Modeling of Subjective Responses to Indoor Air Quality and Thermal Conditions in Office Buildings

Abstract

Air conditioning is used to control air quality and thermal environment and has a significant impact on perceived indoor air quality, thermal conditions, health symptoms, sick-building syndrome (SBS), task performance, and productivity. Building occupants provide a rich source of information about perceived indoor air. Could this information be used to teach and adapt the heating, ventilating, and air-conditioning system to improve satisfaction with indoor air is the research question behind this paper's study. A Bayesian probability model was constructed for perceived air quality and perceived thermal conditions (temperature, draught, and air humidity). Indoor air-condition measurements (temperature, air velocity, velocity fluctuation, and humidity) and human-related background factors, obtained by questionnaire, formed the input data for the model. In the developed model, the human-related well-being factors influenced two threshold values, which describe the indoor air conditions where the occupant changes her/his response on a three-level scale. The three-level scales were as follows: for indoor air quality, fresh/neutral/stuffy; for temperature, too cold/agreeable/too warm; for draught, no draught/weak draught/much draught; and, for humidity too low/comfortable/too high. The data were gathered from five office buildings by Internet questionnaire surveys and by parallel measurements of indoor air quality and thermal conditions. The measured temperature correlated strongly with subjective satisfaction of air quality and temperature. This result confirms the previously reported impact of temperature on perceived air quality. The measured temperature and relative humidity were related to temperature perception. From the human-related background factors, temporary mental well being had the strongest effect on the threshold values of the perceived temperature and air quality. Occupants suffering from decreased temporary mental well being complained of excessive air warmth and nonfreshness of air more readily than others. In our material, the maximum satisfaction of 80% in perceived temperature was predicted by the model to be obtained at 22.5°C. At this temperature, 20% of occupants were predicted to feel too cold, 30% would indicate fresh air quality, about 65% neutral air quality, and 5% bad air quality. The optimum air temperature is in agreement with the guidelines and standards given for temperature. Information gathered by the questionnaire surveys can be used in the model to produce response probabilities for indoor air, which could help to adjust modern ventilation systems, according to the building's occupants, to improve satisfaction with indoor air. However, the possibility for personalized control is still needed to ensure full human comfort.