Title

Decision Support System for Predicting Benefits of Left-Turn Lanes at Unsignalized Intersections

Document Type

Journal Article

Publication Date

2007

Subject Area

operations - traffic, infrastructure - bus/tram lane, ridership - commuting, economics - benefits

Keywords

Unsignalized intersections, Uncontrolled junctions, Uncontrolled intersections, Traffic operations, Traffic engineering, Traffic delay, Neural networks, Microsimulation, Left turn lanes, Highway operations, Geometric design, Decision support systems, Benefits, Artificial neural networks, ANNs (Artificial neural networks)

Abstract

Accommodating left turns at unsignalized intersections is one of the most challenging problems in traffic engineering. Over the past 40 years, a few studies developed guidelines to help traffic engineers in deciding when a left-turn lane was warranted for a given situation. Building on these previous attempts, the current study develops a refined decision support system (DSS) for assessing the likely benefits of left-turn lane installations as an aid in deciding whether a left-turn lane is warranted. The developed DSS is designed to predict these likely benefits on the basis of several measures, including delay savings, reductions in percentage of stops, increases in fuel efficiency, and reductions in emissions. The first step in development of the DSS was the use of microscopic simulation to model several real-world unsignalized intersections with different geometric configurations located in different area types. After careful calibration of these models, several scenarios covering a wide range of operational conditions were simulated. The output from these simulation runs was used to train a set of multilayer perceptron neural networks (NNs) and to generalize the results from the models’ runs. The NNs were then incorporated into a DSS that can help an analyst quantify the impacts of a proposed new development as well as estimate the benefits of installations of left-turn lanes. The paper concludes by summarizing some conclusions derived from the study.