How can mixed integer linear programming assist dairy manufacturers by integrating supply decisions and production planning?

Purpose - The dairy industry needs to make important decisions regarding its supply chain. In a context with many available suppliers, deciding which of them will be part of the supply chain and deciding when to buy raw milk is key to the supply chain performance. This study aims to propose a mathematical model to support milk supply decisions. In addition to determining which producers should be chosen as suppliers, the model decides on a milk pickup schedule over a planning horizon. The model addresses production decisions, inventory, setup and the use of by-products generated in the raw milk processing. Design/methodology/approach - The model was formulated using mixed integer linear programming, tested with randomly generated instances of various sizes and solved using the Gurobi Solver. Instances were generated using parameters obtained from a company that manufactures dairy products to test the model in a more realistic scenario. Findings - The results show that the proposed model can be solved with real-world sized instances in short computational times and yielding high quality results. Hence, companies can adopt this model to reduce transportation, production and inventory costs by supporting decision making throughout their supply chains. Originality/value - The novelty of the proposed model stems from the ability to integrate milk pickup and production planning of dairy products, thus being more comprehensive than the models currently available in the literature. Additionally, the model also considers by-products, which can be used as inputs for other products.