoulos, Forecasting: Principles and Practice, OTexts, 2018.
[12] S. Makridakis, E. Spiliotis, and V. Assimakopoulos, “Statistical and Machine Learning Forecasting Methods: Concerns and Ways Forward,” PLOS ONE, vol. 13, no. 3, 2018.
[13] J. Han, M. Kamber, and J. Pei, Data Mining: Concepts and Techniques, 3rd ed., Morgan Kaufmann, 2012.
[14] D. J. Power, “Decision Support Systems: Concepts and Resources for Managers,” Greenwood Publishing, 2002.
[15] F. Chollet, Deep Learning with Python, Manning Publications, 2017.
How to cite this paper
@article{1715847,
author = {Asan Nainar M, Rajavel J},
title = {AI-Based Demand Forecasting and Inventory Optimization for Seasonal Retail Cracker Stores},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {10},
pages = {129-134},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1715847.pdf},
abstract = {Seasonal retail businesses, particularly firecracker stores, operate within high-stakes, compressed selling windows characterized by extreme demand volatility and significant inventory risks. Traditional manual forecasting methods often fail to capture non-linear surges, leading to prohibitive overstocking costs or lost revenue due to stockouts during peak festival periods. This paper presents an AI-driven framework for demand forecasting and inventory optimization using the XGBoost (Extreme Gradient Boosting) algorithm. By integrating historical sales data with high-impact seasonal features-including festival proximity, holiday indicators, and external environmental factors the proposed model identifies complex patterns that traditional statistical methods overlook. The system further employs dynamic safety stock calculations to optimize inventory levels and maximize serviceability. Experimental results demonstrate that the XGBoost-based approach significantly outperforms traditional moving-average and linear regression models, achieving a reduction in Mean Absolute Percentage Error (MAPE). The findings suggest that AI-based optimization not only enhances profitability but also improves operational safety by minimizing the risks associated with the prolonged storage of hazardous seasonal goods.},
month = {April},
doi = {https://doi.org/10.64388/IREV9I10-1715847}
}