SimpleAgriData

Agriculture is also becoming increasingly digital, with more and more data being collected along the value chain. However, because this data comes from different sources and is stored in different formats, farmers are often unable to make sense of it and leverage its potential. The SimpleAgriData project is exploring how this data can be used in organic farming to improve resource efficiency, animal health and farm profitability.

The project has five main goals:

1. Creating an open source data platform to make the collected data across the entire value chain (in non-uniform formats) more usable for farmers and other stakeholders
2. Enable the easy exchange of data along the value chain by creating a standardized format and a digital twin of the entire process chain
3. Provide concrete recommendations and forecasts based on the data
4. Present data and recommendations in an understandable way to farmers
5. Increase resource efficiency, animal health and profitability in organic farming by using data and AI algorithms

We assume that our solution, in particular the open source data platform, will spread quickly in the market, as there has long been a high demand in the industry for standardised formats for simple data exchange.

The project focuses on stables used for organic livestock farming. They are equipped with a camera system that generates 24/7 video and sound recordings. This information is supplemented by environmental, feeding and veterinary data. Once this data has been successfully transferred to a data platform, extensive data analysis will be carried out using AI models (image segmentation, anomaly detection, forecasting). Moreover, we intend to create a digital twin of the value chain in order to optimize it and generate predictions. Information about animals and every relevant process should be available at any time and users should be able to easily simulate different scenarios. The digital twin and the AI models will be used in different use cases regarding animal health and optimization of processes in organic livestock farming. We intend to implement an open source data platform into which all data from the entire value chain (from parent animals, hatchery, feed, barn, animal health data to slaughter findings as well as animal tracking and barn climate data) can be fed and in which all data receive a standardized and unambiguous semantic description. Therefore we develop connectors that convert the formats of the input data into a generally readable format with clear semantics so that all data can be meaningfully analyzed and exchanged.

Ongoing project from 15.12.2024 to 14.12.2027