Simulating the sensing of building occupancy

Accurate building occupancy information can be beneficial in minimizing energy use by improving the intelligence of a Building Automation System (BAS) and helping designers predict the effect of different design options on occupant behavior. However, current occupancy measurements are quite inaccurate due to limitations in sensing technology and the resulting discrepancies between sensor data and what actually happens. In this paper we explore the use of simulation to model occupant behavior in combination with motion sensors to be able to study the relationship between known and measured occupant behavior. An extensible occupancy model, influenced by computational cognitive science and implemented using established modeling conventions is presented along with a simple experiment comparing the effects of different sensor density levels.