Modelling the behaviour in socio-technical systems for policy assessment - A comparison of modelling approaches using the example of Sponge City concept implementation

Abstract

How are policies affecting the social behaviour of actors and thus the dynamics of the corresponding socio-technical systems? This question is of central relevance to understand what policies might be effective for sustainable transitions and transformation where social changes are immanent. Exploratory modelling approaches such as agent-based modelling (ABM) are suitable and established to analyse how individual, especially heterogeneous behaviour emerges to affect social systems. In contrast to this micro-level perspective of understanding how social behaviour influences system behaviour, System Dynamics models (SDM) focus on the macro-level perspective by analysing how system interactions drive complex system behaviour. Qualitative modelling approaches such as Causal Loop Diagrams belonging to the Systems Thinking Modelling (STM) toolbox remain descriptive but need much less effort for model development. In this article, we compare how the different approaches ABM, SDM and STM can model social behaviour in socio-technical systems and what different kind of insights can be gained to assess the efficacy of policies. It discusses the advantages and limitations of the different approaches for their application in behavioural public policy modelling for sustainability transition. Therefore, three exemplary models (STM, SDM, ABM) are created for the same topic of implementation dynamics of adaptation measures to foster rainwater retention (Sponge City concept) in a multi-residential quarter. As a result, the risk of green gentrification is identified. Three hypothetical policies are implemented to illustrate the differences between the three modelling approaches for their use in policy assessment. Based on that, we discuss how the three modelling approaches conceptualise behaviour in general.