Search Content

This report describes the results of studies on the use and effectiveness of rumble strips, including a survey of Minnesota county engineers and a simulation conducted at the University of Minnesota Human Factors Laboratory. Sixty-eight of 87 counties responded to the survey. Of the 68 respondents, 56 install in-lane rumble strips. The survey also asked respondents to describe the guidelines that they used to designate areas for rumble strip installation. As part of the simulation study, test subjects drove in a simulator through a designed experiment to measure the effectiveness of in-lane rumble strips. The results of this study showed different braking patterns between intersections with in-lane rumble strips and those without rumble strips. Those with rumble strips braked earlier and harder. The report recommends that shoulder rumble strips be used in areas with high rates of run-off road crashes. Before-and-after studies have shown conclusively that shoulder rumble strips have reduced run-off road crashes by 20% to 72%. It also recommends a follow-up study on in-lane rumble strips involving drivers that are sleep deprived, under the influence of alcohol, or driving in poor conditions.

This project involved investigating the effect, if any, of rumble strips on stopping behavior at simulated rural-controlled intersections. Researchers used the wrap-around driver simulator at the University of Minnesota's Human Factors Research Laboratory for the project. Researchers varied the rumble strip type and the number of rumble strips and tested them on two different types of controlled intersections, two-way or four-way, and in the presence and absence of traffic. Results indicate that none of these manipulations seem to affect the point at which drivers stop at the controlled intersections or the point at which drivers start to slow down at controlled intersections. The research did reveal drivers brake more, earlier, when rumble strips are installed than they do if there are no rumble strips. Although they started to slow down at the same time and finished braking at the same time, there was more use of the brake earlier in the slowing down maneuver in the presence of rumble strips. Results also reveal that drivers brake more and earlier with full coverage rumble strips than they do with wheel track rumble strips.

In pursuit of unimpeded high peak traffic flow, the "Design Guidelines for Super Two Highways" (Ekem, 1998) suggested several treatments for the right side of the roadway. But the left side is where drivers experience great speed differentials between their own vehicles and oncoming traffic. The authors examined centerline treatments and possible recommendations for Super Two guidelines. The current US standard (12-ft lane/4-in. dashes) was compared with combinations of wider lanes, wider dashes, and buffer areas. With each of the centerline treatments examined, participants kept the left side of the vehicle in the approximate center of the lane. All treatments resulted in shifting the center of the lane farther from the centerline than it was in the standard condition. Two conditions appear to be most effective in keeping drivers away from the centerline: 1) 14-ft lanes with both longitudinal rumble strips and 4-in.-wide dashes marking the centerline, and 2) 12-ft lanes with 4-ft buffer marked by 4-in.-wide dashes. However, implementing any of the centerline treatments should result in vehicles driving farther from the centerline, thus making it less likely that drivers will meet an oncoming vehicle. Data were gathered in a driving simulator. Further testing should be conducted in real driving situations.

This study is the second in a series investigating rumble strips. The objective was to determine the effect of rumble strips on the stopping performance of sleep-deprived drivers. [The study was nested in a larger fatigue study with components unrelated to rumble strips.] The participants were 20 commercial motor vehicle drivers. Each participant was tested over a twenty- hour period, driving in a driving simulator for one hour in the morning, afternoon, evening and at night. During each drive, the participants encountered two stop-controlled intersections-one with rumble strips and the other without rumble strips. The braking pattern of the drivers was affected by the presence of rumble strips- from the appearance of the first set of rumble strips [218 meters (715.2 ft) from the intersection] until the drivers stopped at the intersection. The mean speed of drivers approaching the intersection with the rumble strips was statistically significantly slower than the mean speed for drivers approaching the intersection without the rumble strips. Though sleep deprivation did not affect the driver's braking patterns on the approach to the intersections, it did affect steering variability throughout the course of the drive.

This Rumble Strip Noise Evaluation study presents results of sound level monitoring of three types of longitudinal rumble strips installed along the edge of two-lane rural roads in Polk County, Minnesota. The study is in response to objections raised by some landowners about the unwanted noise caused by vehicles traveling over rumble strips when they drift over the edge or centerline of the roadway. By changing and modifying the design, the ultimate goal is to provide the maximum safety by capturing the driver's attention through tactile and sound levels while minimizing the associated external noise generated by the rumble strips. Both exterior and vehicle interior sound levels were measured from three longitudinal edge of pavement rumble strip designs - California, Pennsylvania and Minnesota. Simultaneous digital audio files were also recorded. Three vehicles were used - a passenger car, pickup, and semi-trailer truck. Tests were performed at 30, 45 and 60 mph. Comparison of exterior and interior sound levels and audio shows that the Pennsylvania design is the quietest, both interior and exterior. The interior level of the Minnesota and California designs are similar but exterior levels are higher for the Minnesota design.

