Development and improvement of occupant behavior models towards realistic building performance simulation: A review

With the rise of concern about newly-designed or retrofitted buildings to have robust performance under different realistic scenarios, it is of vital importance to providing reliable energy predictions for building design and planning. Occupant behavior (OB), as one source of the significant uncertainties, is generally oversimplified as static schedules or predetermined inputs, which could cause a significant gap between the simulated and measured one. To bridge such gap, growing interests have been raised to understand the role of OB on building energy performance and develop OB models which can be integrated into building simulation tools. This paper aims to provide a systematic review with the focus on three important issues: a) the impact uncertainty caused by OB in building performance simulation and their differences in various spatial scales and temporal granularities; b) main criteria for the comparison and selection of modeling methods; c) requisite considerations to improve the performance of OB models. Based on this review, a framework was proposed towards improving the predictive performance of future OB models. Existing research gaps and key challenges for OB modeling are identified and future directions in this area are highlighted.