In this paper we demonstrate how agent-based modelling can be used to understand the emergence of a new infrastructure system, more specifically, a biogas infrastructure in the Netherlands. The innovative element in our modelling approach is the use of MAIA (modelling agent systems using institutional analysis), a meta-model for agent-based modelling of socio-technical systems, to conceptualise and gain insights into the complexity of infrastructure systems. Through our agent-based simulation model we were able to see how the BioNet infrastructure might evolve over three decades, under various conditions of social norms and institutions. We found that current social norms and institutions allow agricultural biogas production to be economically feasible without any subsidies. The simulations also reveal low expected returns on investment and significant risks for farmers in biogas projects.