Dynamic predictive clothing insulation models based on outdoor air and indoor operative temperatures

Clothing affects people's perception of the thermal environment. Two dynamic predictive models of clothing insulation were developed based on 6333 selected observations of the 23,475 available in ASHRAE RP-884 and RP-921 databases. The observations were used to statistically analyze the influence of 20 variables on clothing insulation. The results show that the median clothing insulation is 0.59 do (0.50 do (n = 3384) in summer and 0.69 do (n = 2949) in winter). The median winter clothing insulation value is significantly smaller than the value suggested in the international standards (1.0 do). The California data (n = 2950) shows that occupants dress equally in naturally and mechanically conditioned buildings and all the data has female and male dressing with quite similar clothing insulation levels. Clothing insulation is correlated with outdoor air (r = 0.45) and indoor operative (r = 0.3) temperatures, and relative humidity (r = 0.26). An index to predict the presence of a dress code is developed. Two multivariable linear mixed models were developed. In the first one clothing is a function of outdoor air temperature measured at 6 o'clock, and the second one adds the influence of indoor operative temperature. The models were able to predict 19 and 22% of the total variance, respectively. Climate variables explain only a small part of human clothing behavior; nonetheless, the predictive models allow more precise thermal comfort calculation, energy simulation, HVAC sizing and building operation than previous practice of keeping the clothing insulation values equal to 0.5 in the cooling season and 1 in the heating season. (C) 2012 Elsevier Ltd. All rights reserved.