Structuring Socio-technical Complexity: Modelling Agent Systems using Institutional Analysis

Abstract

Socio-technical systems consist of many heterogeneous decision making entities and technological artefacts. These systems are governed through public policy that unravels in a multi-scale institutional context, which ranges from norms and values to technical standards. Simulation, agent-based modelling and simulation (ABMS) in particular, is an exploratory approach to gain insights into socio-technical systems and investigate the possible outcomes of policy interventions. However, to really understand and link various levels of behaviour in these systems and increase the usability of ABMS, we need to overcome the conceptual and practical limitations of this approach. In this research, we investigate how we can build social structures in agent-based models and how to increase the utility of ABMS for problem owners. To give social structure to agent-based models of socio-technical systems, we designed a modelling framework based on the Institutional Analysis and Development framework of Elinor Ostrom. To facilitate the use of ABMS by modellers with different levels of expertise, we provided tool support which also facilitates participatory model development. The overall outcome of this research is the MAIA modelling platform which consists of (1) a modelling framework for conceptualizing agent-based models of socio-technical systems, (2) a web tool that facilitates the conceptualization process, (3) transformation protocols that enable the translation of models into simulations and (4) software modules that facilitate semi-automatic translation of MAIA models to simulation code.