Date of Publication :25th July 2024

Abstract:The Bullwhip Effect, a distortion of demand fluctuations inside a supply chain, continues to test traditional supply chain management. This study investigates the revolutionary potential of AI algorithms in minimizing the Bullwhip Effect using Artificial Intelligence Techniques for demand forecasting, dynamic inventory management, supplier relationship management, and real-time data exchange provides a comprehensive approach to optimize successful deployments, with real benefits such as cost savings, increased efficiency, and increased customer satisfaction. This work speaks about the demand fluctuation caused by hotel industry consumers and efficiently predicts the demand using an ensemble machine learning model. Inheriting the ensemble stacking the incorporation of two base learners (KNN and K means clustering) who have different characteristics are used to analyse the meal movement analysis and food type classification both combined to be the input of meta-model linear regression and with these factors, the meta-model predict the demand. Firstly using the k means clustering as a base learner which type of meals are consumed by whom is found and the food type classification is undertaken by the KNN base learner. Both of them are then given to meta-learner linear regressor to predict the food demand of the given region. Outputs look promising and the Mean Square Error (MSE) is used for performance evaluation.

Reference :