Improving Adaptive/Responsive Signal Control Performance: Implications of Non-Invasive Detection and Legacy Timing Practices

Project Details
STATUS

Completed

START DATE

10/01/14

END DATE

02/28/17

RESEARCH CENTERS InTrans, CTRE
SPONSORS

Oregon Department of Transportation

PARTNERS

sub to Northern Arizona University

Researchers
Principal Investigator
Anuj Sharma

Co-Director, REACTOR

About the research

This project collected and analyzed event based vehicle detection data from multiple technologies at four different sites across Oregon to provide guidance for deployment of non-invasive detection for use in adaptive control, as well as develop a true life cycle cost comparison of various detection sources. Data during fair weather from non-invasive detection zones co-located with inductive loops were compared under based upon typical metrics used for driving adaptive algorithms, such as activations, occupancy, and time to gap out. From this analysis, it was recommended that non-invasive detection sources should be used with caution for developing data for adaptive control, as the inherent nature of their operation differs from inductive loops. From a life cycle cost standpoint, non-invasive units were shown to be much more expensive than inductive detection over the typical lifespan, but modest efficiency and safety improvements possible due to expanded features of the non-invasive units may overcome this cost deficit. Lastly, due to performance issues encountered with over 50% of the inductive loop detectors within this study, it is highly recommended to implement some type of continual detector health monitoring program regardless of technology selected for deployment.

Subcontractor to Northern Arizona University on this project.

TOP