This page presents information on the safety of rural, low-volume roads and on research that supports the use of ELRs to improve their safety.
Two significant problems exist on rural, low-volume roads which are addressable with the use of edge lane roads.
Two significant problems exist on rural, low-volume roads which are addressable with the use of edge lane roads.
- Over half of all crashes on these roads are single-vehicle, roadway departure crashes (approx 56% of all crashes per NCHRP 362).
- These roads often lack usable shoulders and sidewalks forcing pedestrians and cyclists into the travel lane.
TELL ME MORE ABOUT THIS PROJECT!
Rural ABL Project
Problem Description
Background
According to 2016 FHWA statistics, rural areas are home to 1.63 million of the 2.75 million miles of paved roads in the US. 1.37 million of those rural road-miles consist of local and collector roads.
Many of these roads have no sidewalks or useable shoulders. Vulnerable road users (VRUs) are forced to use the travel lanes which is both dangerous and intimidating. Without billions of dollars of funding, these roads are likely to continue without substantial improvement for decades or longer.
In addition, these roads have a unique safety profile. Due to their low volumes and physical setting, over half of all crashes, about 56%, are single-vehicle, run-off-the-road crashes. Rates for these crash types drop significantly with added lane or shoulder width.
These two problems appear to be addressable with the use of ELRs.
This project is an effort to research the possibility of, and create design guidance for, the application of ELRs to these roads. The two questions this project needs to answer are:
The first question is best answered by research into existing domestic and international practices (primarily Great Britain and Australia where this treatment is already used), review of findings from the literature and prospective research based on computer simulations and extrapolations from human factors and safety statistics.
The answer to the second question will be informed by answers to the first question, a review of ELR best practices and an approach of creating guidance which targets early successes rather than creating as broad an applicable domain as possible.
Problem Space Description
The roads targeted by this investigation are assumed to have the following characteristics:
While this project is only investigating ELRs as a possible solution, any solution should have at least the following characteristics:
Guidance for Rural ELRs is not a declaration that protection is unnecessary or unwanted on rural, low-volume roads. It is an acknowledgement that sufficient resources to provide safe, comfortable facilities on all roads in the near future are not available but that near-term improvement of those roads is still desired.
Problem Description
Background
According to 2016 FHWA statistics, rural areas are home to 1.63 million of the 2.75 million miles of paved roads in the US. 1.37 million of those rural road-miles consist of local and collector roads.
Many of these roads have no sidewalks or useable shoulders. Vulnerable road users (VRUs) are forced to use the travel lanes which is both dangerous and intimidating. Without billions of dollars of funding, these roads are likely to continue without substantial improvement for decades or longer.
In addition, these roads have a unique safety profile. Due to their low volumes and physical setting, over half of all crashes, about 56%, are single-vehicle, run-off-the-road crashes. Rates for these crash types drop significantly with added lane or shoulder width.
These two problems appear to be addressable with the use of ELRs.
- ELRs add substantial shoulder width which reduces the single-vehicle crash rate.
- ELRs provide VRU facilities until more substantial improvements are possible.
This project is an effort to research the possibility of, and create design guidance for, the application of ELRs to these roads. The two questions this project needs to answer are:
- "In what volume/speed domains are ELRs likely to be safer than the standard two lane configuration?" and
- "If usable domains are suggested by the answer to Question 1, then what design guidance is recommended?"
The first question is best answered by research into existing domestic and international practices (primarily Great Britain and Australia where this treatment is already used), review of findings from the literature and prospective research based on computer simulations and extrapolations from human factors and safety statistics.
The answer to the second question will be informed by answers to the first question, a review of ELR best practices and an approach of creating guidance which targets early successes rather than creating as broad an applicable domain as possible.
Problem Space Description
The roads targeted by this investigation are assumed to have the following characteristics:
- Rural environment - application to other environments is not addressed.
- Two lanes - presence of a center line is standard; edge/fog lines are optional.
- Rural Local or Collectors - Data collected by the FHWA recommends an upper volume for local rural roads as 400 ADT and rural collectors as 2600 ADT.
- Speeds above 35 MPH - the Advisory Shoulder section in the Small Town & Rural Multimodal Networks guide already includes roads of 35 MPH and lower.
- Limited road width - these roads normally consist only of enough paved width to support two vehicular travel lanes, typically 20’-30’ wide.
