Optimizing Micro Cold Storage for Detecting Stale Food and Fruits

Cold storage are a critical element of the food supply and sustainability supply chain for preservation and maintaining the quality as well as the nutritional value of stored items. This work investigates into the adoption of Artificial Intelligence (AI) techniques in cold storage with the aim of reducing food wastage thereby contributing to SDG-2. This work presents an algorithm based on the Inception v3 AI model that aids in assessing the condition or quality of fruits and vegetables that are stored in controlled environment conditions. From the proposed model, one can identify with an accuracy of 99%, the correct condition of the fruit/vegetable. Additionally, other structures have demonstrated high accuracy, such as the compact-CNN achieving 98.54% precision and the sequential model with 98% precision, both of which significantly contribute to the robustness of the results. These high accuracy rates praise the effectiveness of AI in categorizing farm products. Depending on the freshness level, the commodities are classified as fresh and rotten. Classification of such market offers the farmers very crucial information that allows them to make the right decisions, on the right time to sell their produce hence achieve stability of market price. This research not only helps to apply AI into the cold storage systems but also assist in helping the farmers through data-driven decision-making. The feature of recognizing the freshness of stored products proves the opportunities in applying the AI in managing inventory in cold storage and creating efficient and sustainable system. Finally, this work represents a great progress towards the development of smart systems improving the quality of products and a viable economics of storing them in the common granaries of farming industries.