U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

 
Public Roads
This magazine is an archived publication and may contain dated technical, contact, and link information.
Public Roads Home | Current Issue | Past Issues | Subscriptions | Article Reprints | Author's Instructions and Article Submissions | Search Public Roads
Publication Number:  FHWA-HRT-17-003     Date:  March/April 2017
Publication Number: FHWA-HRT-17-003
Issue No: Vol. 80 No. 5
Date: March/April 2017

 

Providing A Shoulder to Drive On

by Jim Hunt, Pete Jenior, and Greg Jones

States are turning freeway shoulders into part-time travel lanes to relieve congestion as a cost-effective alternative to traditional widening.

More and more, States are considering part-time shoulder use, such as this application in Honolulu, HI, as a strategy to improve capacity when traditional road widening is not feasible.
More and more, States are considering part-time shoulder use, such as this application in Honolulu, HI, as a strategy to improve capacity when traditional road widening is not feasible.

Transportation officials are having difficulty simply maintaining, let alone expanding, the Nation’s highway infrastructure. Aging assets, a growing population, and revenue uncertainty add to the hurdles facing State and local transportation agencies.

Relieving congestion on urban freeways through conventional widening projects is often impractical because of costs and the detrimental impacts of construction activities on travelers and surrounding businesses and residents. Plus, once a freeway widening project advances to implementation, it can take years to complete, delaying the realization of any benefits.

Freeway congestion is a recurring performance problem, but it is also a condition that typically occurs only during limited hours of the day. One innovative, relatively low-cost solution States are now implementing is to allow traffic to use freeway shoulders as travel lanes on a part-time basis. Many transportation agencies are implementing or considering part-time shoulder use as a performance-based practical design solution to strategically invest limited transportation funding and maximize system performance. The United States has a growing list of successful projects, and other countries have a long history of effective shoulder use.

Freeway congestion often varies widely throughout the day, and part-time shoulder use can provide congestion relief when it is needed most, while maintaining shoulders for their primary purposes for the remainder of the day and night. Those primary uses include refuges for vehicles during emergency situations, access for first responders, and additional recovery areas for drivers who may have to swerve out of the travel lane to avoid conflicts in the adjoining travel lanes.

Engineers try to preserve wider shoulders when possible; however, the cost to widen a freeway is significant and each foot of pavement width counts. As a result, many examples of urban freeways with narrower lanes and shoulders exist and the value of having shoulders is diminished.

Photo. A bus travels in the shoulder lane on I–55. The other travel lanes are congested with other vehicles.
In Chicago, bus-on-shoulder operations on I–55 have improved trip reliability and increased bus ridership significantly.

To assist State and local agencies in adopting the strategy, the Federal Highway Administration released the Use of Freeway Shoulders for Travel: Guide for Planning, Evaluating, and Designing Part-Time Shoulder Use as a Traffic Management Strategy (FHWA-HOP-15-023). The guide highlights more than 30 shoulder-use installations covering a range of design and operational approaches in 16 States. Part-time shoulder use requires special attention to planning, design, public outreach, implementation, safety, operations, and maintenance. The guide aims to help practitioners consider the strategy and promote consistent practices in evaluating and developing shoulder-use concepts.

How Does Part-Time Shoulder Use Work?

Because part-time shoulder use maximizes existing roadway capacity, the strategy provides a solution that transportation agencies can implement quickly, at a much lower cost and with fewer environmental impacts than traditional projects that expand capacity.

Part-time shoulder use, or “shoulder running,” can take many forms. However, all scenarios involve use of the left or right shoulder of an existing roadway for temporary travel during certain hours of the day.

Current implementations of part-time shoulder use are primarily in locations where recurring congestion exists because of traffic bottlenecks or lack of capacity during peak periods. For example, Pace Suburban Bus, a public transportation agency serving the greater Chicago, IL, area, identified bus-only shoulder use as a strategy that could increase the reliability and attractiveness of public transportation. Compared to constructing new dedicated bus lanes, bus use of shoulders is much more cost effective. In 2011, the agency implemented bus-on-shoulder service on the Stevenson Expressway (I–55) as a demonstration project. Because of the success of this demonstration, the State passed legislation permanently allowing bus-on-shoulder service and expanding that permission to all the region’s expressways and tollways in 2014.

