City-scale Building Occupancy Prediction Using Geographic Information System Data

Abstract

The goal of this research is to assist in quantifying the impact of enhanced occupancy information on the accuracy of building energy models compared to actual 15-minute data. Through gridded population data extraction and cross-reference with building location and area information, daytime and nighttime occupant density data that could be leveraged to modify the occupancy schedules are generated. This work will improve the ability to quantitatively estimate and empirically validate reductions in energy use intensity of individual buildings within a large area. Further, this work will test the value for sources of occupancy data to improve utility-scale models and create measures to assess the energy, demand, emissions, and cost-saving opportunities for a utility. This paper provides the methodology and detailed techniques that are used to achieve this aim.