- Low probability of improvements - The prospect of adding road width or facilities for VRUs in the near future is limited to non-existent.
- Lack of sidewalks or useable shoulders - Lack of dedicated facilities force VRUs to walk on the unpaved shoulder or on the paved roadway. A number of VRUs (those using wheelchairs, personal mobility devices, scooters, etc) may not be able to use a well-developed, gravel shoulder. Open drainage swales on one or both sides is not an uncommon condition.
- At-Grade Intersections - Almost all intersections are at-grade and highly unlikely to be improved beyond that configuration.
- Limited Lighting - Most roads have no lighting save for light from roadside structures or activities.
While this project is only investigating ELRs as a possible solution, any solution should have at least the following characteristics:
- As safe as, or safer than, a standard 2-lane road for all road users
- More comfortable/attractive for VRUs than a standard 2-lane road
- Inexpensive to install and maintain
- Viable at the targeted speed range, 35 - 55/60 MPH - “Viable” covers a lot of ground and an exact definition is avoided for the present.
- All of the solution requirements must be achievable on local rural roads (at or below 400 ADT) and potentially achievable on rural collectors (at or below 2,600 ADT). The volume ceiling for rural ELRs should be founded on a data-based rationale as well as a level which increases chances of initial success. Following a successful rollout at lower volumes, it may be reasonable to explore higher volumes. 400 ADT may be a good initial choice due to the FHWA guidelines on local rural roads and the AASHTO Low Volume guidance.
Guidance for Rural ELRs is not a declaration that protection is unnecessary or unwanted on rural, low-volume roads. It is an acknowledgement that sufficient resources to provide safe, comfortable facilities on all roads in the near future are not available but that near-term improvement of those roads is still desired.
MORE DETAIL
ELRs to make Rural Roads Safer
The Problem
The ELR format provides wide shoulders at little cost. On 2-lane roads with 20-32’ of pavement, ELRs provide 5-6’ of additional shoulder width.
No direct evidence exists to answer this question. Some less-direct evidence suggests that this will not be a problem.
The existing evidence appears to predict that ELRs on these rural roads will significantly improve safety. But more research is needed.
The Problem
- Rural roads host 54% of all fatalities in the US despite serving only 19% of the population.
(https://safety.fhwa.dot.gov/local_rural/) - 56% of crashes on two lane, rural roads are roadway departure, single-vehicle crashes.
(NCHRP Report 362) - Many rural 2-lane roads have narrow lanes and little to no shoulder.
(NCHRP Report 362) - Rates of roadway departure crashes drop significantly with extra lane or shoulder width.
(CMF ID 5285, NCHRP Report 362, FHWA-RD-87/008) - Rural fatalities tend to be randomly located making systemic safety the preferred approach.
(https://safety.fhwa.dot.gov/systemic/fhwasa17010/)
The ELR format provides wide shoulders at little cost. On 2-lane roads with 20-32’ of pavement, ELRs provide 5-6’ of additional shoulder width.
- CMF ID 5285 predicts a 29% crash rate drop after widening paved shoulders from 3 to 8 feet.
- FHWA-RD-87/008 predicts a 40% crash rate drop for a 6’ increase in shoulder width.
- NCHRP 362 predicts crash rate drops ranging from 24% to 59% for addition of 6.5’ shoulders.
No direct evidence exists to answer this question. Some less-direct evidence suggests that this will not be a problem.
- An unknown percentage of head-on collisions result from a road departure on the right followed by an over-correction. Reducing road departures with wider shoulders will reduce this root cause of head-on collision.
- CMF ID 87, which addresses injury crash rate after adding center lines to rural 2-lane roads, shows the most likely outcome is no change in crash rate after center line removal (CMF Clearinghouse, CMF ID 87 = .99 for all injury crashes). This demonstrates motorists' ability to avoid head-on collisions on roads without center lines.
- One can examine the parallels between passing zones on 2-lane rural road and ELRs. Both require oncoming drivers sharing one lane to avoid head-on collisions. The low rate of head-on collisions overall (2-5%) and the very low incidence of head-on collisions as a result of passing movements suggests that this will not be an issue.
The existing evidence appears to predict that ELRs on these rural roads will significantly improve safety. But more research is needed.