States Employing Part-Time Shoulder Use in 2016
States Employing Part-Time Shoulder Use in 2016 - Map. The following States use part-time, bus-on-shoulder operation: Delaware, Florida, Illinois, Kansas, Maryland, Minnesota, New Jersey, North Carolina, Ohio, Virginia, and Washington State. Eight States employ static part-time shoulder use: Colorado, Georgia, Hawaii, Massachusetts, New Jersey, Virginia, Texas, and Washington State. Dynamic shoulder use has been implemented in Minnesota and Virginia.

“Since Pace Bus received approval for bus-on-shoulder operations on I–55 in 2011, bus ridership on that corridor has more than quadrupled,” says Doug Sullivan, department manager of marketing with Pace Suburban Bus. “And on-time performance--which averaged less than 70 percent--is now over 90 percent.”

Dynamic part-time shoulder use, as shown here on I–66 in northern Virginia, provides an “on-demand” strategy for when special events or conditions cause heavy congestion.
Dynamic part-time shoulder use, as shown here on I–66 in northern Virginia, provides an “on-demand” strategy for when special events or conditions cause heavy congestion.

Objectives and Facility Considerations

Currently, 16 States operate some type of part-time shoulder use. Although historically allowing bus use on shoulders has been the most common type of implementation, several States have locations that accommodate general traffic on the shoulder for a portion of the day and many others are exploring such implementations.

Determining the type of part-time shoulder use that is appropriate for a given corridor depends on an agency’s objectives and specific facility characteristics. Typical part-time shoulder use aims to achieve one or more of the following objectives:

  • Offer relief from peak-period congestion for a minimal cost compared to adding new general purpose lanes.
  • Provide additional capacity on an interim basis while a conventional widening project works through the planning, design, environmental, and constructionprocesses.
  • Increase bus ridership by improving bus travel time and reliability.
  • Preserve a full-width shoulder during off-peak periods, which provides safety performance benefits as compared to a widening project that results in narrower lanes and shoulders.

Among the facility characteristics that States need to consider are roadway and interchange geometrics including ramps, physical constraints, pavement type and quality, shoulder maintenance and stormwater management, travel demand patterns, and traffic characteristics.

Several types of design and operations options are available for part-time shoulder use. These include restricting shoulder use to authorized transit buses or allowing use by all (or most) vehicles. If all vehicles can use the shoulder, another option is to open it for travel only during fixed periods, such as each weekday peak period. This is known as static shoulder use. Alternatively, shoulder use can be based on prevailing and predicted conditions to accommodate special events or incidents that trigger heavy congestion. This is known as dynamic shoulder use and typically employs dynamic lane control signs (such as redX or green arrow indicators).

Other design and operations considerations include whether to allow travel on the left or right shoulder and whether to vary speed limits (or post speed advisories) on the general purpose or shoulder lanes when the shoulder is open.

FHWA’s Simulated Part-Time Shoulder Capacity
Part-Time Shoulder-Use Scenario Shoulder Capacity (Vehicles Per Hour)
Short segment and low quality 1,262
Long segment and low quality 1,334
Short segment and high quality 1,610
Long segment and high quality 1,687
Note: Short is defined as a 1,000-foot (305-meter) part-time shoulder encompassing the length of a bottleneck. Long is defined as a 1.5-mile (2.4-kilometer) part-time shoulder encompassing the length of a bottleneck and queue prior to it. A low-quality, part-time shoulder is defined as being 10 feet (3 meters) wide that only 50 percent of drivers are willing to use. A high-quality, part-time shoulder is defined as being 12 feet (3.7 meters) wide with “normal freeway lane design standards” that all drivers are willing to use.

