Yun, SeoyeonLicina, Dusan2023-07-312023-07-312023-07-312023-06-0610.1016/j.buildenv.2023.110459https://infoscience.epfl.ch/handle/20.500.14299/199408WOS:001016824600001Modern health and productivity concerns related to air pollutant exposure in buildings have sparked the need for occupant-centric monitoring and ventilation control. The existing personal exposure monitoring is often restricted to stationary air quality sensors and static occupancy. This study aims to identify optimal stationary sensor placement that best represents exposure to CO2, PM2.5, and PM10 under static and dynamic office occu-pancies. A total of 48 controlled chamber experiments were executed in four office layouts with variation of occupant numbers (2, 4, 6 or 8), activities (sitting/standing and static/dynamic), ventilation strategies (mixing/ displacement) and air change rates (0.5-0.7 h(-1), 2.4-2.6 h(-1), and 3.8-4.2 h(-1)). The breathing zone concen-tration of a reference occupant was monitored with concurrent measurements at seven stationary locations: front edge of the desk, sides of two desks, two sidewalls, and two exhaust vents. The proximity of sensors to the reference occupant and ventilation rate/strategy were important determinants of personal exposure detection. Regression analyses showed that the wall-and desk-mounted CO2 sensors near the occupant (<1 m) best captured CO2 exposure under dynamic-standing activities (R-2 similar to 0.4). The wall immediately behind the seated occupant and the ceiling-mounted exhaust near the standing occupant (<1-1.5 m) were the best sensor place-ments for capturing exposure to particles (R-2=0.8-0.9). Separating static from dynamic occupancy activities resulted in improved exposure prediction by 1.4-6.1x . This study is a step towards provision of practical guidelines on stationary air quality sensor placement indoors with the consideration of dynamic occupancy profiles.Construction & Building TechnologyEngineering, EnvironmentalEngineering, CivilEngineeringiaq sensor placementpersonal exposure detectiondynamic activitiesventilation typelinear regression modelindoor air-qualityparticulate matterventilationemissionsratesparticlestext::journal::journal article::research article