A food upcycling model by food bank collection-distribution networks

Nowadays, the collection and distribution of food products with high nutritional value and freshness for people in poverty has become a global problem due to financial, drought, or other crises. Food banks (FBs) are important entities that mitigate food waste by reusing surplus food at critical points in the food supply chain. This article investigates an FB network design problem for the collection and distribution of food items. An FB network comprises donors mapped from the food supply chain, FB itself, and beneficiaries mapped from charities. The problem addresses synchronous strategic, tactical, and operational decisions, including the location of FBs, the assignment of donors to main streams, the control of inventory, and the routing of vehicles in collection and distribution levels to optimize the amount of food reused. As the demand and supply of food items from charities and donors are uncertain, a robust fuzzy stochastic model is developed to model the problem with three objectives including cost, nutritional value, and freshness of food. An extensive numerical study compares these algorithms with respect to several criteria. The proposed novel MOGGWA heuristic showed superior performance and was ranked first by applying the TOPSIS multi-criteria decision-making method. The value of stochastic programming and the impact of the model on a real-size case study problem are shown, as well. • The collection and distribution network of a food bank is modeled. • A tri-objective model with cost, nutritional value, and food freshness is optimized. • A robust stochastic optimization model with fuzzy numbers is proposed. • Strategic and operational decisions are blended by applying the time value of money. • A novel heuristic approach is proposed and shown as superior to the existing ones.