Spatio-temporal facility utilization analysis from exhaustive WiFi monitoring

Abstract

a b s t r a c tThe optimization of logistics in large building complexes with many resources, such as hospitals, require realistic facility management and planning. Current planning practices rely foremost on manual observations or coarse unverified assumptions and therefore do not properly scale or provide realistic data to inform facility planning. In this paper, we propose analysis methods to extract knowledge from large sets of network collected WiFi traces to better inform facility management and planning in large building complexes. The analysis methods, which build on a rich set of temporal and spatial features, include methods for quantification of area densities, as well as flows between specified locations, buildings or departments, classified according to the feature set. Spatio-temporal visualization tools built on top of these methods enable planners to inspect and explore extracted information to inform facility-planning activities. To evaluate the proposed methods and visualization tools, we present facility utilization analysis results for a large hospital complex covering more than 10 hectares. The evaluation is based on WiFi traces collected in the hospital's WiFi infrastructure over two weeks observing around 18000 different devices recording more than a billion individual WiFi measurements. We highlight the tools' ability to deduce people's presences and movements and how they can provide respective insights into the test-bed hospital by investigating utilization patterns globally as well as selectively, e.g. for different user roles, daytimes, spatial granularities or focus areas.

Task phase recognition for highly mobile workers in large building complexes

“…Hence, the only prerequisite of the positioning method is knowledge of the locations (as well as their emitting power) of the access points, and no calibration is needed -which is in contrast to, e.g., fingerprinting the whole building complex or to methods based on detailed building maps. The method has successfully been utilized in prior studies utilizing indoor positioning systems [13], [15], [14].…”

Understanding the WiFi usage of university students

“…Mobility analysis in hospitals is the main focus in [15] and [16], where movements of people are estimated by leveraging the data coming from a WiFi network deployment composed of 798 access points at Aarhus hospital. Besides characterizing the users mobility, the authors also propose a clustering analysis to classify different types of users (medical staff, visitors, patients, etc.…”

Understanding Crowd Behaviors in a Social Event by Passive WiFi Sensing and Data Mining

“…Existing studies, which make use of WiFi probe requests to understand crowd behaviors in social events or given spaces, can be roughly divided into two categories based on whether the studied spaces are outdoor or indoor. For outdoor environments, related studies are mainly focused on the development of a better WiFi-based passive sensing systems [22], [23], estimation and analysis of crowd size and density [11], [24], analysis of space utilization [25], precise localization of mobile devices [13], and the extraction of trajectories of people [14], [26]. For indoor environments, beyond ordinary visiting and moving statistics, more fine-grained results regarding crowd behaviors have been achieved, such as the social interactions between people studied in [27], [28], because the spatial granularity of collected data is often higher in an indoor space due to bounded space and higher density of existing access points.…”

“…Although existing works have demonstrated the effectiveness of mining WiFi probe requests to understand crowd behaviors in both the outdoor and indoor areas, the thorough analysis of the data collected has not been done with enough focus. Especially, approaches used to analyze the collected data in existing studies mainly include domain knowledge-based processing [9], statistics [29], [24], and visualization [25], [26]. With the rapid development of machine learning algorithms, we believe that more unsupervised learning algorithms should be explored for mining WiFi probe requests.…”