Innovative Real-Time Methodology for Detecting Travel Time Outliers on Interstate Highways and Urban Arterials

Abstract

Bluetooth devices are a rich source of travel time data for transportation engineers. Like any other source, however, Bluetooth devices can generate outliers that may bias travel times and corridor speeds. In this study, the proposed, innovative statistical methodology is capable of real-time deployment when travel time and vehicle speed outliers collected from Bluetooth data collection systems are detected. The proposed statistical methodology identifies outliers by using a data point's standard residual in a robust Greenshields model. The efficiency and the feasibility of this statistical methodology are evaluated for both Interstate highways and urban arterial corridors. These roadways are selected because they display a wide range of traffic patterns and outlier generators. The effectiveness of one methodology, especially on a wide range of traffic patterns, would improve the potential for widespread implementation. Four subsets of the collected data are provided within this study to highlight the efficiency and capability of the methodology. The results of using the Shapiro–Wilk statistics show that in both Interstate highway and urban arterial deployment the proposed method is effective at identifying Bluetooth outliers and is capable of working in a real-time environment. A second comparison between the two road types shows that the methodology performs better for Interstate highways than urban arterials, for which the added performance is based on the higher volume of hits and less fluctuation.

Bluetooth as a traffic sensor for stream travel time estimation under Bogazici Bosporus conditions in Turkey

“…Errors may appear in Bluetooth travel time measurements on arterials as a result of signal delay and non-uniform traffic flow [31]. The 10.24 s required to complete the Bluetooth inquiry process may introduce a major source of error and may result in inaccurate travel times, though this error decreases as the distance between Bluetooth stations increases [32,33].…”

“…In recent years, applications of Bluetooth and cell phones have attracted many researchers in estimating travel time. Consequently, various studies have been presented in detecting and addressing outliers [31,34]. Li et al [35] examined traffic data gathered from taxicabs over 24 days by applying the Temporal Outlier Discovery (TOD) concept to detect temporal outliers.…”

Evaluating the Accuracy of Bluetooth-Based Travel Time on Arterial Roads: A Case Study of Perth, Western Australia

“…A larger detecting zone may prevent this problem. Tsubota et al [9] and Van Boxel et al [29] argued that large BMS detection zone is a main cause of inaccurate TT estimates on arterial roads because a large zone means a long staying time, which leads to multiple detections. Although short-range antenna could provide a more accurate TT estimate because discoverable MAC addresses are captured in a location closer to a BMS, a smaller detection zone can cause a lower penetration rate, leading to a decrease of the accuracy of TT estimates [25]; they argued that there should be a trade-off between the size of the detection zone of a BMS and its penetration rate for configuration and coverage of the antennas.…”

Virtual Sensors

“… Highway segments with rest area or toll plaza between two sensors create noisy data (28).  Pedestrian and bicycle traffic may cause false data measurements for vehicular traffic (31).  High speed vehicles between two sensors are sources for outliers (31).…”