Reinforcement Learning in Business Decision Support Systems: Real-Time Optimization of Pricing and Inventory
- Version
- Download 5
- File Size 461.77 KB
- File Count 1
- Create Date 4 June 2025
- Last Updated 4 June 2025
Reinforcement Learning in Business Decision Support Systems: Real-Time Optimization of Pricing and Inventory
Authors:
Akshat Bajpai, Dr. Shilpa Pandey
1 Akshat Bajpai, Final Year Student, Amity Business School, Amity University Chhattisgarh, Raipur, India
2 Dr. Shilpa Pandey, , Assistant Professor, Amity Business School, Amity University Chhattisgarh, Raipur, India
Abstract - Amid a rapidly evolving market environment, enterprises must consistently recalibrate their pricing and inventory strategies to stay competitive and meet customer needs. Conventional Decision Support Systems (DSS) often depend on fixed models and rules, rendering them ineffective in dynamic scenarios. This paper introduces a smart DSS architecture that leverages Reinforcement Learning (RL), focusing on Q-Learning and Deep Q-Networks (DQN), to enable real-time optimization of pricing and inventory decisions. Through simulations in a retail setting characterized by fluctuating demand and uncertain supply chains, the RL agent is trained to boost overall profit while curbing stockouts and inventory costs. Our findings show substantial performance gains over traditional rule-based models, with adaptive strategies developing through learned behavior. The study underscores RL’s capacity to drive autonomous and data-centric improvements in business decision-making.
Key Words: Reinforcement Learning, DSS, Pricing Strategy, Inventory Optimization, Q-Learning, DQN, Real-Time Decisions
Download