A Bottom-up Approach to Enhance Occupants' Actions Regarding Satisfaction in Buildings

In the context of energy management systems, building energy management is an interesting and important sector as it plays an important role in global energy consumption. Moreover, many studies have shown that occupants' actions considerably impact building energy, but buildings are too complex for occupants to identify good actions. In this approach, there were many approaches using physical models to solve the problem, but there are many limitations to them. This paper proposes a new bottom-up approach in energy management systems without physical models to recommend actions on doors/windows to occupants and enhance their satisfaction without an extra cost, using their historical actions. An office at Grenoble Institute of Technology, France, with a historical dataset in 2015–2016 comprising both contexts and actions, is used to evaluate the approach. The results showed that the approach could suggest better actions for 88% of testing days and improve by 4%-18% global satisfaction of occupants in the testbed.