PrecisionBeef
Lead Research Organisation:
University of Strathclyde
Department Name: Electronic and Electrical Engineering
Abstract
Optimising animal productivity is critical to maintaining a competitive and sustainable UK beef industry with production efficiency the greatest single opportunity to reduce primary production costs. There is considerable inefficiency in the beef sector which increases variable farm costs, reduces the yearly capacity of finishing units, and reduces profitability due to sub-optimal marketing of animals. The reduced revenue associated with these inefficiencies arise for 3 main reasons; (1) retaining cattle on-farm beyond the optimum point of marketability leading to extra feed, bedding and fixed costs; (2) reductions in sale revenue due to these over-finished cattle being out of desired specification; (3) loss of productivity and efficiency due to poor animal health.
The aim is to develop a state-of-the art solution for beef farmers to optimise their business efficiency. At the core of the project is the development of a near infra-red (NIR) system to characterise feed (dry matter content, nutritional composition) as it exits a feeder wagon. Also pivotal to the project is the development of animal-mounted sensors to measure feeding behaviour (eating and rumination patterns). The bulk feed characteristics will be integrated with the feeding behaviour data with a target of providing a robust, accurate and innovative method of determining individual animal feed intake. The final solution will be a cloud-based decision support platform integrating individual animal feed intake and behaviour data, with measures of animal performance. This will provide the support tools necessary to quantify performance and efficiency of individual animals and improve the sustainability of the production process. It is anticipated that by closely monitoring individual animals using the proposed system, the finishing period of the animal will be reduced on average by 14 days, while animals performing poorly dues to illness will be flagged up to the farmer allowing for earlier intervention.
The aim is to develop a state-of-the art solution for beef farmers to optimise their business efficiency. At the core of the project is the development of a near infra-red (NIR) system to characterise feed (dry matter content, nutritional composition) as it exits a feeder wagon. Also pivotal to the project is the development of animal-mounted sensors to measure feeding behaviour (eating and rumination patterns). The bulk feed characteristics will be integrated with the feeding behaviour data with a target of providing a robust, accurate and innovative method of determining individual animal feed intake. The final solution will be a cloud-based decision support platform integrating individual animal feed intake and behaviour data, with measures of animal performance. This will provide the support tools necessary to quantify performance and efficiency of individual animals and improve the sustainability of the production process. It is anticipated that by closely monitoring individual animals using the proposed system, the finishing period of the animal will be reduced on average by 14 days, while animals performing poorly dues to illness will be flagged up to the farmer allowing for earlier intervention.
Technical Summary
Optimising animal productivity is critical to maintaining a competitive UK beef industry with production efficiency the greatest single opportunity to reduce costs. There is considerable inefficiency in the beef sector (biologically and systems-based) which increases variable farm costs, reduces yearly capacity of finishing units, and profitability due to sub-optimal marketing of animals. From a biological perspective large between-animal variation in feed efficiency has been proven but in practice, it is difficult for farmers to measure the performance efficiency of individual animals. Therefore this project will provide producers with an accurate measure of individual animal feed intake and performance i.e. growth rate. These data will be integrated on a decision support system which will allow farmers to assess individual animal performance. New and existing (enhanced) technologies (NIR, KN Pace, animal-mounted sensors) will be used to measure individual animal feed intake in an innovative way achieved by incorporating NIR onto the existing KN system to provide accurate quantification and composition of each batch of feed and integrating this with feeding behaviour data from novel animal-mounted sensors (robust and non-intrusive). Relating this information with measures of performance will enable the farmer to make effective management decisions to optimise use of feed, thus maximising productivity and profitability. iii) There is increased pressure in the beef sector due to rising feed prices and consumer requirements for cheaper produce. The precision farming solutions developed will allow farmers to identify animals that are inefficient (consuming high quantities of food but growing slowly) or performing poorly. iv) The system provides a means of earlier detection of health issues manifest through poor performance e.g. pneumonia and optimising efficiency provides significant potential to reduce greenhouse gas emissions.
