01520133

Component

Collecting travel times through Bluetooth detectors is becoming increasingly popular due to its advantages of: (1) measuring travel time directly; (2) anonymous detection; (3) weatherproof; and (4) cost-effectiveness. However, as with other AVI (automatic vehicle identification) technologies, estimating average travel time based on data collected by Bluetooth detectors requires some form of filter to identify and remove the outliers. Existing real-time travel time outlier filtering algorithms are data driven models which rely on measures of variance in the data obtained in the recent past to discriminate between outliers and valid observations at the current time. However, these models typically perform poorly when travel times change rapidly, such as when an incident occurs. In this paper, the authors propose a traffic flow filtering model which can be applied as an enhancement to existing data-driven outlier detection algorithms as a mechanism to improve outlier detection performance. The authors describe the proposed model and then demonstrate its performance by incorporating the model into two existing data driven outlier detection algorithms and applying the enhanced algorithms to a dataset of freeway travel times collected by Bluetooth detectors. The application results indicate that the proposed method is able to solve the problem of tracking sudden changes in travel times and enhance the performance of the data-driven outlier detection algorithms.

This paper was sponsored by TRB committee ABJ30(3) Travel Time, Speed and Reliability.

01503729

14-0652

English

18p

2014

Transportation Research Board 93rd Annual Meeting

Digital/other

Figures; References; Tables

Algorithms; Bluetooth technology; Data files; Traffic flow; Traffic flow theory; Travel time

TransGuide (Program : San Antonio, Texas)

Data and Information Technology; Highways; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2014 Paper #14-0652

TRIS, TRB, ATRI

Jan 27 2014 2:18PM