Developing new comfort band for adaptive model in Japanese residential building

• The monthly mean comfort temperature was ranging from 19 to 27 °C. • Comfort band for adaptive model was analysed by quadratic and probit analysis. • Comfort band was found to be ±2.0 °C and ±1.5 °C for 80 % and 90 % limits. • 79% of the comfort temperature is within 80% limits of the comfort band. An environmentally sustainable society can be created by reducing building energy usage and achieving carbon neutrality. Adaptive thermal comfort models can aid with the establishment of national guidelines and standards on building indoor thermal conditions. However, many researchers focus on developing adaptive models without concerning the comfort band. A field measurement was conducted for 2 years in 64 selected houses with measured indoor environments and collected 32,988 votes from the thermal comfort survey. The findings indicate that majority of the residents expressed satisfaction with their thermal environments. The monthly mean comfort temperature ( T c ) by Griffiths’ method ranged from 19 to 27 °C. In developing the comfort band for the adaptive model, the data from indoor temperature ( T in ) and the difference between indoor and comfort temperature ( ΔT = T in -T c ) were analysed using quadratic regression and probit analysis. A wider width of the comfort band was found when using the data of T in . By using the data of ΔT , both quadratic and probit analysis produced a similar comfort band. Thus, an appropriate comfort band was found as ±1.5 °C and ±2.0 °C for 90 % and 80 % limits of the adaptive model. Developing these methods for comfort bands of the adaptive models is crucial when creating a reliable standard and guidelines for building design or indoor environmental quality assessment.