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-RD-98-166
Date: July 1999

Guidebook on Methods to Estimate Non-Motorized Travel: Supporting Documentation

2.0 Documentation of Methods


Demand Estimation

2.1 Comparison Studies

Descriptive Criteria: What is it?


Box with an x insideBicycle Box with an x insidePedestrian Box with an x insideFacility-Level Box with an x insideArea-Level

Authors and Development Dates:

Hoekwater (1978); Lewis and Kirk (1997); Wigan (1998)


The simplest form of demand forecasting, comparison studies track bicycle or pedestrian travel levels before and after a change (such as a facility improvement), or compare travel levels across facilities with similar characteristics. The results of a comparison study can be used to predict the impacts on non-motorized travel of a similar improvement in another situation, assuming that all other influencing factors are roughly the same between the two situations.

Two basic types of comparison studies are discussed here:

1. Before-and-after studies. These are based on counts of users both before and after an improvement. The change in users is assumed to be related to the improvement.

2. Similar conditions studies. These studies use counts and/or user survey data from existing facilities, sometimes combined with data on the population in the surrounding area, to estimate the potential number of users on a similar existing or proposed facility. Two examples are documented here:

Lewis and Kirk (1997): To forecast travel on two proposed rail trails, employees at the Central Transportation Planning Staff (CTPS), the regional transportation planning agency for the Boston, Massachusetts metropolitan area, examined a comparable existing rail trail using counts of trail users and travel survey data for area residents.

Wigan, Richardson, and Brunton (1998) compared the characteristics of users and the surrounding population on two existing facilities in Australia, and identified factors that could account for differences in usage levels on the two trails.


Before and After Studies:

These have been widely used in Europe to assess the mode choice impacts of programs to improve bicycle and pedestrian facilities. Some studies have focused on the change in mode split for an urban area as a whole, after a city-wide program of improvements. Others have focused on specific facilities, conducting user counts both before and after an improvement to the facility. An example of the latter is given in Hoekwater (1978), who compared bicycle traffic before and after the addition of bicycle lanes in the Netherlands. In addition to counts on the facility itself, counts were also performed on parallel facilities to attempt to estimate how much traffic was diverted as compared to actual new riders.

Similar Conditions Studies:

Lewis/Kirk: To estimate the potential usage of a proposed rail trail in Massachusetts, planning staff conducted bicycle counts on an existing trail which has characteristics similar to the proposed facility. These counts were then factored based on the ratio of total population within corridors surrounding the two facilities to predict total trips on the proposed facility. Total volumes were distributed throughout the proposed corridor based on the population of communities along the corridor. An alternative method was also applied in which forecasts for the proposed trail were factored by the ratio of bicycle commuting mode share in the two corridors, as determined from census data.

Wigan/Richardson/Brunton: Two existing facilities in Australia were compared: Lower Yarra and Maribrynong trails. A survey of trail users was conducted regarding mode of access to the trail, access distance, personal characteristics, etc. Data on population in the surrounding area were also analyzed using GIS techniques. The characteristics of users and the surrounding population were both used to compare the two trails. The results indicate that the Lower Yarra trail attracts more users from a wider range of distances than the Lower Maribrynong. The authors concluded that with better signage, improved linkages and promotional efforts for the Lower Maribrynong facility, this trail could see higher usage rates, similar to the Lower Yarra trail. The model gives an estimate of the potential users of the Lower Maribrynong trail (see also GIS, Method 1.18)

Calibration/Validation Approach:

Not applicable.

Inputs/Data Needs:

Before and After Studies:

These require counts or mode split data from the facility or area before and after the improvement. Counts should also be obtained from parallel facilities to determine to what extent a change in traffic on a facility is due to diversion as compared to new users. Ideally, counts would also be performed over the same time period in other control areas that are unaffected by the improvements to determine whether traffic levels may have changed for reasons unrelated to the facility addition. Enough counts should be performed so that the statistical significance of any observed change in traffic can be verified.

Similar Conditions Studies:

Lewis/Kirk: The comparison approach requires bicycle counts for the existing facility and population data for the surrounding areas of both the existing and proposed facilities.

Wigan/Richardson/Brunton: The technique uses survey results taken from the VITAL project, which is a continuing household interview survey in Melbourne that covers the origins and destinations of bicycle travel. Another survey was conducted by Melbourne Parks and Waterways (MPW). A third survey questioned only users at the Lower Yarra and Lower Maribrynong trails.

The main inputs that are used from these surveys include:

  • Trip length distributions;
  • Numbers of patrons from different postal code areas (equivalent to ZIP codes);
  • Populations in postal code regions at various distances from the trail; and
  • Distances from the trail to the different postcode area centroids.

Potential Data Sources:

Lewis/Kirk: This approach could use localized mode split information to provide better accuracy.

Computational Requirements:

Minimal computations are required.

User Skill/Knowledge:

Minimal skill is needed.


