redictiWaste: An ML-Powered Framework for Sustainable Food Inventory Optimization in Restaurants
FMDB Transactions on Sustainable Intelligent Networks · 2024
DOI: 10.69888/ftsin.2024.000289 ↗Kurz gefasst
Fachartikel in der Zeitschrift FMDB Transactions on Sustainable Intelligent Networks, 2024. Verfasst von H. Rakesh und 5 weiteren Autor:innen.
Abstract (Originalsprache)
Food wastage is a major concern in the food industry, affecting economic and environmental sustainability. This project introduces a predictive analysis model to reduce food waste in restaurant inventory and menu planning. The system uses machine learning techniques to forecast food wastage based on factors like food type, guest count, quantity, event type, storage conditions, purchase history, preparation methods, seasonality, location, and pricing. A dataset containing food wastage patterns is preprocessed to manage missing values and categorical variables. Label encoding is applied to categorical features, while numerical features are standardized. The data is split into training and testing sets, and a Random Forest Regressor is trained for accurate predictions. The model achieved 92.79% accuracy, with a Mean Absolute Error (MAE) of 1.63 and an R² score of 0.9279, proving its reliability. A Tkinter-based GUI lets restaurant managers input parameters and receive real-time wastage predictions. The system categorizes wastage as minimal, moderate, or high and suggests inventory control measures. Future improvements include integrating Long Short-Term Memory(LSTM) for time-series forecasting, IoT sensors for real-time storage monitoring, and blockchain for supply chain transparency. By minimizing food wastage at the inventory stage, this project helps businesses optimize resources while promoting sustainability.
Zitieren
APA
Rakesh, H., Harshad, R., Kumar, K. S., Regin, R., Anithalakshmi, A., & Padmapriya, G. (2024). redictiWaste: An ML-Powered Framework for Sustainable Food Inventory Optimization in Restaurants. FMDB Transactions on Sustainable Intelligent Networks. https://doi.org/10.69888/ftsin.2024.000289
Vancouver
Rakesh H, Harshad R, Kumar KS, Regin R, Anithalakshmi A, Padmapriya G. redictiWaste: An ML-Powered Framework for Sustainable Food Inventory Optimization in Restaurants. FMDB Transactions on Sustainable Intelligent Networks. 2024. doi:10.69888/ftsin.2024.000289
BibTeX
@article{Rakesh2024,
author = {Rakesh, H. and Harshad, R. and Kumar, K. Sanjay and Regin, R. and Anithalakshmi, A. and Padmapriya, G.},
title = {redictiWaste: An ML-Powered Framework for Sustainable Food Inventory Optimization in Restaurants},
journal = {FMDB Transactions on Sustainable Intelligent Networks},
year = {2024},
doi = {10.69888/ftsin.2024.000289}
}
EUVIMED