Abstract

Demand forecasting in grocery retail encounters considerable difficulties due to fluctuating consumer behavior, as well as external factors such as weather conditions and local events. This research presents an innovative framework that utilizes generative artificial intelligence (AI) to improve forecasting accuracy by incorporating various contextual elements. By employing Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), we integrate weather data, local events, and consumer behavior to better predict grocery sales. The proposed approach aims to optimize inventory management, minimizing stockouts and reducing waste. Experimental results indicate that this context-aware method significantly outperforms traditional forecasting models, providing a powerful solution for improving supply chain operations in grocery retail.