Planned Impact
The following areas will benefit from the proposed research:
1. UK beef producers
a. The proposed system will allow farmers to identify inefficient and poorly performing animals and help them to make informed decisions to increase the overall efficiency of their beef production unit.
b. The system will allow farmers to accurately measure the composition of the diet given to each group of animals and allow for more accurate formation of diets to fulfil the nutritional requirements of the animals.
c. Use of the proposed system will allow the farmer to optimally market their animals (i.e. to meet optimum market specification). This will reduce the number of animals kept on farm beyond their optimal point of marketability, thus reducing finishing times on average. This will reduce the variable costs associated with beef production (such as feeding and bedding) and allow for a higher throughput of animals through finishing units, thus optimising the productive output and improving the economics of their business. It will also prevent abattoir cost-penalties associated with over-finished (i.e. too fat) animals.
d. As the UK commercial partners will have first access to the technology they will be the first to benefit. Reducing farm costs should increase the competitiveness of the UK beef industry and make their products more competitive against foreign competitors.
2. Meat processers
a. By sourcing animals from producers using the proposed system, meat processers will receive animals which are optimally finished, with more desirable carcass conformation and fat grades and killing-out percentage. This will lead to more efficient processing with reduced labour requirements to trim fat from over-fat animals and reduce costs associated with fat disposal.
b. Sourcing animals finished using this system will also allow for increased uniformity of the product for retail, as the animals will be marketed at the optimal market specification.
3. Consumers
a. The use of this system will likely result in cheaper beef being available to the consumer as variable costs during the production process will be reduced.
b. Cheaper UK beef will make it easier for consumers to choose local products over foreign alternatives.
4. The UK
a. The technological systems proposed will enhance the economic efficiency of UK beef production, thus increasing the sectors competitiveness over imported beef and guaranteeing the sustainability of the UK beef industry.
b. Reduced production costs and efficient production methods could enhance the reputation of UK beef and increase the value of UK beef exports.
c. There will be a reduced environmental footprint from more efficient beef production, reduced farm resources and reduced abattoir waste. The quantity of beef produced per unit of greenhouse gas will be reduced.
d. Increasing the profitability of the UK beef farming sector will lead to social benefits including enhanced rural employment.
5. Animal welfare
a. Increased monitoring of animal performance will allow for poorly performing animals to be identified earlier. Poor performance often manifests through ill health, often before the clinical signs of illness become apparent. Therefore, the use of this system will allow for illnesses to be detected and treated earlier, thus reducing the negative impacts of illness (in cost and waste).
1. UK beef producers
a. The proposed system will allow farmers to identify inefficient and poorly performing animals and help them to make informed decisions to increase the overall efficiency of their beef production unit.
b. The system will allow farmers to accurately measure the composition of the diet given to each group of animals and allow for more accurate formation of diets to fulfil the nutritional requirements of the animals.
c. Use of the proposed system will allow the farmer to optimally market their animals (i.e. to meet optimum market specification). This will reduce the number of animals kept on farm beyond their optimal point of marketability, thus reducing finishing times on average. This will reduce the variable costs associated with beef production (such as feeding and bedding) and allow for a higher throughput of animals through finishing units, thus optimising the productive output and improving the economics of their business. It will also prevent abattoir cost-penalties associated with over-finished (i.e. too fat) animals.
d. As the UK commercial partners will have first access to the technology they will be the first to benefit. Reducing farm costs should increase the competitiveness of the UK beef industry and make their products more competitive against foreign competitors.
2. Meat processers
a. By sourcing animals from producers using the proposed system, meat processers will receive animals which are optimally finished, with more desirable carcass conformation and fat grades and killing-out percentage. This will lead to more efficient processing with reduced labour requirements to trim fat from over-fat animals and reduce costs associated with fat disposal.
b. Sourcing animals finished using this system will also allow for increased uniformity of the product for retail, as the animals will be marketed at the optimal market specification.
3. Consumers
a. The use of this system will likely result in cheaper beef being available to the consumer as variable costs during the production process will be reduced.
b. Cheaper UK beef will make it easier for consumers to choose local products over foreign alternatives.
4. The UK
a. The technological systems proposed will enhance the economic efficiency of UK beef production, thus increasing the sectors competitiveness over imported beef and guaranteeing the sustainability of the UK beef industry.
b. Reduced production costs and efficient production methods could enhance the reputation of UK beef and increase the value of UK beef exports.
c. There will be a reduced environmental footprint from more efficient beef production, reduced farm resources and reduced abattoir waste. The quantity of beef produced per unit of greenhouse gas will be reduced.
d. Increasing the profitability of the UK beef farming sector will lead to social benefits including enhanced rural employment.
