Virtual Sensors: A Web-based Real-Time Data Collection Methodology for Transportation Operation Performance Analysis

摘要

Recent advances in mobile networks and increase in the number of GPS-equipped vehicles haveled to an exponential growth in real-time data generation. Over the last decade, a number of onlinemapping or vehicle tracking services haves made their data available for third-party users. Thispaper explores opportunities in utilizing the real-time traffic data provided by online services andintroduces a virtual sensor methodology for collecting, storing and processing large volumes ofnetwork-level data. In order to assess the validity of the collected data using the proposedmethodology, we compare these with data from physical loop detectors and electronic toll tagreaders. Statistical analyses show that there is a strong correlation between the travel timemeasurements from infrastructure based sensors and virtual sensors. We then conduct a travel timereliability analysis using the virtual sensor data methodology and conclude that the results arepromising for future research and implementation.