Use of Data from Point Detectors and Automatic Vehicle Identification to Compare Instantaneous and Experienced Travel Times

    Authors

    Y. Xiao; S. F. Qom; M. Hadi;H. Al-Deek

    Comments

    Authors: contact us about adding a copy of your work at STARS@ucf.edu

    Abbreviated Journal Title

    Transp. Res. Record

    Keywords

    PREDICTION; Engineering, Civil; Transportation; Transportation Science & Technology

    Abstract

    Most traffic management centers use detector data to estimate instantaneous travel times. Interest is increasing in using automatic vehicle identification (AVI) readers to provide travel time measurements as well as in using predictive modeling of travel times. This study aimed to examine the differences between travel time estimation that was based on detector data versus those based on AVI data. In addition, the study compared instantaneous travel time estimates with experienced travel time estimates to determine the adequacy of disseminated instantaneous travel time information and, thus, the potential benefits of using predictive travel time modeling. The results showed that, for uncongested conditions, the difference between point detector- and AVI-based estimates and between instantaneous and experienced travel times was insignificant. During congested traffic conditions, the difference between estimates based on detector data and those based on AVI data (Bluetooth and electronic toll tag reader data) was about 6% to 17%. In addition, a difference of 10% to 20% existed between instantaneous and experienced travel times estimated from both the detector data and AVI data; this difference depended on the tested scenarios. The values of the differences between instantaneous and experienced travel times from both types of data sources are expected to be affected by the queue-forming and -dissipating speeds, route length, and the location of the congestion.

    Journal Title

    Transportation Research Record

    Issue/Number

    2470

    Publication Date

    1-1-2014

    Document Type

    Article

    Language

    English

    First Page

    95

    Last Page

    104

    WOS Identifier

    WOS:000351880200011

    ISSN

    0361-1981

    Share

    COinS