Smart shelves: transforming retail stocking with internet of things and machine learning

Abstract

Retail environments are increasingly leveraging technology to streamline operations and improve efficiency. This paper presents ’Smart Shelves’, an innovative application that utilizes the Internet of Things (IoT) and machine learning (ML) to facilitate shelf replenishment. The application is designed to assist retail stock clerks in managing inventory more efficiently, reducing the potential for human error, and enhancing overall retail performance. The system’s effectiveness is highlighted by its 99.35% accuracy in distinguishing between products, even when dealing with near-identical product features that require restocking. The system’s operational efficiency is demonstrated by its 8.66-second average response time to issue notifications. A significant contribution of the proposed work is the development of a synthetic dataset that closely mirrors real-world retail conditions as obtaining a real-world dataset presented significant challenges. The findings demonstrate the transformative potential of IoT and ML technologies in the retail industry, particularly in the realm of shelf replenishment. Finally, this paper opens the door for future explorations that integrate the proposed system with existing inventory management software.