Dwellings and systems are becoming ever more complex. In achieving high energy performances, the interaction among user, building and systems plays an increasingly crucial role. We studied this interaction in 16 recently renovated low-energy dwellings, to advance the field in terms of methods, and to derive pointers for improved renovation concepts. We used sensors to monitor the indoor environment. And we further developed a method that uses questionnaires, diaries and a so-called 'walk through the house' interview technique to gain insights into occupant behaviour and the reasons for the occupants' behaviours. The results show that the observed behaviour was almost always a normal response to the situation, but often not the behaviour that was expected when designing the renovation concept. A major reason for this was the occupants' efforts to reduce discomfort, which was often the result of the renovation solution itself (for example draught, noise or a warm bedroom). So, we found that behaviour was often a quite normal reaction to the renovation solution. We conclude that the resulting underperformances of renovation concepts are caused by technology centred design. We posit that this can be improved by testing renovation concepts in real world situations.

The effects of pre-set environmental conditions of temperature and lighting on the preferred personal settings, comfort experience and task performance of office workers were investigated in an individually controlled workstation. Twenty subjects performed standardized tasks at a prototype workstation with individually controlled radiant heating and lighting in a climate room. In a repeated measures design, their adjustments to pre-set values were evaluated: low and high radiant heating power, low and high direct illuminance, low and high indirect illuminance. Results showed that preferred personal settings are dependent on the initial, pre-set values of radiant heating power and illuminance. Higher pre-set values result in higher adjusted operative temperatures and higher illuminances on desk, although the differences for heating were too small to show a convincing effect. After adjustment, visual comfort was higher, but it was not dependent of the pre-set values. For thermal comfort no differences were found. Individual task performance was not negatively affected. Providing personal environmental control and the way these concepts and interfaces are designed, play a significant role in user behavior and preferences. The design and control of individually controlled workstations as well as the interaction with the general level of the office environment should be carefully considered in order to obtain maximum comfort and energy efficiency.