Predicting Lane Change Intensity within Urban Interchange Influence Areas (IIA)

Predicting Lane Change Intensity within Urban Interchange Influence Areas (IIA)

This research is intended to assist NCDOT in improving mobility and safety performance at urban interchange influence areas (IIA's) in North Carolina, to remedy the excessive levels of discretionary lane changes occurring at those locations. The research will predict how driver lane changing behavior is impacted by local traffic, by control and by site conditions. A second objective will be to ascertain whether changes in signing, markings or other traveler information near the IIA can induce fewer discretionary lane changes and thus reduce unnecessary traffic turbulence near interchanges. Currently NCDOT has no means to track lane changing behavior. This research will take advantage of an existing (and continuously expanding) NC State high resolution second by second trip database which uses an in-vehicle OBD-II unit (called i2d) in the Triangle Region. Supplemented with controlled experiments and other data sources the research will produce predictions of lane changing behavior at the vehicle scale, based on present IIA geometrics, the prevailing traffic states and any implemented lane-discipline-inducing treatments. To achieve these objectives, we are proposing to develop a statistical model to predict lane change intensity in urban interchange influence areas. Lane changes will be characterized as mandatory or discretionary. Our i2d data can inform us which type of lane change it is, based on knowledge of the trip origin and destination. Initially, however we will assume that all lane changes are discretionary except when it is known that the vehicle is either entering or exiting the IAA. Thus our predictor variable will be the expected discretionary lane change intensity per vehicle mile in the IIA. The explanatory variables will include both road design variables at the IIA as well as traffic and environmental conditions. On the geometric side we will consider the effect of the number of lanes, the spacing between ramps, the length of acceleration/ deceleration lanes, the presence of a lane drop and the distance to the nearest upstream and downstream ramps, along with any significant grades on the mainline and ramp roadways. On the traffic stream side, the current best source of data would be existing NCDOT RTMS devices or portable counting stations that can provide lane by lane volume and speed and could also serve to validate the lane change model. We will also explore the use of the NGSIM datasets with detailed trajectories for all vehicles on several freeway facilities. In addition the research team will extract from i-PEMS sub-TMC or link speed data that have occurred in the same time interval in which the lane changes were executed. This will help our understanding of the effect of local congestion levels on the intensity of lane changes. We will discard any data associated with the presence of incidents in the IIA as these will obviously impact the lane selection behavior of drivers. Upon completion of model calibration, we intend to validate the model at IIA sites not used in the model development process. The final set of influencing variables will be the type of control/treatment at the IIA, which could include the presence of ramp metering, possible additional signing or marking to encourage certain lane shifts or an optimal lane distribution prior to entering the IIA. Depending on the schedule of NCDOT actually implementing any such intervention, the team will compare its prediction of lane change intensity in the baseline case with the observed post-treatment lane change intensity (measured via video, RTMS or i2d) and make an assessment of the effectiveness of the proposed intervention.


  • English


  • Status: Active
  • Funding: $233447.00
  • Contract Numbers:


  • Sponsor Organizations:

    North Carolina Department of Transportation

    Research and Development
    1549 Mail Service Center
    Raleigh, NC  United States  27699-1549
  • Managing Organizations:

    North Carolina Department of Transportation

    Research and Development
    1549 Mail Service Center
    Raleigh, NC  United States  27699-1549
  • Project Managers:

    Penny, Lisa

  • Performing Organizations:

    North Carolina State University

    Institute for Transportation Research and Education
    Raleigh, North Carolina  United States  27695-8601
  • Principal Investigators:

    Rouphail, Nagui

  • Start Date: 20170801
  • Expected Completion Date: 20201031
  • Actual Completion Date: 20201031
  • Source Data: 2019-29

Subject/Index Terms

Filing Info

  • Accession Number: 01746004
  • Record Type: Research project
  • Source Agency: North Carolina Department of Transportation
  • Contract Numbers: FHWA/NC/2019-29
  • Files: RIP, STATEDOT
  • Created Date: Jul 23 2020 1:07PM