A DATA-DRIVEN METHOD TO DYNAMIC PRICING: UNRAVELLING INVENTORY AND COMPETITOR CONTESTS WITH AI IN E-COMMERCE

Shikha Vashishtha; Mani Garg; Megha Vimal

doi:10.29121/shodhkosh.v5.i6.2024.3392

Authors

Dr. Shikha Vashishtha Assistant Professor, Management, Institute of Business Studies, Chaudhary Charan Singh University Campus, Meerut
Mani Garg Assistant Professor, Management, Institute of Business Studies, Chaudhary Charan Singh University Campus, Meerut
Dr. Megha Vimal Professor and Head, Department of Management Studies – DIMS, Dewan, VS Group of Institutions, Meerut

DOI:

https://doi.org/10.29121/shodhkosh.v5.i6.2024.3392

Keywords:

Dynamic Pricing, Bayesian Optimization, Rule-Based Systems, Demand Forecasting, E-commerce Pricing Strategies

Abstract [English]

Dynamic pricing is a critical e-commerce approach that allows firms to modify prices in real time depending on demand, competition activity, and inventory levels. However, successfully adopting such tactics necessitates overcoming obstacles such as accurate demand forecasts, rival pricing monitoring, and inventory turnover optimization. This study presents a data-driven framework that integrates artificial intelligence (AI) techniques, Bayesian optimization, and rule-based systems to provide efficient, flexible pricing strategies. The system use Bayesian Optimization to dynamically balance goals like as revenue maximization, inventory management, and competitiveness, while also including rule-based procedures to assure compliance with business limitations and regulatory norms. Long Short- Term Memory (LSTM) networks are used to estimate demand by modeling temporal trends in sales data, while rival price data is watched and analyzed using web scraping and Natural Language Processing (NLP). Experimental validation of synthetic and real-world e-commerce data shows considerable gains, such as a 22% increase in revenue, a 30% decrease in inventory costs, and improved reaction to competition price. By integrating powerful optimization algorithms with realistic business principles, this framework offers a scalable, efficient, and transparent solution for dynamic pricing in competitive marketplaces. Future research will focus on tailored pricing and explainable AI (XAI) to improve consumer trust and transparency in decision-making.

References

Vives, A. and Jacob, M. (2021). Dynamic pricing at many Spanish vacation hotels. Tourism Economics, 27(2), 398–411. https://doi.org/10.1177/1354816619870652 DOI: https://doi.org/10.1177/1354816619870652

Mohammed, I., Guillet, B. D., Law, R., and Rahaman, W. A. (2021). Predicting the direction of dynamic pricing changes in the Hong Kong hotel business. Tourism Economics, 27(2), 346–364. https://doi.org/10.1177/1354816620903900. DOI: https://doi.org/10.1177/1354816620903900

Ulmer, M.W. (2020). Dynamic pricing and routing for same-day delivery. Transportation Science, 54(4), 1016–1033. https://doi.org/10.1287/trsc.2019.0958 DOI: https://doi.org/10.1287/trsc.2019.0958

Branda, F. Marozzo, & Talia, D. (2020). Ticket sales forecast and dynamic pricing tactics for public transportation. Big Data and Cognitive Computing, 4(4): 36. DOI: 10.3390/bdcc4040036 DOI: https://doi.org/10.3390/bdcc4040036

Najafi S, Duenyas I, Jasin S, & Uichanco J (2024). A cascade click approach is used to provide dynamic pricing for several products with limited stockpiles. Manufacturing and Service Operations Management, 26(2), 554–572. https://doi.org/10.1287/msom 2021.0504. DOI: https://doi.org/10.1287/msom.2021.0504

Hu J, Li L, Zhu X, Zhang H, Yang W (2023). Two-period omni-channel advertising and dynamic pricing approach that takes into account customer behavior. Operations Research and Management Science, 32(8), 114–121.

Sun S. and Han S. (2020). Multi-factor incentive pricing model and algorithm for military equipment purchasing using bi-level decision-making. Systems Engineering and Electronics, 42(6): 1338-1347. https://doi.org/10.3969/j.issn. 1001-506X.2020.0618

Zhao G., Ma C., Zhao Q., and Li J. (2023). Dynamic pricing of airline upgrading services based on revenue management. Journal of Transportation Systems Engineering and Information Technology, 23(3), 76–84. https://doi.org/10. 16097/j.cnki.1009-6744.2023.03.009.

Guo, W., Tian, J., and Li, M. (2023). Deep learning-based improved dynamic recommendation that takes into account price. Journal of Retailing and Consumer Services, 75, 103500. https://doi. org/10.1106/j.jretconser.2023.103500. DOI: https://doi.org/10.1016/j.jretconser.2023.103500

Pandey, V., Wang, E., and Boyles, S. D. (2020). A deep reinforcement learning technique for dynamic pricing of fast lanes with many entrance points. Transportation Research Part C: Emerging Technologies, 119 (102715). https://doi.org/10.1016/j.trc.2020.102715 DOI: https://doi.org/10.1016/j.trc.2020.102715