Neural Network-Based Decision Support Model for Seasonal Demand Strategy Selection

Seasonal demand fluctuations is a constraint encountered by the business personnel in inventory management and production planning. The strategies are formulated by the managerial to handle the situations of demand variations. This research work develops a neural network-based decision-making model to select the most feasible strategy to handle the seasonal demands. The proposed neural network model considers significant features such as historical demand patterns, storage costs, lead time variability, and promotional intensity, to predict the optimal strategy. The simulated data sets are used to train and validate the model. Inferences are drawn from the visualizations and classification results. This decision support system certainly facilitates the industrial decision makers to handle risks of inventory management more effectively.