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

Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

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

This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-HRT-05-138
Date: July 2006

Shared-Use Path Level of Service Calculator

A User's Guide

PDF Version (1298 KB)

PDF files can be viewed with the Acrobat® Reader®




Stone Creek Trail is located in the heart of Rockton, an old, eastern industrial city of about 200,000 residents. The 25-year-old trail is located in a linear park that was developed along a large stream, Stone Creek. Downtown Rockton grew at the confluence of Stone Creek and the Rocky River. The trail is 8.8 km (5.5 mi) long and connects a number of the city's finest neighborhoods with downtown. Along it are located an elementary and junior high school, a major playground, a small hospital, and a small liberal arts college. When it was last repaved 15 years ago, it was widened from the previous 1.8-m or 2.1-m (6.0-ft or 7.0-ft) width to 2.4 m (8.0 ft) throughout.

In recent years, the city council and managing agency, Rockton Department of Public Works and Parks (DPW&P), have been receiving pressure to widen the trail even further. That pressure was coming from a number of community interests:

  • The Tourism Council and Historic Preservation League, which have been successful in boosting the city's economy through historic preservation efforts focused on the old milling industry, the reuse of warehouses as small-scale breweries, and the town's access to outdoor recreation activities in the nearby mountains. They wanted more and better outdoor recreation activities to be offered in town to keep tourist dollars coming into the city.
  • A group of parents who felt that the trail was unsafe because of its age, rough surface, and narrow width. In recent years, there have been a couple of bad crashes between trail users; these involved youths. Skaters, skateboarders, push scooters, strollers, and dog walkers are all common on the trail, in addition to bicycles and pedestrians.
  • A local environmental organization that recently joined with the local bicycle club to promote in-line skate and bicycle commuting to reduce downtown parking demand and keep the air clean. The trail seemed too narrow to accommodate more nonmotorized wheels, however, along with all of the pedestrians, kids, and a growing crowd of fitness walkers organized by the hospital.

The Stone Creek Park Committee, a longstanding advisory committee for DPW&P, favors some widening but is wary of making the trail so wide that too many trees are lost and the 100-year-old park will become a speedway for wheels. There is concern that too many wheeled users will make the park and trail too scary for senior citizens and others (mostly walkers) who have cared for it for many years. Segregation of users is one possibility they are considering, but they are not sure users will comply.

The question is this: How much more width is enough to satisfy the demand and how much is too much? An initial staff survey determined that, in most parts of the park, enough space exists to accommodate a trail of up to 4.3 m (14.0 ft) wide. The DPW&P staff and the committee would like to know what other communities in similar situations have done. However, even on the Internet, they had difficulty finding another community with enough similar characteristics that had already encountered and solved a similar trail problem. Then, in response to a call to the State, the Bicycle and Pedestrian Coordinator at the State department of transportation recommended the Shared-Use Path LOS Calculator, a new FHWA resource.

Because the community had no existing trail counts, nor were any useful volume counts available from other multiuse trails in the State, it was necessary to gather fresh trail user counts to develop inputs for use of the tool. Volunteers from all of the interested stakeholder groups were organized, and they followed the trail count guidelines provided in the SUPLOS User's Guide. To meet the data needs necessary for the results they wanted, the following data collection scope was adopted:

  • Collect data on three warm-weather Spring days in different weeks of May and early June (Friday, Saturday, and Sunday).
  • Take counts to cover the following time periods–three 2-hour periods on Friday
    (7–9 a.m., 12–2 p.m., and 4–6 p.m.) and 10 a.m.–4 p.m. on Saturday and Sunday.
  • Take counts at five locations along the trail, at midpoints between major access points.
  • Count children on skateboards, push scooters, and skates as pedestrians, the user type with the most similar speed profile.

Figure 2, a screen capture from the Shared-Use Path LOS Calculator for this case, shows the results of the average weekend, one-way, per-hour, volume counts, and mode splits. Friday volumes were found to be about 50 percent lower and so were not used for the LOS calculations. As noted above, for data collection and analysis purposes, the trail was divided into five segments. Trail segments 1 and 2 scored F and E levels of service, respectively, while the other segments scored D, confirming suspicions that peak-hour conditions were deteriorating. It was easy for the staff and committee to imagine even worse conditions on summer weekends when more tourists were in town.