This research effort provided the Minnesota Department of Transportation (MnDOT) with field data on the performance of pavement marking materials when used as rumble stripes on MnDOT roadways. These field efforts provide a perspective on the impact that both wear and winter maintenance practices have on retroreflectivity. Given that these markings were installed by a variety of MnDOT contractors and at different times and roadways, this report also serves to document the range of retroreflectivity provided to drivers at any given time on similar two-lane MnDOT roadways under the installation practice guidelines at the time of installation (2012 to 2013). More specifically, these measurements consider the difference in retroreflectivity provided by direction of travel (e.g., for the same marking, what is the retroreflectivity while driving northbound versus southbound?) and by roadway. The long-term evaluation collected field measurements both initially and after two winters (18 months) for centerline rumble stripes only and on seven segments over three different roadways. The in-service evaluation included new centerline and profile rumble stripEs, all of which were installed as part of the 2013 mill and overlay projects on bituminous surfaces within District 4 on two-lane MnDOT roadways. The retroreflectivity data were collected one winter (approximately 12 months) after installation with no initial measurement data being available. This effort included measuring the centerline rumble stripe performance over eight segments on four different roadways and the profile rumble stripe performance over 18 segments on 10 different roadways.

This Sinusoidal Rumble Strip Design Optimization Study presents results of sound level monitoring of four types of centerline rumble strips installed along Trunk Highway (TH) 18 in Mille Lacs and Aitken counties in Minnesota. This study is in response to objections raised by some landowners about the unwanted noise caused by vehicles traveling over rumble strips when they drift over the edge or centerline of the roadway. By changing and modifying the design, the ultimate goal is to provide the maximum safety by capturing the drivers attention through in-vehicle generated sound levels while minimizing the associated external noise generated by the rumble strips. Tests on TH 18 were performed with three different vehicles passenger car, pickup truck and a class 35 tandem dump truck. A single speed of 60 mph was used, as this was shown to provide the most meaningful data in the previous study. For each of the designs, an initial test was performed with vehicles traveling on normal pavement, followed by three passes on the rumble strip. Rumble strip designs 1 and 4 created lower exterior sound level increases but created interior levels similar to designs 2 and 3. The external results correspond to the depth of the rumble strip design, with designs 1 and 4 having a maximum depth of 1/8 inch less than designs 2 and 3. The interior sound level increases are similar for all four designs but vary by vehicle type. All of the designs created increases greater than 10 dBA for the passenger car, which is a desirable level for gaining attention of the driver. For the pickup truck, the interior sound level increases ranged from 4.5 to 6.8 dBA, while the increases for the dump truck ranged from 0.8 to 2.7 dBA.

As a low-cost countermeasure to rural intersection crashes, transverse rumble strips (TRS) provide an audible and tactile warning to drivers approaching an intersection with the primary goal of decreasing crashes that result from running a stop sign. The objective of this project is to evaluate the effectiveness of different TRS patterns on stopping behavior at rural stop-controlled intersections. Eight rural intersections in St. Louis County, Minnesota, were selected as test sites. Milled-in rumble strips were installed at the sites that varied in terms of number of panels (2 or 3) and number of rumble strips per panel (6 or 12). Speed, traffic volume, and video data were collected at each site before, 1 month after, and 9 months after TRS installation to evaluate various crash surrogate metrics. The most significant metrics affected by TRS configuration included percentage of vehicles engaging in a full/rolling stop at the intersection, change in average speeds on the approach near the intersection, percent of vehicles traveling 45 mph or more, and percent of vehicles engaging in late braking. A qualitative summary of the various metrics suggested that the 3-panel, 12-rumble strip design performed the best. Noise analyses were also conducted to assess whether the number of rumble strips per panel (6 or 12) affected exterior and in-vehicle noise. No significant differences in exterior noise were found, and both panels produced sufficient in-vehicle noise to alert a drowsy driver. As a result, noise was not a factor in selecting one panel type over another.

This Technical Summary pertains to Report 2023-17, “Transverse Rumble Strips at Rural Intersections,” published April 2023.

This evaluation determined the change in crash frequency, type or severity associated with longitudinal sinusoidal rumble strips on rural two-lane undivided Minnesota roadways constructed between 2018 and 2022. Crash modification factors (CMFs) were estimated using cross-sectional analysis to compare crash experience of locations with sinusoidal rumble strips (i.e., centerline only, centerline and shoulder, or shoulder only) compared to roads with rectangular rumble strips. The cross-sectional analysis matched sites with sinusoidal and rectangular rumble strips using matched-pair comparisons. Negative binomial (NB) or Poisson log-linear regression models were used to model the crashes at all treatment and non-treatment sites. There was a total of approximately 327 miles of treated (i.e., centerline only, centerline and shoulder, or shoulder only sinusoidal rumble strips) and approximately 302 miles of untreated (i.e., centerline only, centerline and shoulder, or shoulder only rectangular rumble strips) on rural two-lane divided roads. Overall, the results of the models indicated no significant differences in crash rates between rural two-lane undivided roads with sinusoidal rumble strips, and rural two-lane undivided roads with rectangular rumble strips.