Project Abstract

Food waste is a critical global issue, with approximately one-third of all food produced for human consumption wasted annually. This leads to significant economic, environmental, and health consequences. Predicting the shelf life of fresh foods is vital to reduce waste, enhance food safety, and achieve economic benefits. Traditional methods for shelf-life prediction have limitations such as costly and time-consuming laboratory testing and may not consider complex variables impacting food spoilage. Machine learning models offer promising avenues for accurate shelf-life prediction, but their application in this context requires further exploration. This project proposes to apply machine learning techniques, specifically Linear Regression and Neural Networks, to predict the shelf life of fresh foods. The methodology involves gathering a diverse dataset, preprocessing the data, training the models using cross-validation, developing a prediction model, and validating the models performance. The objective is to identify the most effective model for accurately predicting shelf life and contribute to sustainable practices in the food supply chain.

Project Team