Before various widths were tested, participants agreed that if the trail were widened, it should be done not only to improve conditions for existing users, but also to secure some additional capacity for future growth in trail use.

1 ft = 0.3 m


Figure 2. Screen capture. Existing conditions LOS analysis. This screen capture from the Shared-Use Path LOS Calculator shows the results of the average weekend, one-way, per-hour, volume counts, and mode splits for the case being studied in the text. Each of five rows of data, displayed in tabular format across the row, is sorted for the trail segment named in the row and is followed by a button on the screen reading, "Click Here for Default Mode Split."

Figure 2. Screen capture. Existing conditions LOS analysis.

The following general LOS goal and design policy was adopted:

  • The LOS goal should be a high C or low B to ensure space for additional users.
  • A flexible design approach should be used to meet the needs of the users that are attracted to each segment. For example:
    • Segment 1 near downtown should have room to attract more users and provide a trail experience that tourists will want to repeat.
    • Part of segment 3 near the school and a popular playground attracts more youth and should be designed to support trail etiquette education.
    • Segment 5, which runs along the college campus, attracts more skaters and runners.

Using these guidelines, a variety of widths was tested in the calculator to see which achieved the desired levels of service. Figure 3, a screen capture from the Shared-Use Path LOS Calculator for this case, shows the widths and resulting LOS scores that were finally selected.

Figure 3. Screen capture. LOS for selected design widths. This screen capture from the Shared-Use Path LOS Calculator shows the widths and resulting LOS scores that were selected for each row of data for the five trail segments used in the previous figure. Each row is again followed by a button on the screen reading, “Click Here for Default Mode Split.”

Figure 3. Screen capture. LOS for selected design widths.

1 ft = 0.3 m

Working with a trail design consultant, the widths in figure 3 and the general guidelines were used to develop five different cross-sections. Designs for segments 2 and 4 varied only in the centerline treatment, as a means to further test users' response. Table 9 describes the final segment designs selected.


Table 9. Selected cross sections

Segment Number/Distance Width Shoulder Rationale
1 / 0.75 mi 13 ft, centerline Grass The downtown section serves a heavy mix of pedestrians, tourists, and bicycle commuters.
2 / 1.75 mi 11 ft, centerline only on curves Grass This section weaves through an especially cherished part of the park with many old trees and scenic settings along the stream; trail expansion needs to be kept to a minimum.
3 / 0.75 mi 10 ft, centerline 2 ft of rolled stone dust on each side Because of its proximity to the schools, playground, and adjacent neighborhoods, this section should include a centerline and shoulders to facilitate youth education about trail-sharing etiquette.
4 / 1.00 mi 11 ft, centerline Grass This requires a transition design between segments; a separate soft surface jogging path could be added in the future.
5 / 1.25 mi 12 ft, centerline Grass Width could be added to serve in-line skaters and higher overall user volumes from the college; segments 4 and 5 can be promoted as ideal for skating as a way to draw skaters to this section of trail. A separate soft-surface jogging path could be added in the future.
1 mi = 1.6 km; 1 ft = 0.3 m



The De Soto River Trail is a 7.2-km (4.5-mi) segment of multi-use trail proposed for the downtown waterfront in the city of New Metropolis. It will extend the existing De Soto River Trail, a 24.2-km (15.0-mi) trail system, from a nearby suburb into the heart of the city. New Metropolis is a city of 500,000 people in a metropolitan area of approximately 1.5 million.

The Waterfront Trail has been an idea in the city's comprehensive plan for many years; however, no action was taken until a waterfront revitalization effort brought the idea to prominence. Currently, a detailed waterfront redevelopment plan is underway and the city's Office of Planning wants to ensure that the trail component of this plan establishes appropriate path design guidelines. Because it is the first major shared-use path to be built in the city for 25 years, the city's transportation and park departments have no pre-existing guidelines and little trail experience. An advisory committee has been formed to assist the Office of Planning and revitalization consultants with the trail plan.

