Waste reduction model trough DDMRP & Machine learning to achieve zero hunger: A Peruvian fishing company case

The food industry faces the persistent and multifaceted problem of food waste at all levels of its supply chain, significantly impacting sectors that deal with perishable products. This issue primarily arises due to inadequate and inefficient purchasing and replenishment management practices. In the context of Peru, fish stands out as a crucial product, with the fishing and food industry experiencing a remarkable growth rate of 27% over the last decade. This sector now contributes approximately 4% to the national manufacturing GDP and employs a significant portion of the labor force, underpinning its economic importance. However, despite its growth and significance, this industry suffers from substantial levels of waste, accounting for 23% of the total food waste generated in its processes. This is particularly concerning given the global context where hunger levels are growing in Peru and 828 million people around the world suffer from hunger. Motivated by the pressing need to reduce global food waste, particularly in the fish industry, we have developed an integrated and innovative methodology. This methodology combines Demand Driven Material Requirements Planning (DDMRP) for effective inventory management with advanced Machine Learning techniques for accurate demand forecasting. Our approach meticulously identifies the optimal replenishment policy, leading to significant reductions in waste. The study conclusively demonstrates that the application of these advanced techniques can effectively and efficiently address food waste issues within the fish industry. Based on comprehensive bibliographic evidence, it has been shown that it is possible to reduce waste levels by up to 75% using the described tools and methodologies. This research offers a promising pathway for other industries facing similar challenges.