Food Waste Intelligence: A Predictive-Prescriptive Machine Learning Framework for Food Waste Reduction and Surplus Donation Analysis
Food waste is a major environmental, economic, and social challenge, particularly in food-service environments where inaccurate demand estimation can lead to excessive food preparation and unnecessary waste. This study proposes a predictiveprescriptive machine learning framework for food waste reduction and surplus donation analysis to support data-driven food management decisions. The proposed framework uses historical food, demand, preparation, consumption, and waste data to develop machine learning models for predicting expected food waste. It further incorporates what-if scenario analysis to evaluate different food preparation levels and identify their potential impact on waste and demand coverage. Based on the scenario outcomes, the framework provides a suitable waste-reduction scenario recommendation. It also estimates potential surplus food and performs donation potential analysis to identify food that may be considered for redistribution, subject to appropriate food safety and quality conditions. A web-based application is proposed to present waste predictions, scenario comparisons, recommendations, and surplus insights. The framework integrates prediction, scenario evaluation, recommendation, surplus estimation, and donation analysis into a unified decision-support system aimed at supporting responsible food management and contributing to Sustainable Development Goal 12.
Authors
- Nakka Shiva
- Nuligonda Lakshmi Pranay
- Thirunagari Naga Praveena
- Jakka Manideep
- Katikam Nikhitha
Publication Details
- Journal
- International Journal for Research in Applied Science and Engineering Technology
- Published
- 2026-11-16
- DOI
- https://doi.org/10.22214/ijraset.2026.85030
- Primary Topic
- Food Waste Reduction and Sustainability
- Type
- article
- Field-Weighted Citation Impact
- 0.00