Abstract:
Buildings are recognized for their significant role in electricity consumption and carbon emissions. Policymakers and researchers have addressed the necessity of realizing the underlying reasons behind buildings' colossal energy consumption. In this regard, many energy-efficiency strategies have been developed to achieve sustainability and lower energy consumption rates. Building energy performance is also proved to be significantly affected by occupant presence and behavior. However, understanding the dynamic relationship between occupants and buildings is not easy due to the complexities of human behavior. Until a few years ago, many building energy performance tools did not even consider occupant behavior in their analyses, resulting in noticeable gaps between actual and predicted energy performance. Strategies involving behavioral changes are considered low-cost and effective methods in reducing building energy consumption. Although researchers have investigated occupants' role in different building types, the number of studies focused on dormitory buildings is limited. Occupant-building interactions in dormitories are more complicated than office buildings because of the differences in students' lifestyles and daily behaviors. In order to examine the role of students in the energy consumption of dormitories, an agent-based simulation was developed and validated using real-time consumption data collected from a dormitory building located on the Kilyos Campus of Boğaziçi University. Results show satisfying accuracy, and this study explains how the model can be used for energy consumption prediction. Some scenarios are also simulated with the model to demonstrate its capabilities for recommending effective occupant-centric energy-saving strategies. The model is adjustable and can be modified to be employed in other similar buildings. Moreover, this study paves the way for other researchers to use the agent-based simulation for occupancy prediction and building energy analysis and gives recommendations on improving and achieving a more sophisticated model.
