Automated carbon footprint estimation of food products

Details

Supervisors

Barbette, Tom

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] en sciences informatiques, à finalité spécialisée

Abstract

This thesis presents a software solution designed to automatically estimate the carbon footprint of food products based on grocery receipts. The aim is to develop a reliable tool that could eventually be used to print carbon footprint information directly on receipts, helping consumers understand the environmental impact of their daily purchases. Existing tools often rely on generic or incomplete data, limiting their precision and relevance. This project addresses these limitations by developing a complete pipeline that extracts relevant information from receipts, matches products with environmental databases such as Agribalyse and OpenFoodFacts, and calculates greenhouse gas emissions for each stage of the product life cycle: agriculture, transport, and packaging. Natural language processing techniques and similarity algorithms are used to improve product matching. The methodology incorporates factors such as seasonality, country of origin, and transport mode to produce more accurate estimates. The system also falls back on financial emission factors when specific data is unavailable. This approach enables a more transparent and detailed estimation of the carbon footprint, supporting consumers in making more sustainable choices.