ASHRAE Likelihood of Dissatisfaction: A new right-here and right-now thermal comfort index for assessing the Likelihood of dissatisfaction according to the ASHRAE adaptive comfort model

ABSTR A C T The assessment of local and short-term thermal discomfort in buildings has been widely investigated, and different metrics are available in the literature to predict the likelihood of dissatisfied people. These metrics are named right-here and right-now discomfort indexes and constitute the basis for evalu-ating long-term thermal comfort conditions in buildings. Well-known examples are the Predicted Percentage of Dissatisfied (PPD) part of the Fanger comfort model included in the ISO standard 7730 and the Overheating risk index (NaOR), built upon the EN adaptive thermal comfort model. This study proposes a new index for use with the ASHRAE adaptive thermal comfort model to fill a gap in the liter-ature and standard. It is called the ASHRAE Likelihood of Dissatisfaction (ALD) and is obtained from a logistic regression of the right-here and right-now thermal comfort field data contained in the 1990s ASHRAE RP-884 database. The recent release of another, more extensive database of thermal comfort field studies, the ASHRAE Global Thermal Comfort Database II, provides an opportunity to validate ALD with an independent dataset and assess its generalisability. The successful external validation of ALD and its agreement with NaOR give support to the reliability of the novel right-here and right-now index and open to the possibility to use it for assessing short-term thermal comfort conditions in buildings, cal-culating long-term thermal comfort indices based on the ASHRAE adaptive model, optimising both the design of new buildings and renovations and for assessing the operational thermal comfort performance of existing buildings. (c) 2021 Elsevier B.V. All rights reserved.

Automated summary

This study proposes a new index ALD for assessing short-term thermal comfort conditions in buildings, which has been successfully externally validated and shown to be reliable and generalizable.