Artificial intelligence and machine learning models for predicting and evaluating the influence of shelf-life environments and packaging materials on garlic (Allium Sativum L) physicochemical and phytochemical compositions.

Abstract

The nutritional content and quality of garlic, a crop widely consumed, must be preserved after harvesting by overcoming several challenges. The necessity of this study arises from the growing demand for effective postharvest management solutions that can extend shelf life, maintain the nutritional integrity of garlic and enhance consumer satisfaction. This study explores the application of machine learning (ML) and artificial intelligence (AI) to predict the effects of storage environments and packaging materials on garlic's physicochemical properties, enzymatic activities, phytochemical content, and antioxidant characteristics. The results show that temperature significantly influenced most parameters, except for anthocyanin, and that packaging impacted all variables. Storing garlic at 4 °C was found to preserve its quality better than at 25 °C, offering insights into optimizing storage conditions and packaging for superior product quality. This research provides valuable guidance on controlling factors that affect garlic's postharvest performance, aiming to improve preservation and reduce food waste.