Exploring agent-based modeling for human-centered energy consumption prediction

People interact with buildings to make the indoor environment more comfortable thereby impacting energy consumption. However, these occupants' energy-related actions in buildings are not well accounted for using traditional simulation programs and normally are represented in a simplistic manner. Agent-based modeling (ABM) is widely used to predict real world situations and this approach has been used to model human behavior in different contexts. ABM has also been used in relation to energy consumption prediction. This study explores how ABM can be used to better model not only human behavior but also look into how occupant values (for instance thermal comfort, visual comfort, indoor air quality, perceived health and personal productivity) and indoor environmental preferences can be maintained while ensuring energy savings. The human behavior considered in this paper include occupants adjusting layers of clothing, using portable heating and cooling devices, adjusting thermostats, adjusting lighting levels, and using shading devices. Following a review of the available literature, the authors highlight the gaps and present recommendations to better address occupant preferences and behavior.