Capacity and Safety Performance

Based on FHWA-sponsored research and simulation studies, the extent to which a shoulder may provide additional capacity compared to a general purpose freeway lane (albeit on a part-time basis) varies and is influenced by the quality of the shoulder. Quality factors include shoulder width and length, distance from the travel way to roadside features, and other elements that might make some drivers uncomfortable or unwilling to travel on the shoulder.

Predicted Crash Frequency for Freeway Conversions
Line graph. The graph shows the correlation between the predicted change in annual crash frequency and the two-way annual average daily traffic for an 8- to 10- lane freeway conversion, a 6- to 8-lane freeway conversion, and a 4- to 6-lane freeway conversion. The vertical axis is labeled in increments of two from negative 12 percent to 8 percent for the predicted change in annual crash frequency. The horizontal axis is labeled in increments of 10,000 vehicles per day from 50,000 to 120,000. For a 4- to 6- lane freeway conversion, the predicted change in annual crash frequency is 3.12 percent at 50,000 vehicles per day, 1.03 percent at 60,000, negative 2.39 percent at 70,000, and negative 6.54 percent at 80,000. For a 6- to 8-lane freeway conversion, the predicted change in annual crash frequency is 6.22 percent at 60,000 vehicles per day, 4.43 percent at 70,000, 3.42 percent at 80,000, 1.43 percent at 90,000, negative 1.19 percent at 100,000, negative 3.41 percent at 110,000, and negative 5.11 percent at 120,000. For an 8- to 10-lane freeway conversion, the predicted change in annual crash frequency is 4.71 percent at 60,000 vehicles per day, 2.56 percent at 70,000, 0.72 percent at 80,000, negative 0.86 percent at 90,000, negative 2.10 percent at 100,000, negative 3.70 percent at 110,000, and negative 5.73 percent at 120,000.

The results of FHWA’s simulations are generally consistent with capacities observed in the field and the proven relationship between design features and capacity. For example, the lane designated for part-time shoulder use on I–66 in northern Virginia is 12 feet (3.7 meters) wide, has large portions of paved shoulder several feet wide beyond it, and has overhead dynamic lane control signs. Observed capacity of the shoulder during in-use periods is similar to the adjacent general purpose lanes (approximately 2,000 vehicles per hour per lane).

Illustration. Shown here is a section of highway with three general purpose lanes and a shoulder lane. The shoulder lane is shaded. Arrows illustrate how the conversion of a taper-style ramp to a parallel-style ramp help to remove conflict between ramp and shoulder traffic. In the original design, traffic exited directly from the rightmost general purpose lane to the ramp, as is typical for a taper-style ramp. Crossing the shoulder would create a conflict with a shoulder lane for travel, so the pavement markings were modified to have exiting traffic change lanes onto the shoulder for several hundred feet prior to the exit ramp. Creation of a short lane prior to the exit is effectively a parallel-style ramp design.
FHWA included this illustration in its guide on part-time shoulder use to help practitioners understand various design scenarios. It depicts a freeway ramp junction where shoulder use is permitted and the pavement markings were modified to have exiting traffic change lanes onto the shoulder for several hundred feet prior to the exit ramp.

In contrast, the lanes designated for part-time shoulder use on I–93 in Massachusetts are less than 12 feet (3.7 meters) wide, generally have only a 1- to 2-foot (0.3- to 0.6-meter) paved shoulder beyond the shoulder lane, use limited dynamic signs, and include interchanges with constrained geometries. The shoulder on I–93 was observed to have one-half to two-thirds the capacity of adjacent general purpose lanes. The range of potential capacity increases from shoulder use illustrates the influence of road geometrics such as shoulder lane width, effective shoulder width when the shoulder lanes are in use, and interchange spacing and design.

With regard to safety performance, the sources of safety-related research evaluating the specific changes in crash frequency and severity as a result of implementing part-time shoulder use is limited to only a few Federal and State empirical studies. Crash frequency has increased in some locations following part-time shoulder-use projects and decreased as a result in others, suggesting the effect of part-time shoulder use on crash frequency is influenced significantly by site-specific operational and geometric characteristics. No transportation agencies have had to discontinue the practice of part-time shoulder use because of safety concerns.

