Characterisation of occupancy in building performance simulation for non-residential buildings

Building performance simulation (BPS) is a powerful tool for understanding energy consumption in buildings. Occupant behaviour in general, and occupancy in particular, has been identified as an important factor contributing to the gap between building performance simulation results and actual energy consumption. Oversimplified assumptions regarding occupancy, often used by modellers, are argued to widen this gap. Therefore, this thesis investigates the role of occupancy inputs in building performance simulation. It aims to examine various methods of representing occupancy, covering applications mainly designed for code compliance assessments and those used for design and operational assessments of buildings. In order to achieve these, a detailed analysis of the technical aspects of thirteen Energy Performance Certification (EPC) methodologies is conducted, focusing on their approaches to representing occupancy in calculations. This analysis includes ten official methodologies currently implemented in ten European countries, as well as three projects proposing “next-generation EPCs”. The findings reveal qualitative and quantitative differences among these methodologies regarding their occupancy input approaches. Key observations include variations in the magnitude of default parameters, temporal resolution, spatial scale, the level of detail in definitions, and the degree of standardisation applied in methodologies. These findings underscore the future efforts required to achieve the long-term goals of harmonisation of EPC assessment methodologies across the European Union countries. The results of the analysis can also inform the development of next-generation EPC methodologies. Additionally, the analysis highlights a growing trend of integrating compliance assessments with tools traditionally used for design and operational applications, as well as broader use of data-driven approaches. In order to gain insight into the attitudes of practitioners towards occupancy inputs in compliance simulation, a survey is conducted across Europe as part of this study. The results of the survey indicate that occupancy and activity inputs are among the least trusted inputs of EPC calculations. A comparison of survey responses between different countries reveals potential links between the level of standardisation of methodologies and the level of trust the practitioners have towards occupancy inputs. These results may indicate that EPC practitioners prefer models that represent the actual occupancy of the building to those applying default assumptions to simulation. The study also reviews the applications of data-driven occupancy inputs in three case study buildings to explore the implications. Data-driven occupancy inputs suitable for integration into BPS tools are generated from collected occupancy data and implemented in the simulations. These inputs are compared with the standard version (mainly designed for code compliance purposes), revealing significant differences. The standard profiles are shown to misrepresent the peak occupancy by values ranging between 34% and 98%, and the overall occupancy patterns across all case study buildings. Notably, the most substantial differences (72%-98%) were observed during academic break periods and weekends throughout the year. The implications of using data-driven occupancy on the simulation results are also examined. Additionally, the relative importance of occupancy compared to two other commonly standardised simulation inputs is assessed. These inputs include weather files, and electrical equipment and lighting usage. The influence of using actual weather files instead of using Typical Meteorological Year (TMY) files on the simulation results is investigated. Furthermore, the impact of employing data-driven electrical equipment and lighting inputs rather than standard versions is studied. The results indicate that the importance of empirical occupancy inputs relative to other inputs varies across different performance indicators. For peak heating loads, data-driven occupancy inputs have a more significant impact on the results (2% to 6%) compared to data-driven electrical equipment and lighting profiles (1%-3%), and a lower impact than using actual weather files (15% to 28%). Whereas for other indicators, like annual cooling demand, data-driven occupancy inputs are less important (with an impact of 22% to 39%) compared to data-driven electrical equipment and lighting profiles important (with an impact of 32% to 58%). The study also investigates the improvements in the performance gap resulting from the use of data-driven occupancy inputs. Results show that while occupancy inputs decrease the performance gap up to 7%, the magnitude of their impact varies across buildings and energy use categories. Finally, based on insights gained from the case study simulations and the analysis of EPC methodologies, a decision support framework is presented to help practitioners effectively incorporate occupancy inputs in BPS. This framework provides an overview of the potential pathway for implementing variations of occupancy inputs in BPS.

Grantor dc:publisher

Heriot-Watt University

Year dc:date.issued

2025

Author dc:creator

• Sayfikar, Mahsa

Advisor dc:contributor.advisor

• Jankins, Professor David

Statement dc:rights

• All items in ROS are protected by the Creative Commons copyright license (http://creativecommons.org/licenses/by-nc-nd/2.5/scotland/), with some rights reserved.

Language dc:language.iso

en

Repository record dc:identifier.uri

https://www.ros.hw.ac.uk/handle/10399/5262

OAI identifier oai:identifier

oai:ros.hw.ac.uk:10399/5262