Analysis of Inventories in the Agri-Food Sector through Probabilistic Models and application of Neural Networks.

Resumen

Objective: To evaluate the effectiveness of this hybrid approach in improving forecasting accuracy and optimizing inventory decisions under demand uncertainty for the agri-food sector.

Design/methodology/approach: Based on the determination of the coefficient of variation, the Newsboy model was implemented, along with demand forecasting techniques using neural networks, allowing inventory levels to be adjusted according to demand fluctuations and reducing the risk of both overstocking and stockouts.

Results: This model and the neural network optimize product availability, balance inventory costs, and maintain a solid foundation for decision-making within the agri-food supply chain.

Limitations on study/implications: This study focuses on an agri-food company, where probabilistic inventory models are applied to improve certainty in managing product availability under demand variability.

Findings/conclusions: Inventory optimization is essential in the agricultural sector, where demand variability can significantly impact both costs and service levels.