Strategies for thermal comfort improvement and energy savings in existing office buildings using occupant feedback

Log it While it's Hot

“…In shared work environments, participatory sensing mechanisms designed to assist smart thermostats in calculating peoples' thermal comfort thus far amount to an assumption that users are willing and able to engage with the system at regular intervals over extended time periods [10,18,23,25,27]. These assumptions contradict social research which finds everyday life to be messy and unpredictable [31] and that engagement with personal informatics tends to wane over time [15,29,33].…”

“…Recent participatory sensing approaches for thermal comfort in offices include affording users the ability to influence the set-point of their offices' thermostat via a mobile [10,18,25] or PC application [27]. In each of these studies, users select their thermal comfort level between 'Cold' to 'Hot' on a scale.…”

“…Shared work environments such as offices represents a relatively under-explored context for smart thermostat research in HCI, with thermal comfort sensing and modelling work thus far constrained to the fields of artificial intelligence and machine learning [3,10,13,18,21,23,25,26,27]. Shared work environments provide a unique set of challenges for smart thermostats as distinct from the home.…”

“…This does not allow the system to adapt for a change of occupancy or user preferences over time. On the other hand participatory sensing systems such as [10,23,25,27] all rely on regular user input over longer timescales, which may not be realistic given the tendency for user engagement in situated technology to evolve over time to less frequent interactions, or diminish all together [14,15,33].…”

“…Various thermal comfort models and participatory sensing systems for office-based smart thermostats exist [3,10,13,18,23,25,26,27], however the validation and practical implementation of these models does not always extend to in-the-wild user trials [10,18,25,27].…”