The findings of an FHWA project entitled "levels of service in shared, permissive left-turn lane groups at signalized intersections"are described. The project objective was to evaluate and improve the 1985 Highway Capacity Manual (HCM) model for estimating the impact of shared-permissive left-turn movements on lane group saturation flow. The study was based on a nationwide data collection effort that yielded 1,492 15-min data samples taken for 45 intersection approaches (at 25 intersections) in four regional areas. Field measurements included prevailing and ideal saturation flow rates for each period, critical portions of the green phase as defined in the 1985 HCM, subject and opposing flow rates, lost times, and other critical information. Because of the lack of off-peak-period queues sufficient to yield saturation flow rate measurements, only 22% of the data was used in model evaluations and calibrations. The resulting analysis documents flaws in the 1985 HCM model for the impact of shared, permissive left turns on saturation flow rates, and recommends simplified models that could be used in conjunction with the general analysis procedure of the 1985 HCM. Unique characteristics of single-lane approaches opposed by single-lane approaches are identified, and revised procedures for the analysis both of single-lane and multilane approaches involving shared-permissive left turns are recommended.
Abstract