REFERENCES & ADDITIONAL DATA
A Selection of References & Supporting Data
ZEGEER, C.V., HUMMER, J., REINFURT, D., HERF, L., and HUNTER, W., "Safety Effects of Cross-Section Design for Two-Lane Roads." Report No. FHWA-RD-87/008. Federal Highway Administration and Transportation Research Board, Washington, D.C. (Oct. 1987). NCHRP Report 362 summarizes the results of this study as:
“For lane widths between 8 and 12 ft, the predictive accident model showed that related accidents were reduced by 12 percent for 1 ft of lane widening (for example, 10-ft lanes to I I -ft lanes), 23 percent for 2 ft of lane widening, 32 percent for 3 ft of lane widening, and 40 percent for 4 ft of lane widening. The study also modeled the effects of shoulder widening on related accidents for paved and unpaved shoulders. For shoulder widths between 0 and 12 ft, the percent reduction in related accidents resulting from widening paved shoulders was predicted to be 16 percent for 2 ft of widening, 29 percent for 4 ft of widening, and 40 percent for 6 ft of widening.”.
40% reduction in crash rate predicted for increase of shoulder width by 6 feet.
GRIFFIN, L.I. and MAK, K.K., "Benefits to be Achieved from Widening Rural, Two-Lane, Farm-to-Market Roads in Texas." Paper Presented at the 1988 Annual Transportation Research Board Meeting.This study looked at 36,215 miles of rural Texas roads with 1500 ADT or less. NCHRP 362 claims that the crash reduction rates for lane widening produced by Griffin and Mak are very similar to those produced by Zegeer, 1987.
Independent confirmation of results in 1987 Zegeer study.
CMF for Shoulder Widening
CMF ID 5285 addresses crash rate after widening paved shoulder from 3 to 8 feet.
CMF ID 5285 has 4 stars but no error range available.
CMF ID 5285 (for all crash types in unspecified area) = 0.71.
Widening of shoulders from 3 to 8 feet can reduce all crash types by 29%.
CMF for Center Line Addition/Removal
CMF ID 87 addresses injury crash rate after adding center lines to unmarked rural 2-lane roads.
CMF ID 87 has 3 stars and an adjusted standard error of 0.1 or less.
CMF ID 87 (for all injury crashes) = .99, CMF Value Range: .87 – 1.11
Adding center lines can reduce injury crashes up to 13% or increase them up to 11%.
Most likely outcome is no change in the crash rate with removal of center line.
Reminder: CMF * crash rate = new crash rate, e.g. .8 (CMF) * 100 crashes/year = 80 crashes/year.
NCHRP Report 362
Figure 5 drawn from NCHRP Report 362
Compare crash rates from Figure 5 for the 11’/2’ configuration of the 2-lane road and for ABL equivalents. The 11’/2’, 2-lane road produces a related accident rate of 1.87 crashes per million vehicle miles. The ABL equivalents produce accident rates of 1.43 for 10’/8’, 1.31 for 11’/7.5’, 1.31 for 12’/7/, and 0.76 for 13’/6.5’.
Crash rate reductions range from 24% (1.43) to 59% (0.76).
90% of rural 2-lane roads carry volumes of 1,000 vehicles per day or less.
Per NCHRP Report 362, “of the 3.1 million miles of two-lane rural roads, approximately 90 percent (2.8 million miles) have an average daily traffic (ADT) of less than 1,000 vehicles per day (vpd). About 80 percent have ADTs of less than 400 vpd, and 38 percent carry less than 50 vpd.”
Many rural 2-lane roads have narrow lanes and little to no shoulder
Per NCHRP Report 362, “over one-quarter of the mileage of such roads have lane widths of 9 ft or less, and two-thirds have shoulder widths of 4 ft or less. In addition, 11.5 percent of two-lane highway mileage (356,500 miles) have no shoulders. ” Doing the math over 3.1 million miles tells us that over 775,000 miles have lane widths of <= 9’ and 2.07 million miles have shoulders <= 4’ width.
Scotland ELR Example
Highway B764 in East Renfrewshire, Scotland has used an ELR format since 2005.
60 MPH posted speed limit, 1400 ADT, 10-100+ bikes/day, 22’ pavement width.
Location: 55.719918, -4.331990.
