In part 2 (chapters 5–7), the factors influencing on the reference levels for thermal comfort were elaborated and a method to quantify the relative importance of those factors were established. The first analysis of the data obtained from the measurement in the student dormitory showed the significance of individual factors in relation to differences in the behaviour. Examples for such individual factors are a preference for air-conditioned spaces and thermal background, which refers to the climate a person spent the majority of their first 10 years of life. In the next step, the influence of external factors such as temperature and humidity as well as various individual factors were evaluated according to their significance in describing the observed behaviour by the mean of multivariate logistic regression analyses. As a result, statistical models of occupant‐behaviour prediction including an exponentially weighted running mean of the outdoor temperature during the night and 17 individual factors such as above mentioned preference and thermal background were presented. The factors included in these models were then judged by introducing an importance factor. It was found that, on the one hand, the running mean of the temperature has a similar impact to the individual factors in summer, while, on the other hand, the latter have a much higher influence in winter than in summer.
Part 3 (chapters 8–9) analyses whether and to what degree long‐term or short‐term experiences can influence the set‐up or a change of the reference levels. Firstly, the effects of long‐term experiences such as cultural influences are shown by comparing the stated behaviour with respect to sleeping conditions of those participating in the Internet‐based survey. This revealed, on the one hand that the behaviour is influenced by the type of building as well as the surrounding, and on the other hand, it was shown by means of a detailed analysis of Germans living in Japan compared to those living in Germany that behaviour can be influenced by such external factors. Second, the influences of knowledge transfer methods on the reference levels, which can be regarded as short‐term experiences, were evaluated. It was shown that the knowledge transfer methods can motivate the recipients to try out strategies not practised before, but there is a huge discrepancy between the stated intention immediately after the knowledge transfer and the actions performed in reality. With respect to the usage of AC‐unit, the effect of knowledge transfer processes such as workshops or information papers could be quantified to lead to a reduction of up to 16% in the case of the workshop, but no significant change was detected in response to the information paper.
Part 4 (chapter 10) describes the effect of differences in the reference levels compared to building envelope improvements on the exergy consumption for heating and cooling. Based on a steady‐state exergy calculation, the influence of the occupant behaviour was highly significant (more than 90% decrease of exergy consumption) when the difference between indoor and outdoor air temperature is small, which is the case for long periods in those regions with moderate temperatures during summer and/or winter. On the other hand, the building envelope improvements have a much higher influence wherever the difference in temperature
between indoors and outdoors are larger. Nevertheless, both measures combined lead to a reduction of up to 95% and should be the final goal.