TECHNICAL FEATURE Implementation Challenges Several challenges are associated with the implementation of OCC in real buildings: * Coarse Granularity of Zones: The practice of combining several rooms into a single thermal zone diminishes the savings achievable through zone-level OCC algorithms. In multi-occupant zones, the earliest expected arrival, latest expected arrival and latest expected departure times will be governed by the most extreme occupant in each zone. Case studies demonstrating the benefits of OCCs will make an increasingly compelling case for smaller terminal devices serving 38 ASHRAE JOURNAL ashrae.org M AY 2021 FIGURE 8 Logistic regression models predicting the frequency of thermostat overrides. Probability of a Thermostat Override in The Next 15 Minutes the frequency of thermostat use is minimized between 21.5°C (70.7°F) and 23.5°C (74.3°F). This indicates default setpoints during occupied periods of 21.5°C (70.7°F) in the heating season and 23.5°C (74.3°F) in the cooling season for this building. If preference data were available for each zone, similar models could be developed at the zone level, enabling personalized indoor conditions. Setpoint Decrease Setpoint Increase Setpoint Change 0.01 0.008 0.006 0.004 21.5°C 23.5°C 0.002 0 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Indoor Temperature (°C) smaller thermal zones. * Data Plumbing and Quality Assurance: The integration of disparate data sources within a building automation system may hinder explicit occupancy sensing approaches. Some motion detectors may exist in the lighting automation system, Wi-Fi data will be in the IT network and people-counting cameras will be in the security network. In many buildings, while at least onehttp://simprogroup.com/ http://www.ashrae.org