Predictive system for demand planning using machine learning in Bookstores - Bazar M & M of Metropolitan Lima (#113)

Authors

Corrales Jhong, Jose

Aldave Paredes, Miguel

Aliaga Cerna, Esther

Abstract

The present research work aims to develop a predictive system for demand planning using machine learning techniques in M ​​& M Bookstores - Bazar in Metropolitan Lima. Currently, the company faces difficulties in accurately predicting the demand for its products, resulting in inefficient inventory management, unnecessary stock accumulation, and loss of sales due to lack of products. Various predictive models were evaluated, with the random forest model having the best performance, with an accuracy of 92% in predicting demand. Thanks to the implementation of the system, the company was able to reduce excess stock by 15% and optimize the product replenishment process, improving operational efficiency. These results demonstrate that the use of machine learning can significantly contribute to decision making in inventory management, with the possibility of being replicated in other retail stores to obtain similar benefits.