A data-driven approach to defining acceptable temperature ranges in buildings

Abstract

Current thermal comfort standards use Predicted Mean Vote (PMV) classes as the compliance criteria despite previous critiques. The implicit assumption is that a narrower PMV range ensures higher thermal acceptability among building occupants. However, our analysis of a global database of thermal comfort field studies demonstrates that PMV classes are not appropriate design compliance criteria, and reinforces the need for a new and robust approach to thermal comfort compliance assessment. We compared two statistical methods to derive acceptable temperature ranges from occupant responses applied one to the ASHRAE Global Thermal Comfort Database II. Derived acceptable temperature ranges in real buildings (7.4K-12.2 K) using this new method are wider than the current standards mandate (2 K-6K). Our findings support the call for a relaxation of suggested temperature ranges in thermal comfort standards so as to minimize unnecessary space conditioning. The proposed data-driven statistical methods to determine temperature design compliance criteria are viewed as an important step forward in the age of continuous and pervasive monitoring and the associated large databases of building comfort measurements.