Cameras and sensors provide data on pig housing behaviour, AI algorithms identify behavioural patterns. Image: ILV / University of Bayreuth.
Prof. Dr. Agnes Koschmider, University of Bayreuth. Photo: UBT.
The ProcessPig research project is an example of the increasing importance of artificial intelligence (AI) and data-driven process analysis in agriculture. The close cooperation in a network of partners from science and practice are expected to lead to important new insights for the future design of animal husbandry and sustainable action in the agricultural sector.
The project is linked to current European developments in animal husbandry. Due to legal requirements and growing public interest, more and more stables are being built with free ventilation and increased space for the animals or are being converted with these aspects in mind. Innovative housing systems envisage that the pigs can use different functional areas such as resting and defecation areas. In this way, the living conditions in the barn are to be more closely adapted to the natural needs of the animals and their behaviour is to be promoted accordingly. At the same time, there are efforts throughout Europe to reduce the harmful environmental effects of animal husbandry, especially the release of ammonia, which contributes to the acidification of soils and the eutrophication of water bodies. The new EU project aims to sustainably support these developments, which are aimed at promoting animal welfare and environmental protection, and to use the possibilities of artificial intelligence (AI) for this purpose.
"Sensors, video data and AI algorithms will in future form a monitoring system that will make it possible to analyse the pigs' behaviour in real time. Deviations from expected behaviour patterns are detected and visualised as key indicators. This will provide farmers with indications of potential problems and enable them to redesign the animals' housing conditions, not least with regard to the climatic conditions in the barn. By using AI-supported algorithms for behavioural recognition, the well-being of the animals can be significantly improved and environmental protection significantly promoted," says Prof. Dr Agnes Koschmider.
Dr. Andreas Melfsen from the Institute of Agricultural Process Engineering at Christian Albrechts University in Kiel emphasises the fundamental importance of this approach: "Our goal is to capture the correlations between pig behaviour and climatic conditions using AI-supported analysis of video and sensor data. In this way, we can detect at an early stage when animals deviate from their usual behaviour patterns due to unfavourable weather conditions and take appropriate measures."
The partner institutions from science and practice
The ProcessPig project is based on a network of research institutions, commercial enterprises and agricultural businesses. These include the Chair of Business Informatics and Process Analytics at the University of Bayreuth, the Institute of Agricultural Process Engineering (ILV) at Christian Albrechts University in Kiel, the company Lock Antriebstechnik GmbH in Ertingen, Baden-Württemberg, as well as five agricultural companies from Schleswig-Holstein and Baden-Württemberg. Associated partners from the fields of research, experimental facilities and agricultural extension strengthen the network.
Contact for scientific information:
Prof. Dr. Agnes Koschmider
Chair of Business & Information Systems Engineering and Process Analytics
University of Bayreuth
Phone: +49 (0)921 / 55-4583
E-mail: agnes.koschmider@uni-bayreuth.de
