Use of Machine Learning to Predict Transition Success in Dairy Cows in an Automatic Milking System.

The transition period (3 weeks pre- to 3 weeks post-calving) is a critical time for dairy cattle and is recognised as a major contributor to health problems in dairy cows. Precision management of cattle, specifically prediction and/or identification of high risk individuals and groups, allows for corrective practices to minimize risk or impact of disease. Automated milking by robots provides an opportunity to utilise sensor data to predict transition success and aid in prevention of associated diseases and therefore offering the potential for large impact through improved transition management. This collaborative project, with a world-leading robotic milking technology company (Lely International), aims to develop algorithms to predict transition success using advanced data analytical techniques and multiple data streams. Tools developed will be integrated into Lely systems with the potential to provide significant impact on the dairy industry worldwide. The successful applicant will gain knowledge in the use of machine learning algorithms such as Neural Networks, K-nearest Neighbour, Support Vector Machines and Decision Trees. In addition, the industrial partner will provide the student with the opportunity to participate in work related to translation of research outputs to the industry and offer a unique training experience through exposure to Lely networks.

The research will be conducted at the 'Centre for Dairy Science Innovation' (CDSI) at Nottingham, utilising recent investments in this high-level research infrastructure. The successful student will also spend a period of time with the industrial partner, Lely International. Further information and Application: Applicants should have a first or 2.1 undergraduate degree (or a minimum of a 2.2 degree in addition to a Masters degree) in Animal Science, Veterinary Science, Applied Statistics, Veterinary Epidemiology or similar subjects, and should have a strong interest in quantitative analysis and epidemiology.

This industry linked PhD project, based at the School of Veterinary Medicine and Science, University of Nottingham and in collaboration with Lely International, aims to explore and develop algorithms to predict transition success in dairy cattle, using advanced data analytical techniques and multiple data streams. Prediction and/or identification of high risk individuals and groups, allows for corrective practices to minimize risk or impact of disease associated with transition, offering the potential for huge gains in transition cow management. Automated milking by robots provides an opportunity to utilise sensor data to predict transition success and algorithms developed from this project will be integrated into the systems of a world-leading robotic milking technology company. In addition the industrial partner (Lely) will provide the student with a unique training opportunity via their industry networks and offer the student opportunities to participate in work related to translation of research outputs to the industry.