The SLED (Shelf Life Expiration Date) Tracking System: Using Machine Learning Algorithms to Combat Food Waste and Food Borne Illnesses

TL;DR

This paper presents a machine learning-based system that uses sensory observations and an app to accurately predict food spoilage, aiming to reduce food waste and related environmental impacts.

Contribution

It introduces a novel approach combining sensory data, machine learning, and consumer tools to improve spoilage detection beyond traditional date labels.

Findings

Sensory observations are more accurate than best-by dates.

Actual spoilage time exceeds label estimates.

The app and test kit effectively engage consumers in waste reduction.

Abstract

The issue of food waste is a major problem contributing to the emission of greenhouse gases into the environment in addition to causing illness in humans. This research aimed to develop a correlation between the amount of time until a food spoiled and dates on food labels in conjunction with sensory observations. Sensory observations are more accurate as they are immediate observations that are specific to the food. This experiment observed bananas, bread, milk, eggs, and leafy greens over a period of time using characteristics specific to the food to quantify food spoilage. It was shown that the actual time until spoilage for all foods was longer than that of the best by date and that sensory observations proved to be a more accurate factor in determining spoilage. From this data, a machine learning algorithm was trained to predict if food was spoiled or not, in addition to the number of days until spoilage. This was presented to the consumer as an app, where the user can track foods and are reminded to check on them to prevent wastage. In addition, the experimental procedures were incorporated into a test kit for the consumer to take instructed observations to assess the spoilage of their food, which are then entered into the app to improve the algorithm. This paper discusses the individual effects of sensorial observations on each food and examines the shifting of consumer habits through an app and test kit to combat environmental consequences of food waste.