The American Association of State Highway and Transportation Officials’ Highway Safety Manual (HSM) is a tool that officials can use to help estimate the predicted highway crash frequency and severity as a result of a project. Currently, the HSM is not capable of analyzing part-time shoulder-use scenarios during the period of time when traffic is using the shoulder. However, the current edition of the HSM can inform project decisions by approximating the crash frequency of a freeway widening project that would result in narrower lanes or shoulders. Because narrower lanes and shoulders have a negative effect on safety, officials can use the HSM to estimate crash frequency and severity of a part-time shoulder alternative during the periods of the day when the shoulders are not being used as a travel lane. This comparison underscores the clear safety advantages of providing a full-width shoulder under less congested conditions in lieu of a freeway widening project that results in narrower lanes and shoulders.

Excerpt of Questions Covered in
Use of Freeway Shoulders for Travel
Roadway Design Implementation
  • Has vertical clearance under bridges been
    checked?
  • Have drainage patterns been checked?
  • Has stopping sight distance been checked on curves adjacent to barriers?
  • Have fixed object offsets been checked? Guardrails, signs, and other objects may need to be moved farther away from the roadway.
  • Will safety turnouts be provided, and have locations been established?
  • Are ramps taper-style or parallel-style, and will any need to be modified?
  • Is a Manual on Uniform Traffic Control Devices request to experiment necessary?
  • Are stakeholders, such as police and emergency responders, engaged?
  • Have State-specific legal issues, such as laws prohibiting driving on the shoulder, been addressed?
  • For bus-on-shoulder implementation, is driver training occurring?
  • Is a public outreach plan established?
Planning and Preliminary Engineering Maintenance and Operations
  • Will physical roadway conditions permit shoulder use?
  • Is the shoulder pavement strong enough to carry traffic?
  • Will the right or left shoulder be used?
  • If an area has a congestion management process, is shoulder use a compatible strategy?
  • Is there a plan for plowing snow from the shoulder?
  • Is there a maintenance plan for aggressive debris removal from the shoulder because it will be used for
    travel at times?
  • What specific actions will occur each time the shoulder is opened or closed?

In time, transportation agencies will have a more substantive understanding of the safety effects of part-time shoulder use. Until then, surrogate methods of assessing safety can help inform agencies about whether and how to implement this practice. For example, reducing congestion by allowing part-time shoulder use for all or most vehicles enables greater headways between vehicles and reduces stop-start activity that contributes to rear-end crashes. However, it may require compromises of other geometric design elements known to adversely influence crashes, such as effective shoulder width remaining when the shoulder lane is open to traffic or lateral offset to roadside features (such as median barriers and guardrails). It is important that FHWA continue monitoring the safety performance of part-time shoulder-use installations to build the body of safety knowledge.

FHWA’s Guide on Part-Time Shoulder Use

FHWA developed its guide on part-time shoulder use in response to increasing local, State, and Federal legislative interest in shoulder use and performance-based practical design. Efforts promoted by States and FHWA to evolve the highway design process from one driven primarily by design criteria to one that considers cost-effectiveness and systemwide performance also fueled the guide’s development.

FHWA developed the guide based on interviews with its own subject matter experts and staff from agencies that have deployed shoulder--use treatments. The guide developers also researched appropriate analytical techniques, including those to estimate the safety and operations effects.

The guide provides information on all phases of the life cycle of a proposed shoulder-use project, including planning, environmental considerations, design, operations, and maintenance. It also covers a range of issues--from costs to design considerations to maintenance--that can help agencies advance shoulder-use concepts in their States in a more consistent manner. Planners and designers can use the guide to help address a number of questions at various stages in the project development process. However, knowledge gaps remain in some areas of the guide, such as estimating induced traffic and air quality impacts, estimating effects on crashes, and determining the optimal thresholds for opening up a shoulder for travel to maximize operational and safety performance. More experience with shoulder use and additional research is needed.

