Drowsy or inattentive drivers who drift off the right side of the road present a significant safety concern. In an effort to alert these drivers to their potentially dangerous situation, a number of states are testing or using "an innovative type of shoulder rumble strip called the Sonic Nap Alert Pattern (SNAP)." When a vehicle's tires roll over the strip, they "produce a distinct warning sound and vibration alerting drowsy or inattentive drivers that their vehicles are drifting along the shoulder of the roadway."

Engineer Neal E. Wood developed SNAP, and the innovative Rumble strips were installed and tested on an abandoned section of the Pennsylvania Turnpike. Because raised strips on a roadway's shoulders had been found to create problems for snowplows and service vehicles, the Turnpike engineers "tested only narrow and recessed rumble strip patterns with varying lengths and depths and selected an effective design with enough sound and vibration to be perceptible in a truck cab and yet not too severe for cars or motorcycles."

John J. Hickey, Jr. reported the findings of the Turnpike installation and testing in "Shoulder Rumble Strip Effectiveness: Drift-Off-Road Accident Reductions on the Pennsylvania Turnpike" (Transportation Research Record 1573). Results showed that SNAP reduced the number of yearly drift-off-road (DOR) crashes by about 100, which was a 60 percent reduction. These results were evidence that SNAP "demonstrated a substantial safety benefit."

METHODS

All the Turnpike testing spaced the indentations every 0.305 m (1 ft.) along the vehicle's travel direction "based on vehicles drifting off at a shallow 3-degree departure angle, striking a succession of indentations to produce a tone with enough duration to awaken a drowsy driver."

All test patterns were transverse to travel direction and were 40.64 cm (16 in.) in width so that wide truck tires would contact enough indentations. A number of tests for indentation depth showed that "1.27 cm (0.50 in.) was deep enough for tire drop to produce an alerting sound, provided that the opening (length in the direction of travel) was large enough for various sized tires." The openings were 10.16 cm (4 in.) long, with 20.32 cm (8 in.) between, which produced sound adequate for both cars and trucks--86 decibels in a truck cab at 65 mph.

After the first installation, engineers tried milling SNAP in existing pavement rather than installing it during paving. The milling head resulted in 17.78 cm (7 in.) openings that were 12.7 cm (5 in.) apart, which in turn resulted in a "perceptible difference" in sound level (89 decibels rather than the earlier 86). At the time of this report, "all but a few miles of initial SNAP installation projects [were] . . . 17.78 x 40.64 cm (7 x 16 in.) indentations, 1 groove per 0.305 m, milled 1.27 cm deep starting 10.16 cm outside the roadway edgeline along the shoulder." These dimensions produced enough sound and vibration to effectively alert both cars and trucks traveling at highway speeds.

Costs for SNAP are less than $0.30 per 0.305 m of asphalt shoulder, and it can be installed on both shoulders of 1.62 km (1 mile) of highway in six hours for under $5000. Installation for the entire 815 km (506 miles) Turnpike was expected to cost between $2 and $3 million.

RESULTS

The first installation project (1989) experienced "a dramatic reduction in DOR accidents" after 18 months, such that "a decision [was made] to install SNAP systemwide as a part of all roadway resurfacing projects." Monthly accident reports (for crashes involving fatality, injury, or vehicle damage that required towing) for the first five projects showed that "accidents decreased from an average of 0.518 per month before SNAP was installed to 0.155 after SNAP treatment." This was a 70 percent reduction in crashes.

The next four installation project areas showed high DOR rates for the 37 months preceding SNAP installation; these rates were comparable to those for similar roadways. Resurfacing schedules rather than DOR rates were the selection basis for these four installations. Even so, the average DOR rates for these four highway segments decreased by 69 percent. All crashes on the Turnpike decreased by 12 percent during this same period; therefore, after adjustment for this overall decline, "a 65 percent reduction in DOR rates [was] attributable to SNAP."

Given such "impressive reductions," it was reasonably expected that more SNAP installations on the Turnpike would result in "a noticeable decline in single-vehicle, off-right-side accidents." Such a decline would be further "persuasive evidence of SNAP's effectiveness." The numbers appeared to verify those expectations. From 1990 to 1995, the number of crashes susceptible to SNAP treatment decreased from 172 to 74, a 57 percent reduction. Crash reduction for large trucks was similar--at 60 percent.

Certain types of single-vehicle crashes not susceptible to SNAP treatment were excluded from the evaluation, including those attributable to weather conditions. However, SNAP can be a positive factor in weather-related conditions. For example, the pattern can be seen and heard through some snow-and-ice conditions, and it can be used as an edgeline guide through snow. Snowplow operators have also used SNAP as a guide. In general, "weather-related accidents were subtracted to remove an annual variation that could mask SNAP's effectiveness, not necessarily because SNAP could have no effect."

After 1991, all repaving projects included SNAP and Recessed Reflective Pavement Markers (RRPM), though the latter were "not believed to have much effect on the off-right-side accidents considered in this evaluation of SNAP effectiveness." DOR rates were compared for all roadways both before and after SNAP/RRPM treatment. Fifty-three segments (totaling 348 miles) "experienced a 60 percent reduction in treatable-type accidents per Vehicle-Miles-Traveled (VMT)."

Twenty-five segments (totaling 186 miles) were treated with only SNAP/RRPM and no repaving or shoulder work. They experienced a 63 percent reduction in crashes, which "suggest[ed] that the other work that was usually performed along with SNAP treatment did not contribute to the DOR accident reductions." Thirteen untreated segments (totaling 90 interspersed miles) experienced a 31 percent reduction in crashes. While these segments "could represent control sections," the "DOR accidents selected for comparison were very likely to have been affected by SNAP installed in adjacent sections of roadway." Since roadway segments averaged 6.6 miles (10.6 km) in length, a "spillover effect" was likely on the untreated segments "as drivers alerted to their drowsiness or inattention could remain affected for the 2 to 10 minutes required to traverse untreated segments at highway speeds."

CONCLUSIONS