No abstract

This research effort explored the factors associated with higher driver yielding rates at pedestrian crossings with traffic control signal (TCS), pedestrian hybrid beacon (PHB), and rectangular rapid-flashing beacon (RRFB) treatments in Texas. The percentages of drivers yielding to a staged pedestrian were collected at 7 TCS sites, 22 RRFB sites, and 32 PHB sites. Overall, TCSs in Texas had the highest driver yielding rates, with an average of 98%. The average driver yielding rate for RRFBs in Texas was 86%, whereas the average for PHBs was 89%. The RRFB results for Texas were slightly higher than those found in other studies, perhaps because all RRFB sites included in this analysis had “School Crossing” signs and were located near a school. The number of devices within a city may have an impact on driver yielding. Those cities with a greater number of a particular device (i.e., Austin, Texas, for the PHBs and Garland, Texas, for the RRFBs) had higher driver yielding rates than cities where the device was used at only a few crossings. A comparison of the number of days since installation revealed statistically significantly higher driver yielding rates for those devices that had been installed longer. For PHBs, the results support the use of PHBs on roadways with multiple lanes or a wide crossing. For RRFBs, a lower rate of compliance was observed for longer crossing distances and indicated that for a certain crossing distance width, a device other than the RRFB should be considered.

No abstract

The Oregon Department of Transportation developed segment arterial safety performance functions (SPFs) to help quantify the safety performance of driveways located on state urban and rural arterial highways. The research team determined that the crash reporting indicating that a driveway may have been involved in the crash was not a dependable variable, so the team developed SPFs for all non-intersection-related arterial crashes (many were likely the result of vehicle interactions at driveway locations). The information included in this paper reviews the subsequent validation effort and highlights innovative techniques used for the analysis. A common validation approach is assessing model performance for spatial transferability. For this effort, however, the authors evaluated spatial transferability, spatial–temporal transferability, and individual coefficient stability and significance. These procedures were highlighted and applied to an example urban model. The model performed well with the spatial transferability resulting in statistically equivalent values. The spatial–temporal transferability provided similar values but was not statistically equivalent at the 95% level, and all but one of the model variables were determined to be statistically significant.

No abstract

Several methods, including one that supplements signage with beacons or embedded LEDs, have been used to emphasize the presence of a pedestrian crossing. A device that has received national attention is the rectangular rapid-flashing beacon (RRFB). A question that is being asked about the device is whether the beacons need to be rectangular. Could they be circular? For a comparison of the two devices, RRFBs and circular rapid-flashing beacons (CRFBs) were installed at 12 sites in four cities (i.e., Milwaukee, Wisconsin; Flagstaff, Arizona; Austin, Texas; and College Station, Texas). CRFBs were installed initially at half of the sites, while RRFBs were installed at the other half. The same flash pattern was used in all cases, regardless of beacon shape. A staged pedestrian protocol was used to collect driver yielding data to ensure that oncoming drivers received a consistent presentation of approaching pedestrians. The preliminary findings showed only minor, if any, differences between the CRFB and the RRFB (i.e., average daytime yielding rate with CRFB was 63%; it was 59% with RRFB). The results from the generalized linear mixed model indicated no significant differences between the two beacon shapes ( p-value = .4717). For a subset of the sites, the luminous intensity (also called brightness) of the beacons was measured. There was evidence of an increasing yielding rate with increasing intensity at night. In conclusion, the shape of the yellow rapid-flashing beacon did not have an impact on whether a driver decided to yield to waiting pedestrians.

This research effort identified the operational and safety implications of the use of reduced lane and shoulder widths for a variety of urban freeway configurations. The study used speed, crash, and geometric data for freeways in Dallas, Houston, and San Antonio, Texas. The operational analysis identified an increase of about 2.2 mph in speed for a 12-ft lane compared with an 11-ft lane. The shoulder width is significant when the adjacent lane is 11 ft wide, but not when it is 12 ft wide. These findings suggest that the width of the left shoulder is more important with a reduced lane width. Operating speeds on Texas freeways are 2 mph lower during nighttime (with roadside lighting present) than during the day. Speeds were higher (by 1.5 mph) on the weekends (Saturday) than on the weekday studied (Wednesday). The safety analysis determined a crash difference when freeways with 12- to 11-ft lanes were compared. There was a reduction in fatal and injury crashes that ranged from 5% for two-lane freeways up to 12% for five-lane freeways. Similarly, some crash reductions were associated with each additional lane, increased left shoulder widths, and increased right shoulder widths. This paper concludes with a series of sample problem scenarios that collectively demonstrate how the resulting operating speed and crash prediction models allow practitioners to assess the trade-offs that can be expected when urban freeway lane and shoulder widths are modified.

Distraction has been a point of interest in transportation research for many years. Recently, this interest has extended from distracted drivers to distracted pedestrians. Past research suggests that distracted pedestrians are less likely to show cautionary crossing habits and tend toward increased crossing times. This study provided information about the pedestrian waiting to cross and investigated how distraction and other factors may have affected pedestrian start-up time and crossing behaviors. Researchers surreptitiously observed 760 pedestrians at three intersections in College Station, Texas, and coded their characteristics and behavior in real time. Characteristics recorded were age, gender, distraction, and grouping. Distractions recorded were talking on a phone, texting, listening to music, eating, drinking, smoking, and other. The pedestrian groups were no group (crossed alone), mixed-age group, and peer group. Behaviors recorded were glancing before entering the crosswalk, entering the crosswalk early, walking within the crosswalk markings, and hurrying across the crosswalk. Pedestrians who texted and pedestrians who talked on a phone had 21% and 31% more start-up time, respectively. Additionally, pedestrians in mixed-age groups had 14% more start-up time. Texting pedestrians were about two times less likely to glance before entering the crosswalk compared with undistracted pedestrians. Similarly, a pedestrian engaged in a phone conversation was about five times less likely to glance. These results may provide insight into ongoing discussion on how to address distracted pedestrians crossing the street.