Morning Peak Period Speeds Northbound on SH 161 in Irving, TX
Line graph. This graph shows the differences in average speeds during the morning peak period before and after implementing part-time shoulder use on SH 161 in Irving, TX. The vertical axis is labeled speed in miles per hour, and it ranges from 0 to 80 miles per hour (0 to 129 kilometers per hour) in increments of 10. The horizontal axis is labeled time, from 6:00 a.m. to 9:30 a.m., in 15-minute increments. The average speeds before implementation dropped significantly from nearly 70 miles per hour (113 kilometers per hour) at 6:15 a.m. to about 25 miles per hour (40 kilometers per hour) at the slowest around 8:00 a.m. Average speeds picked up after 8:15 a.m., returning to around 65 miles per hour (105 kilometers per hour) by 9:30 a.m. After implementation of part-time shoulder use, average speeds stayed more consistent, between nearly 70 miles per hour (113 kilometers per hour) and about 55 miles per hour (89 kilometers per hour) at the slowest, around 7:45 a.m.

The average speeds for each 15-minute interval in the 4-hour morning peak period improved drastically after implementation of part-time shoulder use.

The guide also includes photos and illustrations depicting signing and design scenarios. In addition to these components, the guide provides a list of most known applications of shoulder use in the United States and presents several case studies of successful applications. What follows are two examples of applications of part-time shoulder use that mitigate different types of congestion.

Texas Addresses Weekday Congestion

In September 2015, the Texas Department of Transportation(TxDOT) and the Regional Transportation Council opened a shoulder lane for travel during peak periods on State Highway (SH) 161 in Irving. The 3-mile (4.8-kilometer) stretch of highway between SH183 and SH114 has two general purpose lanes in each direction and connects with the President George Bush Turnpike, operated by the North Texas Tollway Authority, which has three general purpose lanes in each direction. The drop from three lanes to two creates a bottleneck that recurrently causes speeds to drop during peak periods.

The interim phase of implementation consisted of shoulder restriping to enable three general purpose lanes to operate during peak travel periods. The project team restriped the existing inside shoulder and main lanes to provide two general purpose lanes and an interim operational travel lane during morning and evening peak hours of travel (in each direction). The project also incorporated a variety of operational improvements, including closed circuit cameras, improved lighting, and strategically pre-positioned tow trucks, to ensure that the lanes function safely and efficiently. The staged tow trucks assist with opening and closing the lanes, while also providing expedited incident management.

The shoulder-use project improved travel speeds immediately. The average northbound (morning) rush-hour speed on the freeway for 8 workdays before implementing shoulder use was 30.7 miles per hour, mi/h (49 kilometers per hour, km/h). After opening the inside shoulder, the average northbound speed in the first 8 days increased to 66 mi/h (106 km/h). Speeds in the southbound direction during the evening peak period experienced similar improvements.

“The SH 161 shoulder lanes have achieved their goal of improving traffic flow through a major bottleneck,” says Kelly Selman, P.E., an engineer with the TxDOT Dallas District. “TxDOT is pleased with their performance. Where appropriate, TxDOT will consider implementing them in other areas.”

I–70 Performance Improvements (Georgetown to Veterans Memorial Tunnels)
Bar graph. This graph compares the total delay, average travel time, and number of hours with greater than 30 minutes of delay for the section of I–70 from Georgetown to the Veterans Memorial Tunnels before and after implementation of the Mountain Express Lane outside Denver, CO. The chart displays data collected from 10 a.m. to 10 p.m. on Sundays, comparing the winter before the Mountain Express Lane was constructed and the winter after the Mountain Express Lane was constructed. Before shoulder use the total delay was 35.02 hours, the average travel time per vehicle was 26.47 minutes, and there were 24 hours with greater than 30 minutes of delay. After implementing part-time shoulder use, the total delay was 12.62 hours, the average travel time was 19.57 minutes, and there was 1 hour with greater than 30 minutes of delay.

Note: The Mountain Express Lane stretches from the Veterans Memorial Tunnels to the U.S. 40 interchange (Empire Junction), but travel time and delay were studied on a slightly longer corridor from the Veterans Memorial Tunnels to Georgetown.