The major question facing the Waterfront Trail Advisory Committee (WTAC) is this: How wide must the trail be to serve the volumes and diverse user groups expected? A WTAC member suggested using a new resource developed by USDOT-funded research called the Shared-Use Path LOS Calculator. The following discussion describes how WTAC used the calculator to inform their design process.

After reading in the User's Guide how the LOS concept should be applied to a trail with various access points, WTAC recognized that a LOS evaluation should be performed for three separate segments:

  • Segment A–3.2 km (2.0 mi) of path in newly acquired public lands that pass through a natural area slated to become an urban conservation area, with limited park development. This segment has no access points between the existing trailhead in Jefferson County and Reed Mill Park.
  • Segment B–a 2.4-km (1.5-mi) segment that passes through Reed Mill Park, the oldest park in the city, which is slated for revitalization and will serve as a major trailhead area.
  • Segment C–a 1.6-km (1.0-mi) segment that will be located in a new waterfront park to be built as an extension of Reed Mill Park. The new park is very close to downtown New Metropolis, which is expanding toward the river, now that upgraded flood control infrastructure is in place. From downtown to Water Street, the old warehouse district is being redeveloped with commercial and residential uses. The park will be about 0.40 km (0.25 mi) wide along the shoreline. Future expansion of the path 4.8 km (3.0 mi) down-river may be feasible if and when an old railroad line is abandoned.

Given the history of the De Soto River Trail in neighboring Jefferson County, it seemed logical to employ trail usage data from that trail, at least as a starting point for volumes and user mixes that might be expected on the city's section of the trail.

The existing De Soto River Trail is a 3.0-m (10.0-ft) wide asphalt path. Counts taken the previous year show that the trail is averaging about 250 users per hour (total 2-way volume). The mode split for various trail user types was as follows:

  • Bicycles–51 percent.
  • Pedestrians–16 percent.
  • Joggers–19 percent.
  • Skaters–10 percent.
  • Other–4 percent.

The volume data provided by Jefferson County were generated by taking counts every hour between 12:00 p.m. and 4:00 p.m. on one nice-weather weekend in both the spring and fall. The location where these counts were taken is just outside the city, along a 2.4 km (1.5-m) segment between a large trailhead near the city limits and a Jefferson County regional park facility in Jeffersonville (a suburb of 85,000 people). Based on this volume and mode split data, the calculator gave the existing De Soto River Trail in Jefferson County a LOS score of 2.81, which is a high D.

Because the Waterfront Trail will be located in the city, which has higher population density and fewer trail opportunities than suburban Jefferson County, WTAC wants a plan to accommodate higher future-use levels. As a result, WTAC developed the following adjustments to the Jefferson County baseline counts, upon which to base assumptions for LOS analysis of segments A–C:

  • Each trail segment should be developed to meet a low-C LOS for projected 10-year volumes and mode splits as defined below.
  • Each trail segment has a different character and context; therefore, the 10-year projections and mode splits should vary accordingly:
    • Segment A, plus 20 percent of baseline, with fewer pedestrians and more bicycles.
    • Segment B, plus 50 percent of baseline, at the same mode split.
    • Segment C, plus 100 percent of baseline, with more pedestrians and fewer bicycles.
  • To accommodate additional increases in usage beyond these estimates, the trail should be designed with right-of-way reserved for expansion.
  • It is assumed that, in the first few years after a trail segment is built, the trail will experience user volumes lower than the baselines, and thus levels of service during those periods will be above C.
  • If the new trail is very popular from the start, it may need to exist at an LOS D for a period of years until expansion can be funded.

Using these assumptions, the Office of Planning staff developed a series of volume and mode split estimates to use as inputs into the Shared-Use Path LOS Calculator to develop potential trail design cross sections for segments A–C of the Waterfront Trail, as shown in table 10.


Table 10. Volume and mode split estimates to test in the calculator.

