Predictive model of indoor temperature from ambient levels

This study aimed to estimate the relationship between atmospheric and indoor temperature in typical residences, and further to establish the predictive model for indoor temperature based on outdoor information collected from ambient monitoring station. Thirty homes, selected from 11 cities throughout Taiwan, were recruited for indoor environmental monitoring. Mixed effects model was constructed. Preliminary data suggest as much as 44% of the indoor temperature variance could be explained by the model, depending on the region of interest, and the prediction was more precise during the period of December to April. The model is expected to provide a better estimate of individuals' thermal exposure levels in ecological studies, and to be considered as the refrain the susceptible group from the risk of sudden of temperature changes.