Data science for precision management of beef cattle nutrition: a solution leveraging artificial intelligence, big data and sustainability (funded PhD project)


   Department of Agriculture and Environment

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
  Dr James McCaughern  No more applications being accepted  Funded PhD Project (UK Students Only)

About the Project

Please note this project is only open to applicants who qualify for home (UK) student fees.

Primary supervisor: Dr James McCaughern, [Email Address Removed], Department of Agriculture and Environment  

Second supervisors: Dr Ed Harris, Professor Karl Behrendt, and Professor Jude Capper

Project title: Data science for precision management of beef cattle nutrition: a solution leveraging artificial intelligence, big data, and sustainability

 Project Description:

Background: Recent technological advancements, particularly in relation to sensor technologies, create opportunities to monitor and manage animal nutrition, following the example set by other fields such as livestock health and welfare [1]. These range from methodologies for data analysis and decision support, to using automated feeders and individual animal tracking systems [2,3]. Precision animal nutrition aims to utilise technology to manage animal diets for optimal performance, profitability, and environmental outcomes [4]. The nutritional requirements of beef cattle vary as a result of a plethora of factors including: age, performance, activity and environmental conditions [5]. To implement precision management of beef cattle nutrition on farm, three key parameters are required: animal live weight, average daily gain (ADG), and feed intake [4]. The traditional methods of measuring these metrics on farms, however, can be time consuming, expensive, and may require trained technical expertise [6]. Recent work indicates that it may be possible to improve beef finishing profitability by over 38%, while concurrently reducing enteric greenhouse gas emissions by 30% if key performance parameters were available to support real-time decision making [7]. Despite the benefits of these key performance indicators, a large proportion of farms do not collect basic liveweight information, with only 42% of livestock farms in the UK owning or using weigh scales to monitor animals [8]. 

The ability to analyse both pictures and video via artificial intelligence (AI) may serve as a solution to this problem, by providing farmers with a real time overview of what is going on in beef production systems, plus detailed individual animal information. Evidence that this technology can be utilised for this purpose, comes from pig production, where 3D images have been used to successfully predict pig live weight [9], and dairy production systems, where images have also been used to accurately predict cow body composition [10] Similar systems could therefore be developed to monitor the key performance metrics in beef production systems, facilitating management changes that optimise nutrient supply to the animal, and reducing environmental impacts, whilst enhancing overall system profitability [4] The National Farmers Union has pledged to reach net zero by 2040, which presents a significant challenge to livestock producers, particularly for those producing beef cattle given that enteric methane contributes a considerable proportion of total greenhouse gas emissions [12]. Real time analysis using AI has the potential to significantly improve the sustainability of British beef production, and therefore the global competitiveness of the sector, by taking decades of animal science knowledge and making it wholly accessible at the farmers fingertips, for the benefit of all.  

Project Aim: To demonstrate the use of artificial intelligence to manage beef nutrition, and thereby create more sustainable beef production systems.

Proposed methodology: Cattle performance on beef finishing systems will be monitored on-farm from arrival through to slaughter at target end weight, with curation of all associated performance data. Data to be collected will include (but not be restricted to): images and video of individual cattle, body condition and growth (ADG), feed intake, start and end weights, slaughter data and wider farm information (e.g. dietary feed composition). This data will be used to generate algorithms, such that a farmer can use the images and data to predict precisely what feed should be provided and when the animal should be slaughtered to minimise resource use and greenhouse gas emissions whilst improving productivity. Once developed, this information system will be used to collect ground truth data in a variety of beef production systems, which will help to validate the system across the range of environments in which it maybe deployed.

The stipend for this studentship is £16,062 per annum based on the standard UKRI 2022/23 rate. The expected start date for the studentship is 27th September 2022, although a later start date will be considered for the best applicant. The student will be registered for a PhD at Harper Adams University and based at Harper Adams University, Edgmond, Shropshire, UK. During this project, the successful applicant is expected to develop sought-after technical skills in the fields of data science and animal nutrition.

Person specification:

Candidates will normally be expected to hold a bachelor's degree with a first or upper second class, or a high GPA in an appropriate subject area (e.g. data science, computer science, engineering, agricultural science or a related subject area). A relevant masters degree maybe an advantage. Please note this project is only open to applicants who qualify for home (UK) student fees.


References

References: NASEM (2016) NAP, USA; 2. Nkrumah et al (2006) J Anim Sci 84:145-153; 3. Deng et al (2017) J Clean Prod 142:758-766; 4. González et al (2018) Animal 12:s246-s261; 5. AFRC (1993) CABI, UK; 6. Hewitt et al (2018) Livestock 23:72-78; 7. Ford et al (2022) CIEL, UK; 8. DEFRA (2019) FPS, UK; 9. Pezzuolo et al (2018) Comput Electron Agric 148:29-36; 10. Fischer et al (2015) 98:4465-4476; 11. Kamilaris et al (2020) J Clean Prod 253:119888.
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs