Evaluation of the 2010 Highway Capacity Manual Urban Street Free-Flow Speed Prediction Model

The performance of the new 2010 Highway Capacity Manual (HCM 2010) procedure that predicts free-flow speed by using posted speed limit and eight additional variables was evaluated. The eight additional variables were the proportion of segment length with restrictive median, the proportion of segment with curb on the right-hand side, the number of access point approaches on the right side in the subject direction of travel, the number of access point approaches on the right side in the opposing direction of travel, the segment length, the width of the signalized intersection, the number of through lanes, and the distance between intersections. Multiday speed data from 104 Florida urban street segments with a speed limit ranging from 25 to 55 mph were used in the study. Field mean free-flow speed was determined for each analysis segment as well as the geometric and traffic attributes required by HCM 2010 to predict free-flow speed. Statistical and graphical analyses showed that the HCM 2010 speed prediction methodology had a low coefficient of determination and underpredicted free-flow speed when the segment posted speed limit was 35 mph or higher. The results further showed that the use of free-flow speed (predicted by the HCM 2010 methodology) in determining arterial level of service produced lower levels of service for 10 of the 20 segments analyzed. In addition, calibration of the HCM 2010 model coefficients showed that speed prediction by this model could be improved if the speed limit coefficient were doubled and that the influence of cross section and access density factors substantially decreased.