Determinants of spatio-temporal patterns of energy technology adoption: An agent-based modeling approach

• We present an agent-based model of residential solar photovoltaic (PV) adoption.
• Model integrates social, behavioral, and economic elements of agent decision-making.
• Real-world, large-scale integrated dataset used for model validation and testing.
• We study the importance of using disaggregated empirical data on model performance.
• Social and attitudinal components are critical for spatial and demographic accuracy.

Energy technology adoption is a complex process, involving social, behavioral, and economic factors that impact individual decision-making. This paper uses an empirical, geographic information system (GIS)-integrated agent-based model of residential solar photovoltaic (PV) adoption to explore the importance of using empirical household-level data and of incorporating economic as well as social and behavioral factors on model outcomes. Our goal is to identify features of the model that are most critical to successful prediction of the temporal, spatial, and demographic patterns that characterize the technology adoption process for solar PV. Agent variables, topology, and environment are derived from detailed and comprehensive real-world data between 2004 and 2013 in Austin (Texas, USA). Four variations of the model are developed, each with a different level of complexity and empirical characterization. We find that while an explicit focus only on the financial aspects of the solar PV adoption decision performs well in predicting the rate and scale of adoption, accounting for agent-level attitude and social interactions are critical for predicting spatial and demographic patterns of adoption with high accuracy.