AI-Based Dynamic Pricing Engine in E-Commerce: Designing Intelligent, Real-Time, and Ethical Price Optimization Systems
Abstract: E-commerce platforms operate in markets where demand, competitor behaviour and inventory levels shift by the minute. Traditional cost-plus and rule-based pricing cannot react at this speed or scale across catalogues containing millions of SKUs. This paper examines the design and impact of an AI-based dynamic pricing engine that ingests demand signals, competitor prices, inventory position and customer behaviour to recommend or automatically apply optimal prices in near real time. The study surveys the analytical foundations of dynamic pricing, including price-elasticity estimation, demand forecasting and reinforcement-learning-based price optimization and proposes a modular system architecture spanning data ingestion, feature engineering, model training, decision policy and guardrail layers. A prototype workflow using gradient-boosted regression for elasticity estimation and a Q-learning-based agent for sequential price-setting is evaluated on simulated retail data. Findings suggest that AI-based pricing engines can improve gross margin and revenue relative to static pricing while remaining sensitive to fairness, transparency and regulatory constraints. The paper ends by talking about risks. These risks include things, like price discrimination and algorithmic collusion. The paper also gives some tips on how to use these tools in a way.
Keywords: dynamic pricing, e-commerce, machine learning, reinforcement learning, price elasticity, algorithmic pricing
How to Cite?: Somnath Kar, Renuka Mahto, Shilpa Oraon, "AI-Based Dynamic Pricing Engine in E-Commerce: Designing Intelligent, Real-Time, and Ethical Price Optimization Systems", Volume 15 Issue 8, August 2026, International Journal of Science and Research (IJSR), Pages: 1779-1784, https://www.ijsr.net/getabstract.php?paperid=SR26824211602, DOI: https://dx.doi.org/10.21275/SR26824211602