Why and how can agent-based modelling be applied to community energy systems? A systematic and critical review

Community energy systems (CESs) are key elements for the local energy transition and are receiving considerable attention. In this context, agent-based modelling and simulation (ABMS) is becoming one of the prominent computational modelling approaches for studying CESs. However, no systematic and critical review of such models has been conducted. Therefore, this study investigates the studies that used ABMS to study CESs, comprehensively analyse and structure this branch of literature, and suggest future research avenues. The six following elements from the Overview, Design concepts and Details (ODD) are used: (i) modelling purposes, (ii) agents and their variables, (iii) concepts and theories, (iv) initialisation and narrative, (v) network structures and interactions, and (vi) input data. The analysis demonstrated that particular purposes (e.g. social learning) and specific agents (e.g., energy companies) are neglected in the modelling practices. Furthermore, the developed ABMS are mainly focused on economic discipline, including topics such as local market design, and topics related to environmental and behavioural/institutional disciplines are largely missing. Such domination also reflects on the narratives and the networks used in the modelling. The most studied cases are focused on European countries and electricity-generating CESs based on solar photovoltaics. Further insights and future research avenues are elaborated on in detail.