@article{ref1,
title="Simulation-neural network model for evaluating dilemma zone problems at high-speed signalized intersections",
journal="Transportation research record",
year="1994",
author="Huang, Peter and Pant, Prahlad D.",
volume="1456",
number="",
pages="34-42",
abstract="The commonly used traffic control devices or techniques used to address dilemma zone problems at low-volume, isolated, high-speed signalized intersections include detector configurations, advance warning signs with or without flashers, and timings of change intervals or green extensions. The levels of effectiveness of these devices or techniques are sensitive to roadway geometric, speed distribution, and traffic volume. The measures of effectiveness of traffic control at a high-speed intersection approach are expressed as the (a) probability of being caught in a dilemma zone, (b) speed of a vehicle in different segments of the intersection approach, and (c) vehicle conflict rate. A simulation model has been developed to dynamically represent each element of the traffic control system such as roadway geometric, traffic control devices (advance warning signs, flashers, detectors, and signals), and vehicular movements. Artificial neural networks have been developed to estimate vehicular speeds in different segments of the intersection approach in response to different advance warning signs, flashers, and signal indications. The simulation model has been integrated with the neural networks to provide better accuracy of the simulation. A case study showed that the results of the simulation-neural network model compared well with the field data collected at several low-volume, high-speed signalized intersections in Ohio. The model can be used as a non-accident-based safety evaluation procedure for high-speed signalized intersections.<p /><p>Language: en</p>",
language="en",
issn="0361-1981",
doi="",
url="http://dx.doi.org/"
}