Knowledge Platform

Simulation of system/agents behaviour

In this group we include methods aimed at allowing representation of behavioural process suitable to measure effects of upstream changes in context or policy variables. These may be feed by all of the above, which provide parameters and behavioural rules into the simulation models. This method group also encompasses Agent-based models (ABMs) that simulate actions and interactions of autonomous agents (e.g. people, organisations in social sciences; resources in life sciences).

General Purpose and Application

Simulation of agents' behaviour is usually used to understand the effect of potential drivers, in particular policy and market variables, on agents' decisions and downstream effects. A whole stream of literature includes for example the effects of CAP reforms on farms' economic parameters (e.g. profitability) and choices (e.g. land use). Simulation models are largely used to support ex-ante policy evaluation and impact assessment exercises.

Lessons learned

Potential difficulties are related to the high effort needed to build new models, which tend to bring to the consolidation over time of existing models. This also implies that simulation models tend to be used for areas and problems for which they are already available, data-fed and calibrated. In addition, due to the above, there is some rigidity in the way usual economic models may adapt to new problems. Finally, being often based on aggregated data (e.g. regional averages) and aggregated economic concepts (e.g. elasticities) they represent spatial and technical phenomena in a very simplified and approximated way.

The ABM approach has been used in the region of Ferrara Lowlands (Italy) to address some of these limitations and hence improve our understanding of socio-economic and biophysical components in the rural landscape by taking into account the heterogeneity of farm-agents as well as the spatial dynamics in land ownership and intensity of land uses modified by the agro-environmental schemes application. ABM has the ability to represent complex mechanisms, by explicitly including in the model formulation the relevant social interactions, and to apply it to broad context. The methods' strengths are also their weaknesses. They require specification of social interactions and mechanisms that are not yet fully understood and their results are difficult to validate, given the complexity they take into account. The experience of the analysis of impacts of shelterbelts and CAP's greening measures on landscape composition and farm performance in the Chlapowski landscape park (Poland) revealed the interest for expanding well established agricultural economic models in the direction of accounting for landscaper elements, particularly for detailed representation of policy impact at farm level.

Further Reading

Grimm, V. & Railsback, S.F. (2005) Individual-based Modeling and Ecology. Princeton University Press, Princeton.