Urban Arterial Speed–Flow Equations for Travel Demand Models

This paper describes the effort to improve the speed–flow relationships for urban arterial streets that are contained in the Southern California Association of Governments (SCAG) metropolitan area travel demand model. Intersection traffic counts and floating car runs were made over 4-h-long periods on 1-mi-long sections of eight different arterial streets within the city of Los Angeles. The field data were then filtered to identify which speed measurements were taken during below-capacity conditions and which measurements were made during congested conditions when demand exceeded the capacity of one or more intersections on the arterial. Because the traditional manual intersection traffic count method that was used to gather volumes did not measure queue buildup, and therefore demand, the speed data points obtained during congested conditions were not used in the fitting of speed–flow equations. Several different speed–flow relationships were evaluated against the field data for below-capacity conditions. The most promising speed–flow equations for below-capacity conditions were then evaluated for their ability to predict delays for congested conditions where one or more intersections on the arterial are above capacity. The theoretical delay due to vehicles waiting their turn to clear the bottleneck intersection on the arterial was computed by using classical deterministic queuing theory. Speed–flow equations that underpredicted the delay to clear a congested intersection were rejected. Of the speed–flow equations tested, the Akcelik equation performed the best for above-capacity situations and performed as well as other possible equations for below-capacity conditions.