The main objective of the AGRICORE project is to develop a new generation of ABM tool taking advantage of the latest progress in computational science and ICT (including advances in big data, artificial intelligence algorithms, mathematical solvers and cloud computing services) as a means to overcome the challenges that are still hampering their capacity for improving new policies design and for performing the related socio-economic and environmental assessments at various geographic scales – from regional to global.

This main goal can be further split in the next list of specific objectives:

  1. To develop a European data-sources index tool. The partners will perform a comprehensive survey which will include EU statistics datasets, geo-reference databases, national and regional information sources and previous research results for the modelling of land use, policy, biophysical, social, economic and environmental aspects related to farming activities. The partners will identify synergies and analyse the integration of such information into the tools in AGRICORE.
  2. To minimise the time and user efforts currently required for the parameterisation and calibration of ABM models. The partners will develop a state-of-the-art combination of big data extraction and fusion followed by a combinatorial optimisation step to construct synthetic populations mimicking the distribution and characteristics of the real populations of interest.
  3. To develop an evolved agent-based model with improved capacity to model policies dealing with agriculture. The partners will elaborate on a dynamic quadratic model explicitly accounting for agent interactions and which computation is to be enabled by recent advancements in the capacities of mathematical solvers and ICT. This model structure will be a step ahead with respect to current ABM models to address main policy modelling challenges of today.
  4. To produce a behavioural model of farmers mimicking their decision-making rationale. The partners will collect direct feedback from target groups through participatory research involving farm representatives and associations. This will improve the understanding of factors acting on farmers and their possible responses to these factors. This work will imply the identification of key drivers and parameters likely to influence farmer decision-making and understanding how policies could lead to the implementation of different options by farmers. Thanks to the collected information and complementary data extracted from the multiple databases considered, a behavioural model will be developed by means of the application of artificial intelligence algorithms.
  5. To develop a flexible and integrated simulation suite. The partners will integrate all the modules composing the AGRICORE suite as a simulation environment ready for its use either for normative or positive purposes and which will have both ex-ante (for policy design) and ex-post (for monitoring) analysis capacity. Such an integrated suite will allow its connection with biophysical models and a large set of databases including multiple data sources and geo-referenced datasets (interoperability).
  6. To compile, analyse and show the produced information in an optimal way. The potential amount of information that influences the development of a policy at local, sectorial and global levels is huge. The proper visualisation of such an amount of information is challenging, and it is key to ensure an adequate decision-making process. Accordingly, the partners will rely on big data analysis and big data-oriented visualisation tools and libraries (such as visualisation maps and results from previous EU research projects on interactive visualisation charts and plots).
  7. To provide social, economic and environmental impact assessments of agricultural policies at farm, sector and global levels. The partners will design the AGRICORE suite to support the process of monitoring and assessing the impact of policies at farm, sector and global levels. Some of the impacts at farm level will be related to farm structure, production costs and land balance while at sector and global levels will include environmental and socio-economic factors. In addition, impacts will be assessed for the whole rural area with an emphasis on environment, development and jobs.
  8. To effectively integrate stakeholders’ knowledge and to cooperate with policymakers. The partners will cooperate with main stakeholders’ groups (including policymakers, researchers, data analysts and farmers) and will gather their needs and requirements in order to guide and enrich the AGRICORE design.
  9. To build a basis for credibility of the policy modelling work. The partners will apply the AGRICORE suite to the ex-post (2014-2017) and ex-ante (2018-2020) policy assessment of three use cases (UC1, UC2 and UC3), which have been selected to test the tool at various geographic scales (UC1 corresponds to the regional level while UC2 and UC3 aims to the national level) and for different policy instruments (UC1 policy instrument relates to environmental impacts, UC2 relates to the delivery of ecosystem services and UC3 relates to the socioeconomic aspects of the integration of agriculture in rural society). The results to be obtained will offer opportunities to publish in good level journals to keep researchers and policymakers at their institutions sufficiently motivated and to build a basis for credibility of the policy modelling work.
  10. To develop a highly modular and customisable tool to allow further improvements as needs arise. The AGRICORE suite will be implemented as a highly modular IT architecture composed of interchangeable and expandable modules so other researchers can contribute to its development as well. To facilitate this, AGRICORE will be released as an open-source project to be abundantly documented, communicated and disseminated (indeed, almost all deliverables will be made public). Additionally, the partners plan a set of dedicated actions towards clustering, coordination with policymakers and transfer of project’s results.