Agriculture Data Space: AI-based multicriteria approach for Advanced information in agriculture
- Introduce the key challenges addressing the sector related to data sharing and interoperability
- Introduce the key components of an effective Agriculture Data Space addressing the main challenges identified
- Elaborate on existing technologies and processes able to support knowledge sharing and development of advisory services for policy makers and farmers.
- Stimulate the discussion among the panelists and the audience, encouraging the debate between different perspectives
This session will address the Agriculture domain, one of the most critical sectors in Europe both in terms of GDP and in terms of the social welfare of the European population; in fact food sector plays a critical role for addressing most of the SDGs (Strategic Development goals) directly (SDG 1,2, 3, 12, 15) and indirectly (all the others). This was further demonstrated by both the COVID pandemic and through the current geopolitical crisis , where ensuring food safety, sovereignty and independence from regions outside Europe is of critical value. Therefore, the need to take steps to ensure our continent remains productive, sustainable and climate-neutral, in line with the Green Deal objectives.
The availability of huge data and the use of Farm Management Information Systems (FMIS) in agriculture are themselves insufficient to guarantee efficiency, prosperity and control of agriculture sector:
- not all FMIS are suited to all types of production, environments and local conditions;
- not all FMIS use the same level of inputs, place equal restrictions on the use of pesticides, or similar levels of ecological and environmental sustainability approaches.
- Not all FMIS support efficient knowledge management aimed towards knowledge sharing and innovation.
- Not all farmers are able to successfully interpret and utilise data coming from available FIMS;
- Having real time data in this topic doesn’t provide the entire scope of what is needed, what is going on around the field and on the market.
A new generation of FMISs will target Knowledge AI and will emerge in the next few years on the basis of multi-criteria decision models capable of including environment-oriented parameters and provide output more akin to advisory services then a mere set of data or information. The availability of domain specific Data Spaces in agriculture will help to collect more interoperable, high-quality data, make systems from different vendors interoperable, and promote data sovereignty. However, this is not enough to guarantee real advancement for the Agricultural Data Economy. AI and ML will play a crucial role in handling multi-criteria data management, where data will be integrated using specific algorithms that will in turn send back structured, more aware information. Strong investment will be needed from a range of organizations in order to build a Knowledge Management (KM) infrastructure that cleans, organizes, and provides structure to this information. Without this investment, these organisations will be unable to successfully deliver on their AI plans. KM is foundational to achieve some of the more advanced information management, findability, and semantic web capabilities organizations refer to as AI1.
This approach can benefit both business and public interest based on the relevant stakeholder where the solution is addressed. This workshop aims to give the opportunity to describe and discuss conditions whereby business, public and social/environment strategies can co-exist harmoniously, and where needs, surplus and potential benefits can be shared via a real ecosystem. Key innovations and challenges coming from running EU projects, such as DIVINE, AgriDataSpace CSA, AgriTEF, will be introduced .
Presentations of the session: