Mathematical modeling of the animal feed manufacturing process (UK)

AB Agri is a £1.2 billion animal nutrition and agricultural services business. It operates across the agri-food supply chain research to retailer, straddling high-performance compound feeds and feed co-products the food, drink and biofuels industries.

The recipe of the animal feed is usually set by AB Agri’s customers and seek for the minimum ingredient costs, fulfilling the requirements of nutrients. AB Agri mills, as a service provider, is responsible for grinding and pelleting the feed. It appears that the cheapest grains were causing higher damages to the blades and require a lower production pace to guarantee a good quality product which is the responsibility of AB Agri. So they wanted to analyze whether the original recipe was the best cost effective, considering all these impacts. And consequently know how to price their services depending on the recipe.

Cassotis created a nonlinear model integrating the processes of grinding, mixing, pelleting and coating. The main variables that were optimized according to the recipe are: lifetime of the machines, steam consumption, productivity and spray quantity. The objective of the optimization is the overall cost, including material and process related costs.

The model presented combinations of decisions across the mill that reduced the overall cost (mainly in the coating phase), improving the quality of the final products and consequently reducing the rejection rate by the customers.