6Reinforcement Learning for Demand Forecasting and Customized Services
Sini Raj Pulari1, T. S. Murugesh2*, Shriram K. Vasudevan3 and Akshay Bhuvaneswari Ramakrishnan4
1Bahrain Polytechnic, ISA Town, Bahrain
2Department of Electronics and Communication Engineering, Government College of Engineering Srirangam, Tiruchirappalli, Tamil Nadu, India (On Deputation from Annamalai University, Department of Electronics and Instrumentation Engineering, Faculty of Engineering & Technology, India)
3Lead Technical – Evangelist (Asia Pacific and Japan), Intel India Pvt. Ltd., Bengaluru, Karnataka, India
4Department of Computer Science and Engineering, Sastra Deemed to be University, SASTRA University Thanjavur Campus, Thanjavur, Tamil Nadu, India
Abstract
Reinforcement Learning (RL) is a strong way for machines to learn that has shown promise in areas like predicting demand and providing personalized services. This chapter investigates how strategies based on RL can improve the accuracy of demand forecasting and make it possible for businesses to provide individualized services to each of its clients. The principles of RL, its use in demand forecasting, and the implementation of individualized services are covered extensively as key components of the topic. A real-life case study of a large retail chain highlights the practical benefits of applying RL in optimizing inventory management and providing individualized product recommendations. Researchers at the University of California, Berkeley, ...