5. Animal welfare
a. Increased monitoring of animal performance will allow for poorly performing animals to be identified earlier. Poor performance often manifests through ill health, often before the clinical signs of illness become apparent. Therefore, the use of this system will allow for illnesses to be detected and treated earlier, thus reducing the negative impacts of illness (in cost and waste).
Organisations
Publications
Davison C
(2020)
Detecting Heat Stress in Dairy Cattle Using Neck-Mounted Activity Collars
in Agriculture
Davison C
(2021)
Predicting feed intake using modelling based on feeding behaviour in finishing beef steers.
in Animal : an international journal of animal bioscience
Davison C
(2023)
Feed Conversion Ratio (FCR) and Performance Group Estimation Based on Predicted Feed Intake for the Optimisation of Beef Production.
in Sensors (Basel, Switzerland)
Hamilton AW
(2019)
Identification of the Rumination in Cattle Using Support Vector Machines with Motion-Sensitive Bolus Sensors.
in Sensors (Basel, Switzerland)
Michie C
(2017)
Wireless MEMS Networks and Applications
Michie C
(2020)
The Internet of Things enhancing animal welfare and farm operational efficiency.
in The Journal of dairy research
Miller GA
(2020)
Using animal-mounted sensor technology and machine learning to predict time-to-calving in beef and dairy cows.
in Animal : an international journal of animal bioscience
Pavlovic D
(2021)
Classification of Cattle Behaviours Using Neck-Mounted Accelerometer-Equipped Collars and Convolutional Neural Networks.
in Sensors (Basel, Switzerland)
Pavlovic D
(2022)
Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars.
in Sensors (Basel, Switzerland)
Description | The project has developed animal-mounted (collar-mounted) sensor technology for the beef production sector to predict eating and rumination behaviours on an individual animal basis. We have quantified for the first time the degree of precision that individual animal feed intake can be estimated from feeding time budgets measured using experimental feeding technology (fee weigh systems) or using animal-mounted technology. We are progressing this work to calculate the feed efficiency, using integrated data streams (feeding behaviour, feed characterisation and growth) which was in essence, the goal behind the project. While there remains work to be done to improve on the overall prediction accuracy, we have demonstrated that high performing animals can be identified within the herd using animal-mounted technology and integrated data. This aligns entirely with what was proposed within the application and is of strong interest to industry partners Afimilk , Harbro and Keenan. |
Exploitation Route | The application of the measurements is of interest within the beef cattle production sector and also within the dairy sector. It enables farm operatives to gauge the performance of their stock in relation to production of outputs (beef or milk) with respect to feed intake. Initial gains are likely to emerge from identifying poorly performing lines and high performing animals in order to optimise breeding selection. |
Sectors | Agriculture Food and Drink |
Description | 5G Rural-First: Rural Coverage and Dynamic Spectrum Access Testbed and Trial |
Amount | £493,268 (GBP) |
Funding ID | 5GRuralFirst |
Organisation | Department for Digital, Culture, Media & Sport |
Sector | Public |
Country | United Kingdom |
Start | 03/2018 |
End | 09/2019 |
Description | Digital Dairy Value-Chain for South-West Scotland and Cumbria (Strength in Places Fund) |
Amount | £21,000,000 (GBP) |
Organisation | Innovate UK |
Sector | Public |
Country | United Kingdom |
Start | 02/2022 |
End | 02/2027 |
Description | FOSTERING PRECISION AGRICULTURE AND LIVESTOCK FARMING THROUGH SECURE ACCESS TO LARGE-SCALE HPC-ENABLED VIRTUAL INDUSTRIAL EXPERIMENTATION ENVIRONMENT EMPOWERING SCALABLE BIG DATA ANALYTICS |
Amount | € 14,309,650 (EUR) |
Funding ID | 825355 |
Organisation | European Commission H2020 |
Sector | Public |
Country | Belgium |
Start | 01/2019 |
End | 12/2023 |
Description | Internet of Food and Farming 2020 |
Amount | £30,000,000 (GBP) |
Funding ID | 731884 |
Organisation | European Commission |
Sector | Public |
Country | European Union (EU) |
Start | 01/2017 |
End | 12/2020 |
Description | Precision Livestock Farming (PLF) Technologies to Reduce Greenhouse Gas (GHG) Emission Intensity |
Amount | £125,000 (GBP) |
Funding ID | SCF0218 |
Organisation | Department For Environment, Food And Rural Affairs (DEFRA) |
Sector | Public |
Country | United Kingdom |
Start | 09/2019 |
End | 09/2022 |
Title | Bolus Sensor Acceleration Data With Timestamp and Behavioural Classification |
