Prediction of indoor air temperature for assessment of people's thermal stress

Climate change is expected to increase the frequency and intensity of extreme weather events. Individualized and timely advice on how to cope with thermal stress is therefore needed to encourage protective strategies and reduce morbidity and even mortality among vulnerable populations. Such advice can be based on integration of human thermal models, weather forecasts and individual user characteristics. The current study focused on development of an algorithm to predict indoor air temperature and assess indoor thermal exposure with incomplete knowledge of the actual thermal conditions. The algorithm provides discrete predictions of temperature through a decision tree classification with six simple building descriptors and three parameters harvested from weather forecast services. The data used to train and test the algorithm was obtained from field measurements in seven Danish households and from building simulations considering three different climate regions ranging from temperate to hot and humid. The approach was able to correctly predict approximately 68% of the most frequent temperature levels. The findings suggest that it is possible to develop a simple method that predicts indoor air temperature with reasonable accuracy.