Automation of the modern commercial feed mill: Optimization of pellet quality using machine learning 2.0

Lead Applicant: Jennifer Ellis

Research Priority: Competitive Production Systems 

Program Type: Tier 1

Funding Cycle: 2023/2024

Research Centre: NA

Research Summary: Feed manufacturing is often described as an art over a science. Many factors including raw material attributes and quality, formulation, season and temperature, and operator and feed manufacturing parameters (temperature, pressure, speed, etc) influence the quality of the end product, such as the pellet durability index (PDI). Machine learning as applied to big data is a disruptive technology that has the ability to map patterns in complex data and is ideally suited for the commercial feed manufacturing environment. This project aims to take the next big step beyond previous work to expand involvement to additional feed mills across multiple companies in Ontario to create a robust and versatile machine learning system to predict PDI across Ontario mills and develop an optimization algorithm to aid in decision making when using the model.