Predictive Food Demand Forecasting and Waste Reduction Using Machine Learning in Modern Catering Systems

This study has demonstrated that a multi-algorithm machine learning forecasting framework, integrating LSTM networks, Random Forest, XGBoost, and an ensemble model with IoT-derived real-time features, offers materially superior demand forecasting performance relative to traditional ARIMA approaches in large-scale catering management contexts. The ensemble model achieved an MAPE of 5.4% and an R² of 0.94 on out-of-sample test data, and its deployment in a six-month pilot implementation was associated with a 36% reduction in food waste, a 26.5% decrease in procurement costs, a 66.3% reduction in stockout incidents, and an 18.4% improvement in customer satisfaction scores. These outcomes are not merely technical achievements. They represent a meaningful advancement in the operational capability of catering organisations to manage food resources sustainably, efficiently, and responsively. The study contributes to the management literature by providing an empirically grounded, managerially framed account of how AI-driven forecasting systems can be designed, evaluated, and implemented to generate strategic value across multiple dimensions of hospitality operations performance. Future research should explore the following directions. First, the application of explainable AI frameworks—including SHAP value analysis and counterfactual explanation methods—would enhance the managerial accessibility and trust-building capacity of ML forecasting tools, particularly for non-technical end users. Second, adaptive online learning approaches that continuously update model parameters as new operational data arrive would address the temporal drift limitations of static model training. Third, multi-site comparative studies spanning diverse geographic, cultural, and institutional catering contexts would enhance the generalisability of findings and identify context-specific moderating factors that influence AI forecasting adoption outcomes. Finally, research examining the organisational change management dimensions of AI system implementation in hospitality settings would complement the technical and operational findings of the present study by addressing the human and cultural factors that ultimately determine whether analytically superior systems are adopted and used effectively in practice.  

Bandara, K., Bergmeir, C., & Smyl, S. (2020). Forecasting across time series databases using recurrent neural networks on groups of similar series: A clustering approach. Expert Systems with Applications, 140, 112896. https://doi.org/10.1016/j.eswa.2019.112896 Cavaleri, M., Mariani, M., & Baggio, R. (2021). Demand forecasting in food service: A machine learning approach. International Journal of Hospitality Management, 95, 102945. https://doi.org/10.1016/j.ijhm.2021.102945 Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785 Chiu, M. C., & Hsieh, Y. C. (2023). Food waste reduction strategies in the hospitality industry: An evidence-based review. Journal of Cleaner Production, 390, 136126. https://doi.org/10.1016/j.jclepro.2023.136126 Duan, Y., Liu, J., Fu, Z., & Zhang, Q. (2022). IoT-enabled smart inventory management in hotel food and beverage operations: Evidence from a longitudinal field study. International Journal of Contemporary Hospitality Management, 34(6), 2214–2238. https://doi.org/10.1108/IJCHM-08-2021-1027 FAO. (2023). The state of food and agriculture 2023: Revealing the true cost of food to transform agrifood systems. Food and Agriculture Organization of the United Nations. https://doi.org/10.4060/cc7724en Filimonau, V., & Gherbin, A. (2017). An exploratory study of food waste management practices in the UK grocery retail sector. Journal of Cleaner Production, 167, 1184–1194. https://doi.org/10.1016/j.jclepro.2017.07.229 Hevner, A. R., March, S. T., Park, J., & Ram, S. (2004). Design science in information systems research. MIS Quarterly, 28(1), 75–105. https://doi.org/10.2307/25148625 Hu, Y., Li, J., Hong, M., Ren, J., Lin, R., Liu, Y., Liu, M., & Man, Y. (2021). Short term electric load forecasting model and its verification for industry sectors in China. Energy, 237, 121501. https://doi.org/10.1016/j.energy.2021.121501 Kim, S., & Lee, H. (2023). Deep learning approaches to food demand prediction in institutional catering: Temporal patterns and operational insights. Computers & Industrial Engineering, 175, 108887. https://doi.org/10.1016/j.cie.2022.108887 Nikolopoulos, K., Syntetos, A. A., & Boylan, J. E. (2021). An empirical study of the accuracy of forecasting methods for catering demand management. Journal of the Operational Research Society, 72(4), 832–849. https://doi.org/10.1080/01605682.2019.1686138 Papargyropoulou, E., Wright, N., Lozano, R., Steinberger, J., Padfield, R., & Ujang, Z. (2022). Mapping the causes of food waste in the hospitality industry: A systematic review. Journal of Hospitality and Tourism Management, 51, 165–178. https://doi.org/10.1016/j.jhtm.2022.02.018