Report # UNL: SPR-P1 (11) M307 Final Report SPR-P1 (11) M307
This report presents the results of Nebraska Department of Roads (NDOR) research project SPR-P1 (11) M307, which evaluated the traffic operations and safety effects of 5 mph and 10 mph speed limit reductions in the vicinity of high-speed, signalized intersections with advance warning flashers (AWF). The methodology involved two studies: 1) field study of the impact of speed limit reduction at seven high-speed intersections, 2) crash analysis using the 10-year history from 28 high-speed intersections. In the field study, traffic operational effects of the reduced speed limits were analyzed for seven high-speed, signalized intersections with AWF, using the Quantile regression model and Seemingly Unrelated Regression Estimation (SURE). The Quantile regression models indicated that reduction of speed limit from 60 mph to 55 mph did not lead to any statistically significant reduction in the 15th, 50th, or 85th percentiles. It was found that a speed limit reduction from 65 mph to 55 mph led to a 4.6 mph reduction in 85th percentile speed. Also, the speed dispersion based on an inter-percentile range between 15th and 85th percentiles was reduced by 1.4 mph in the vicinity of the intersection. SURE was used to estimate the mean and standard deviation of grouped average speeds simultaneously. The SURE model was chosen to account for any potential correlations between the mean and standard deviation of speed. It was found that a speed limit reduction of 10 mph, when the upstream speed limit was 65 mph, reduced the mean speed of vehicles by 3.8 mph, or by six percent. This result was statistically significant at the 95% percent level of confidence. It was also found that reducing the speed limit by 5 mph when the speed limit was 60 mph did not produce any statistically significant reduction in mean speed. In addition, the standard deviation of the speeds downstream of the speed limit sign was not statistically significantly different from the upstream for either 10 mph or 5 mph reductions. In the second study, a crash analysis based on 56 approaches from 28 intersections was performed to study the safety effects of speed limit reductions. The dataset included four approaches of 10 mph reduction from 65 mph to 55 mph, seven approaches of 5 mph reduction from 60 mph to 55 mph, two approaches of 5 mph reduction from 55 mph to 50 mph, and 43 approaches with no limit reduction (i.e., the control group). The 10 mph speed reduction from 65 to 55 mph was found to reduce, on average, 0.4 crashes per approach per year with a 90% level of confidence. Also, the studied approaches with 10 mph reduction were found to have a lower probability of possible injury crashes and a higher probability of possible damage crashes with a 90% level of confidence. The 5 mph reductions from 60 mph to 55 mph and from 55 mph to 50 mph were found to reduce 0.6 crashes per approach per year at a 95% significance level. It was also found that lower speed limits in the vicinity of signalized intersections reduced the probability of fatal and injury crashes. The conclusions of this study, however, are limited by the low number of intersections with speed limit reductions. For example, only two intersections with 10 mph reduction were available for the study, where the speed limit was reduced from 65 mph to 55 mph. Based on this dataset, for a highway with speed limit at 65 mph, the reduction to 55 mph at intersections with AWF has been found to reduce mean speed and crash frequency, and alleviate possible crashes in comparison to the intersections with only AWF. It is recommended that future research include other speed limit combinations, such as a 5 mph reduction from 65 mph to 60 mph, and utilize larger datasets to provide better generalizability and transferability of results. A before and-after study could also provide partially controlled conditions to isolate the impacts of speed limit reduction.
Available at: http://works.bepress.com/anuj_sharma1/16/