Accurate assessment of thermal comfort requires comprehensive analysis of the environmental effects contributing to the heat transfer to and from the human body. A common comfort evaluation approach (e.g. PMV/PPD, Equivalent Temperature) is to find a direct correlation of comfort to environmental conditions (e.g. air temperature, relative humidity, clothing), thus implicitly accounting for the relationship between physiological response and thermal comfort. An alternate approach (e.g. Berkeley Comfort Model, Fiala's DTS) is to explicitly correlate comfort to basic physiological response (e.g. skin and core temperature), thereby separating the thermal analysis portion of the problem from the more subjective comfort analysis portion.
While it has been shown that comparable results can be obtained between environment-based comfort metrics and physiology-based comfort metrics, the latter should be employed for optimal prediction accuracy. This is largely due to the fact that environment-based models, having been developed under steady-state conditions, do not account for the initial thermo-physiological state of the human body, or its transient response to rapidly changing thermal environments. A series of numerical simulations have been developed to illustrate the merits of using physiology-based comfort metrics. The intent of this paper was not to validate or recommend a particular comfort model, but rather to demonstrate the differences in thermal comfort predictions that can be obtained between environment-based metrics and physiology-based metrics.