Paper Conference

Proceedings of eSim 2022: 12th Conference of IBPSA-Canada


Crowdsourcing Building Occupant Complaints to Aid Fault Detection and Diagnosis

Pedram Nojedehi, Burak Gunay, William O’Brien
Carleton University, Canada

Abstract: Numerous fault detection and diagnosis (FDD) methods for buildings have been developed but many faults are still undetectable. There are also several crowdsourcing applications developed to collect building occupants’ input. However, there is no solid framework enabling building operators to make the most of the collected data, and specifically directly integrating it into FDD. To end this, a framework is proposed to transform occupant feedback into insightful technical contextual information, which supports building operators to make more informed decisions. According to the spread (e.g., multiple on a floor or zone) of problems and their severity computed from sensor readings, local and/or central rulesets are used. Through this framework, the data collected by crowdsourcing occupants’ feedback applications are used more effectively for FDD purposes. The advantages of this framework are demonstrated through occupants’ feedback and measured data extracted from maintenance management and building automation systems, respectively for an academic office building in Ottawa, Canada.
Keywords: crowdsourcing, thermal comfort, occupant feedback, fault detection