Analyzing WiFi connection counts in commercial/institutional buildings to estimate/predict occupancy patterns for optimizing buildings’ systems operation

Accurate occupancy information can help in optimizing the operation of building systems. To obtain this information, previous studies suggested using WiFi connection counts due to their strong correlation with occupancy counts. However, validating this correlation and investigating its variation have remained limited due to challenges regarding collecting ground-truth data. Moreover, the difficulty of integrating real-time WiFi traffic data in building automation systems hinders wide-scale deployment of this approach. This research addressed these gaps by proposing a methodology, including two modules focused on developing frameworks, for (i) validating the correlation between WiFi connection counts and actual building occupancy counts by using continuous ground-truth data collected from camera-based occupancy counters; and (ii) extracting occupancy indicators from WiFi connection count data which can then be used for updating control sequences. The proposed research was implemented in two institutional buildings to validate the proposed methods in two case studies. Results of the first case study showed Hour of the day, Day of the week, as well as occupancy level, affect the correlation between WiFi and occupancy counts. Furthermore, the proposed models could successfully estimate real-time occupancy counts and predict day-ahead occupancy counts with an average accuracy (R2) of 0.97 and 0.87, respectively. Moreover, the results of the second case study revealed that the proposed models could successfully predict weekly building occupancy patterns, with an average accuracy (RD2) of 0.90. Furthermore, the analysis identified peak occupancy timing, as well as arrival/departure times variations between different zones. These findings provided a proof-of-concept for the proposed methodology and demonstrated the potential of using WiFi connection count for estimating/forecasting occupancy counts at a large scale and extracting actionable information to optimize buildings’ system operation based on buildings’ unique occupancy patterns