Segment Assumptions(conditions) One-way volume per hour Bikes Peds Runners Skaters Child Bikes
Existing trail in Jefferson County Experiencing some user conflicts 125.0 51% 16% 19% 10% 4%
Segment A Same as above 150.0 56% 11% 19% 10% 4%
Segment B Same as above 187.5 51% 1% 19% 10% 4%
Segment C Same as above 250.0 40% 25% 20% 13% 2%


Various widths were tested in 0.2-m (0.5-ft) increments, starting at 2.9 m (9.5 ft). With each width change, the LOS score was observed to determine when a low-C (3.0-3.2) was achieved. It took 6.1 m (20.0 ft) width to get a low-C for segment C. Because 6.1 m (20.0 ft) was considered too wide a footprint, a multiple treadway design was considered that would provide pedestrians and runners with their own dedicated treadways. An 80 percent compliance rate was used for pedestrians and runners, and the user volume and mode splits were reduced accordingly. Figure 4, a screen capture from the Shared-Use Path LOS Calculator for this case, shows the results from the calculator after revising segment C for multiple-treadway options.

Based on these results, WTAC adopted the following width design guidelines for each trail segment:

  • Segment A: Build a 3.4-m (11.0-ft) asphalt path, with AASHTO-recommended 0.6-m (2.0-ft) shoulders to be planted in grass. In natural areas, boardwalk sections may be reduced to a total clearance between railings of 3.7 m (12.0 ft).
  • Segment B: Build a 3.7-m (12.0-ft) asphalt path, with AASHTO-recommended 0.6-m (2.0-ft) shoulders to be planted in grass.
  • Segment C: For environmental and aesthetic reasons, building a path 6.1 m (20.0 ft) wide was not considered optimal or desirable. However, given the 0.40-km (0.25-mi) width of the waterfront park and the high volume that is expected in this segment near downtown, at least 6.1 m (20.0 ft) of width was determined to be desirable if it was provided in separate treadways, using space, landscaping, and elevation as buffers:
    • Closest to the river, a 1.8–2.4-m-wide (6.0–8.0-ft-wide) concrete promenade for pedestrians to stroll and observe the river (3.0–3.7 m (10.0–12.0 ft) wide for the expected highest-use segment).
    • Located away from the river, a 3.4-m (11-ft) asphalt path for bicyclists and skaters, with 0.4-m (1.0-ft) concrete shoulders and 0.6-m (2.0-ft) grass shoulders on each side (4.0 m (13.0 ft) of usable treadway) and a minimum 10.7-m (35.0-ft) buffer between path and promenade.
    • Located adjacent to the path (on the riverside), a 1.2-m (4.0-ft) soft surface trail for runners, with a minimum 1.5-m (5.0-ft) grass buffer.
    • Assuming an 80 percent compliance rate for pedestrians and runners using their separate treadways, the bicycle/skate path will provide a high LOS C (see segment C, revised, in figure 4).


Figure 4. Screen capture. LOS for 10-year projected volumes and selected design widths. This screen capture from the Shared-Use Path LOS Calculator shows the results of a revision to one of five segments studied in the text (not the same segments as for the previous two figures). The visual format is quite similar to that of the two previous figures. Each row is again followed by a button on the screen reading, “Click Here for Default Mode Split.”

Figure 4. Screen capture. LOS for 10-year projected volumes and selected design widths.

1 ft = 0.3 m

To ensure the potential to serve higher than expected volumes of trail users in the future, the following recommendations were included in the design guidelines:

  • For segment A: Identify a separate alignment for a 1.2–1.5-m (4.0–5.0-ft) hiker/runner path with a soft surface, to be developed if demand warrants. Because this is a natural area, if more capacity is ever needed, a natural surface pedestrian track should be created rather than expanding the paved treadway. To minimize the path footprint and visual impact in the natural area, the pedestrian track could be located away from the paved trail.
  • For segment B: Reserve sufficient right-of-way on the waterside of the main path to allow the development of a 1.8-m (6.0-ft) stone-dust pedestrian and runner path with a minimum 1.5-m (5.0-ft) buffer between the two paths. The second path can be added if and when additional capacity is needed. It could also be paved if a hard surface will better meet users' needs and if environmental impacts are acceptable.



Previous | Table of Contents | Next

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