Master Thesis: Combating User-Behaviour Caused Variations with Robustness in Building Design

Motivation: In the current design practice, many assumptions about the building operation are considered by the complex dynamic computer simulations to estimate building performance. But occupant behaviour is one of the most fluctuating boundary conditions that has a significant influence on a building’s performance. A holistic solution to tackle the regularly expected variations in occupant behaviour and, in turn, reduce Performance Gap is to design robust buildings. Robustness in building performance is the ability of the building to withstand variations without significantly affecting its performance in terms of energy consumption or thermal comfort. In simpler terms, robustness is the stability and reliability of a building’s performance. An evaluation of robustness of constructed buildings can provide insights about the practical implications of their design decisions and support the theoretical studies of building robustness and robust optimization.
Methodology: An uncertainty analysis of variable user behaviour of two constructed office buildings in Germany was done to compare their performance robustness in terms of energy consumption as well as thermal comfort. A Post Occupancy evaluation and a behaviour analysis was conducted. The offices of four occupants in each building were monitored and their interaction with the building was studied.
Conclusion: The observed behaviour of the occupants of these two buildings does not match the assumed behaviour used for BP simulations. This can lead to inaccurate energy demand consumption predictions. The behaviour analysis found that occupant behaviour is inconsistent and unpredictable. For example, the measured CO2 levels when windows were opened ranged from 400-2000 ppm; Automatic shading controls were often overridden by 80% of the sample group; Decentralized mechanical ventilation systems were under used because the occupants may not know how to operate it.
The theoretical robustness analysis found that the more robust building had smaller windows, higher thermal mass, automated shading systems and slightly higher room volume had a lower overall heating and cooling energy demand. The parameters that most influence energy demand (amongst those than can be altered by an occupant) are: thermostat set point temperatures, shading, CO2 threshold and number of occupants. It was found that a two degree increase in the heating setpoint could double the heating energy demand. In conclusion, the findings indicate that a more robust building can perform better in combating the regular variations caused by occupants than a sensitive one.