GAIA: Great-Distribution Artificial Intelligence-Based Algorithm for Advanced Large-Scale Commercial Store Management
Cettina Giaconia, Aziz Chamas · Applied Sciences · 2022
Today, the intelligent management of market stores in the large distribution field represents one of the most difficult tasks to address, considering the various problems to be managed. Specifically, from the classic issues of managing out-of-stock to the reconstruction of customer sentiment and the optimal management of shelves, scientific research has placed considerable effort on producing robust and efficient solutions to the aforementioned problems. In this context, modern deep learning techniques have allowed for the development of intelligent and adaptive systems capable of automating and significantly improving the management of a large-scale distribution market. Specifically, the authors have designed and implemented an innovative full pipeline that integrates modern deep learning technologies. More in detail, an innovative pipeline embedding a visual AI-based engine for customer sentiment assessment merged with a deep framework for stock management and market store cashflow monitoring is proposed. The innovative proposed system has been tested and validated in a large-scale distribution supermarket, confirming the effectiveness of the proposed solution. Specifically, in the performed testing sessions, the designed pipeline was able to show ad hoc visual customer sentiment assessment with an accuracy of 95% as well as intelligent stock monitoring with an accuracy of 93% in cross validation.