ZEGEER, C.V., HUMMER, J., REINFURT, D., HERF, L., and HUNTER, W., "Safety Effects of Cross-Section Design for Two-Lane Roads." Report No. FHWA-RD-87/008. Federal Highway Administration and Transportation Research Board, Washington, D.C. (Oct. 1987). NCHRP Report 362 summarizes the results of this study as:
“For lane widths between 8 and 12 ft, the predictive accident model showed that related accidents were reduced by 12 percent for 1 ft of lane widening (for example, 10-ft lanes to I I -ft lanes), 23 percent for 2 ft of lane widening, 32 percent for 3 ft of lane widening, and 40 percent for 4 ft of lane widening. The study also modeled the effects of shoulder widening on related accidents for paved and unpaved shoulders. For shoulder widths between 0 and 12 ft, the percent reduction in related accidents resulting from widening paved shoulders was predicted to be 16 percent for 2 ft of widening, 29 percent for 4 ft of widening, and 40 percent for 6 ft of widening.”.
40% reduction in crash rate predicted for increase of shoulder width by 6 feet.
GRIFFIN, L.I. and MAK, K.K., "Benefits to be Achieved from Widening Rural, Two-Lane, Farm-to-Market Roads in Texas." Paper Presented at the 1988 Annual Transportation Research Board Meeting.This study looked at 36,215 miles of rural Texas roads with 1500 ADT or less. NCHRP 362 claims that the crash reduction rates for lane widening produced by Griffin and Mak are very similar to those produced by Zegeer, 1987.
Independent confirmation of results in 1987 Zegeer study.
CMF for Shoulder Widening
CMF ID 5285 addresses crash rate after widening paved shoulder from 3 to 8 feet.
CMF ID 5285 has 4 stars but no error range available.
CMF ID 5285 (for all crash types in unspecified area) = 0.71.
Widening of shoulders from 3 to 8 feet can reduce all crash types by 29%.
CMF for Center Line Addition/Removal
CMF ID 87 addresses injury crash rate after adding center lines to unmarked rural 2-lane roads.
CMF ID 87 has 3 stars and an adjusted standard error of 0.1 or less.
CMF ID 87 (for all injury crashes) = .99, CMF Value Range: .87 – 1.11
Adding center lines can reduce injury crashes up to 13% or increase them up to 11%.
Most likely outcome is no change in the crash rate with removal of center line.
Reminder: CMF * crash rate = new crash rate, e.g. .8 (CMF) * 100 crashes/year = 80 crashes/year.
NCHRP Report 362
Figure 5 drawn from NCHRP Report 362
Compare crash rates from Figure 5 for the 11’/2’ configuration of the 2-lane road and for ABL equivalents. The 11’/2’, 2-lane road produces a related accident rate of 1.87 crashes per million vehicle miles. The ABL equivalents produce accident rates of 1.43 for 10’/8’, 1.31 for 11’/7.5’, 1.31 for 12’/7/, and 0.76 for 13’/6.5’.
Crash rate reductions range from 24% (1.43) to 59% (0.76).
90% of rural 2-lane roads carry volumes of 1,000 vehicles per day or less.
Per NCHRP Report 362, “of the 3.1 million miles of two-lane rural roads, approximately 90 percent (2.8 million miles) have an average daily traffic (ADT) of less than 1,000 vehicles per day (vpd). About 80 percent have ADTs of less than 400 vpd, and 38 percent carry less than 50 vpd.”
Many rural 2-lane roads have narrow lanes and little to no shoulder
Per NCHRP Report 362, “over one-quarter of the mileage of such roads have lane widths of 9 ft or less, and two-thirds have shoulder widths of 4 ft or less. In addition, 11.5 percent of two-lane highway mileage (356,500 miles) have no shoulders. ” Doing the math over 3.1 million miles tells us that over 775,000 miles have lane widths of <= 9’ and 2.07 million miles have shoulders <= 4’ width.
Scotland ELR Example
Highway B764 in East Renfrewshire, Scotland has used an ELR format since 2005.
60 MPH posted speed limit, 1400 ADT, 10-100+ bikes/day, 22’ pavement width.
Location: 55.719918, -4.331990.
A short video of a ride on a Dutch country road. Note the "shoulders" that wouldn't exist on a two lane road and the widening of the clear zone for motorists traveling in the center lane.
Photos courtesy of Richard Sparks.