Unless very carefully designed, comparison studies may not control for other factors unrelated to the facility improvement which may affect usage levels, such as weather conditions on the day of the count, improvements to parallel facilities, etc. Also, the specific factors causing differences in impacts for different facilities may not be readily explained, or may only be described qualitatively. Because of possible differences in the situation, transferring results from one situation to another may lead to incorrect usage forecasts. Therefore, the comparison method is best used in conjunction with a qualitative assessment of environmental factors to gauge an approximate level of impact, rather than for quantitatively predicting an actual change in usage levels.

Wigan/Richardson/Brunton: This technique assumes that the main reason that the two trails have different user rates is because of inadequate signage and connections at the trail with the lower usage rate. The differences in level of usage between two apparently similar trails illustrates why care must be taken in using a simple comparison approach to predict demand.

Facility Design Factors:

Lewis/Kirk and Wigan/Richardson/Brunton: This approach requires planners to compare facilities that are similar in type and length.

Photo of a bike lane
Figure 2.1 A similar conditions study uses data from an existing facility, such as the bike lane shown here, to estimate the potential number of users on a proposed facility.


Output Types:

These methods supply the planner with rough estimates of bicycle usage for proposed facilities.

Real-World Examples:

Hoekwater: Counts of bicycle traffic were performed before and after the addition of bicycle lanes at two locations in the Netherlands. Counts were also performed on parallel facilities to attempt to estimate diversion vs. new riders. In one location, bicycle counts increased by 30 to 60 percent on the route with a slight increase on parallel routes. For a different location, bicycle traffic on the route also increased but there was some decrease on parallel facilities; the authors concluded that roughly two-thirds of the increase in bicycle traffic came from parallel routes and one-third from new trips.

Lewis/Kirk: Bicycle counts from the Minuteman Commuter Bikeway in the Boston area were used to predict the bicycle volumes for the proposed Central Massachusetts Rail Trail Bikeway. Weekend and peak-hour weekday counts were taken at four locations along the 48-km long Minuteman facility. Weekday counts then were estimated by assuming that the peak period represents 10 percent of the daily usage. An average of the four survey sites was taken to obtain a weekday estimate of 1,600 and a weekend estimate of 3,400 users. The population of the Central Massachusetts Bikeway corridor, at 138,556, is 80 percent that of the Minuteman corridor's population of 172,606. Usage estimates for the Central Massachusetts Bikeway, therefore, total 1,280 for the weekdays and 2,720 for the weekends. The volumes then were distributed along the corridor according to population share.

The same approach was used on the Norwottuck Rail Trail, which is at the western end of the proposed Central Massachusetts Bikeway. Weekday bicycle volumes total 700, weekend/holiday volumes total 1,900 and the regional population is 69,000. The surrounding area for the proposed facility has two times the Norwottuck population so the daily estimates total 1,400 per weekday and 3,800 per weekend/holiday.

Wigan/Richardson/Brunton: The Lower Yarra and Lower Maribrynong trails differ in that the latter lacks linkages and promotional opportunities. Since the populations surrounding the two trails are similar as well as each trail's length, the user rates also should be similar. The Lower Maribrynong trail has a significantly higher usage rate. The authors hypothesized that the proportional difference in user rates reflects the potential usage that could occur on the Lower Yarra trail.


Cathy Buckley Lewis, Central Transportation Planning Staff, 10 Park Plaza, Suite 2150, Boston, MA 02116.

Marcus Wigan: Oxford Systematics, GPO Box 126, Heidelberg, Victoria, Australia 3084.


Hoekwater, J. Bicycle Routes in the Hague and Tilburg. Published in Bicycling as a Mode of Transport: Proceedings of a Symposium held at the Transport and Road Research Laboratory, Crowthorne, U.K. (TRRL Supplementary Report 540), October 1978.

Lewis, Cathy Buckley and James E. Kirk, Central Massachusetts Rail Trail Feasibility Study, Central Transportation Planning Staff, Boston, MA, April 1997.

Wigan, Marcus, Anthony J. Richardson and Paris Brunton. Simplified Estimation of Demand for Non-motorized Trails Using GIS, Transportation Research Board, Preprint #981203, 1998.

Evaluative Criteria: How Does It Work?


Examples where these methods had been validated in practice were not identified.

Use of Existing Resources:

Lewis/Kirk: These methods are simplified approaches that capitalize on the use of existing, albeit somewhat limited, resources.

Wigan/Richardson/Brunton: Since it is difficult to quantify the benefit of signage and linkage improvements as well as the impact of promotions on the usage of an existing facility, this method attempts to calculate the benefits using a comparable facility that has more sophisticated signage, linkages, and promotional opportunities.

Travel Demand Model Integration:

Not applicable.

Applicability to Diverse Conditions:

The methods use survey results that vary depending on the situation.

Usage in Decision-Making:

The methods provide a rough estimate concerning the demand that is likely to occur on proposed facilities.

Ability to Incorporate Changes:

The methods are able to incorporate changes into the analysis since the data inputs and computations are not complex.


Lewis/Kirk: The method is easy to use because it uses simple and widely available data.

Wigan/Richardson/Brunton: This approach is somewhat more complex than Lewis/Kirk, requiring the use of GIS and local travel surveys for analysis.



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