Residential and tertiary buildings in France consume 46% of the global final energy (2017) and are responsible for 30% of all energy related carbon dioxide emissions (2016) (CGDD, 2019). However, even if this represents an important potential of energy consumption reduction, this potential is not totally exploited and the building sector remains a silent consumer. With the development of Internet of Things and the integration of Big Data and sensors in many fields of daily life, there is an important need to develop intelligent approaches in order to create sustainable environments where less energy is consumed while responding to comfort needs. Hence, more and more importance is nowadays given to the industrial and academic issue of using the potential of digital solutions in order to make buildings more energy efficient from design to operation phase. The development and the use of simulation tools aimed to help designers predict the energy consumption and performance of the building. However, recent researches and experiences have shown a considerable gap between predicted energy consumptions and real ones, and occupant behaviour is now essentially recognized as the main responsible of this disparity (Darakdjian et al., 2018) (Vorger, 2015). The interaction between users and buildings, occupant presence and the adaptive actions they may have to maintain their comfort can make energy consumption five times higher than what was predicted (IPEEC Building Energy Efficiency Taskgroup, 2019). Thus, occupant behaviour must be well understood and modelled in a more accurate way in order to identify its impact on energy consumption. This will help identifying the adequate energy saving recommendations and turn occupants from consumers to active users of energy in buildings.