Description | The dataset consists of 1 .csv file containing 107 hours of continuously captured 3-axis acceleration data from a bolus sensor that was placed inside the rumen of a cow. The file also contains behavioural classification: 0 = other, 1 = rumination and 2 = eating. The classifcations are derived from an activity collar that was placed on the cow during the same period to provide the classifications. |
Type Of Material | Database/Collection of data |
Year Produced | 2020 |
Provided To Others? | Yes |
URL | https://pureportal.strath.ac.uk/en/datasets/5e1dfedc-87c7-4af0-bb20-3f3e1388d9d9 |
Title | Bolus acceleration data 3x |
Description | Python pickle files containing raw acceleration data from internal bovine bolus acceleration sensors. Collected from 3 cows in period July 2015. |
Type Of Material | Database/Collection of data |
Year Produced | 2019 |
Provided To Others? | Yes |
Description | "Advances in monitoring of livestock" European Association of Animal Science, Warsaw August 2015. |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Plenary talk at European Association of Animal Science, Warsaw August 2015, Session 18. The talk included elements of Precision Beef and built on other programmes. Dave Ross, Craig Michie, Carol-Anne Duthie, Shane Troy, Ivan Andonovic were contributors. The talk was delivered by Dave Ross. |
Year(s) Of Engagement Activity | 2015 |
Description | Application of Motion Sensors within Beef finishing |
Form Of Engagement Activity | Participation in an open day or visit at my research institution |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Professional Practitioners |
Results and Impact | The application of accelerometer based collars to identify eating and rumination patterns was delivered by Hugh Thomson, research collaborator (Harbro) to around 130 farmers engaged in beef finishing. |
Year(s) Of Engagement Activity | 2015 |
Description | Artificial Intelligence Enabled by Sensor Innovations |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | Workshop: Sensors and Measurement for Agricultural Applications, Harper Adams, Newport, UK. This work was delivered to a special interest grouping of practitioners who are responsible for the development of technology to support the delivery of precision agriculture. |
Year(s) Of Engagement Activity | 2018 |
Description | Cloud-based Animal Health Service Provision |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Postgraduate students |
Results and Impact | This was a presentation at the 11th Annual International Symposium on Agricultural Research, Athens, Greece, July 2017 |
Year(s) Of Engagement Activity | 2017 |
Description | Innovation in Scottish agriculture - the Internet of Agricultural Things event 27 February 2018 |
Form Of Engagement Activity | A formal working group, expert panel or dialogue |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | The understanding of the role of technology within agriculture that has been developed through working within the Precision Beef program and similar activities was presented to MPs at an event hosted within the Scottish Parliament. Prof I Andonovic presented an assessment of how 'Internet of Things' solutions can assist in the support of beef and dairy production and beyond. The presentation was supported by Prof R Stewart also of Strathclyde who outlined the opportunity for optimising rural connectivity through the delivery of 5G services within rural environments. This presentation outlined plans for trailing new business models and developing innovative farming practices through the combination of emerging sensor and communications technologies. The presentations form part of an information gathering process carried out by the Scottish Government in relation to policy formulation. |
Year(s) Of Engagement Activity | 2018 |
Description | Presentation to the IET Policy Group on Internet of Things and Application to Agriculture |
Form Of Engagement Activity | A formal working group, expert panel or dialogue |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Policymakers/politicians |
Results and Impact | Dr Craig Michie presented an overview of the application of wireless sensor technology, Internet of Things, within precision agriculture. Through a description of how such technology has been applied with the Precision Beef project, its objectives and progress, an illustration of the application of technology within this context was provided. As a result of this presentation a request was made to follow this up with a presentation at a Holyrood briefing event in order to communicate concepts and understanding to MPs and policy makers within the Scottish Parliament. |
Year(s) Of Engagement Activity | 2017 |