More recently, traffic managers have observed increasing traffic volumes on SH 161, which may dampen some of the speed gains. One possible cause for the increase in traffic is diversion from parallel routes to SH 161 because it now provides a more reliable trip. The North Central Texas Council of Governments and the Texas A&M Transportation Institute are finalizing the collection of data for a study that will provide more insight into speeds and volumes, and information on origins and destinations.

Colorado Takes on Seasonal Weekend Congestion

The Colorado Department of Transportation (CDOT) employs part-time shoulder use on I–70 eastbound between the U.S. 40 (Empire Junction) Interchange and the Veterans Memorial Tunnels (formerly the Twin Tunnels) west of Denver. Referred to as the Mountain Express Lane, the project officially opened in December 2015 and operates about 100 days out of the year. When open, the shoulder lane is tolled.

The Mountain Express Lane differs from most shoulder-use projects because it does not address daily peak period congestion. Instead, the project targets seasonal congestion on the weekends and on holidays related to tourism and recreation. This section of I–70 experiences high congestion on Fridays (westbound) and Sundays (eastbound) while travelers are on their way to and from the mountains west of Denver.

In addition to safety and mobility challenges, other factors made traditional roadway widening problematic. For example, this section of I–70 passes through sensitive natural environments and historic communities. CDOT needed to preserve these areas while accommodating the recreational activities that attract many travelers to Colorado.

“Part-time shoulder running is a perfectly matched operational solution to the unique patterns and problems of our intense recreational traffic demands,” says Ryan Rice, director of Transportation Systems Management & Operations at CDOT. “We have seen high benefit relative to the cost of the project.”

Compared to the winter season prior to implementation (December 2014–March 2015), the Mountain Express Lane has reduced the total amount of delay (evident during the first winter season after implementation December 2015–March 2016). This improvement is especially noteworthy considering that throughput volumes increased an average of 14 percent despite 12 percent more snowfall the season following implementation. Drivers have experienced improved average travel times, reduced delays, and significantly less time spent in excessive traffic (delay greater than 30 minutes).

“The I–70 Mountain Express Lane has impacted the frontage [local county] roads, which aren’t as congested with people getting off the highway,” says Megan Castle, communications manager with CDOT’s High-Performance Transportation Enterprise. “Business is up and people are getting into their communities.”

FHWA’s Next Steps

With the completion of FHWA’s Use of Freeway Shoulders for Travel: Guide for Planning, Evaluating, and Designing Part-Time Shoulder Use as a Traffic Management Strategy, agency officials now plan on developing additional guidance, sharing best practices, providing outreach through workshops and webinars, and continuing to collaborate with national organizations to raise the collective understanding of part-time shoulder use.

FHWA also will undertake additional research on the crash and safety implications of part-time shoulder use and the optimal conditions to open shoulders dynamically. The ultimate goal of the research, collaboration, and additional outreach is more efficient use of the Nation’s urban freeways.


Jim Hunt, P.E., is a transportation specialist in FHWA’s Office of Operations. He manages projects related to operations planning and design and active transportation and demand management. Hunt has a B.S. in engineering from Hofstra University and an M.S. in transportation planning from Iowa State University.

Pete Jenior, P.E., P.T.O.E., is a senior engineer with Kittelson & Associates, Inc., in Baltimore, MD. He was the lead author of FHWA’s guide on part-time shoulder use and has led numerous traffic engineering and safety projects in the mid-Atlantic and beyond. Jenior has a B.S. and an M.S. in civil engineering, both from Georgia Tech.

Greg Jones is a transportation operations specialist who splits his time working with the FHWA Resource Center in Atlanta, GA, and the Office of Operations in Washington, DC. He has worked for FHWA since 1984. Jones provides national technical support in the areas of congestion pricing, managed lanes, active transportation and demand management, emergency transportation operations, intelligent transportation systems, freeway management, and regional operations partnerships. Jones has a B.S. in civil engineering from the University of Tennessee.

For more information, see www.ops.fhwa.dot.gov/publications/fhwahop15023 or contact Jim Hunt at 717–221–4422 or jim.hunt@dot.gov.

 

 

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101