Search Google Appliance

Group Stations

Group count stations by pattern types. For example, commute, non-commute and mixed.1 

There are many ways to group count stations, including by visual interpretation of graphs, statistically based criteria2  and cluster analysis.3  On this page we present the simplest method, grouping by visual interpretation of graphs. This is also known as the traditional or intuitive approach.  

Below is an example from Colorado of how stations can be grouped by the monthly and daily patterns.4  In the example, the counters are divided into two groups:

  • Commute
  • Non-commute

Commute locations typically have higher counts on weekdays then weekends and show weekday peaks during morning and evening commute hours. Non-commute locations may have higher weekend counts than weekday and typically have one peak in the middle of the day. Here we describe patterns as “non-commute” instead of “recreational” or “utilitarian” because trip purpose is hard to determine from count data. For example, shopping patterns and recreational patterns can be similar.

The factors computed for each group are shown as the thick black line. Each of the other lines shown represent a unique permanent count station.  

Monthly Patterns (Source: Colorado Department of Transportation)

Daily Patterns (Source: Colorado Department of Transportation)

Hourly Patterns for Workdays (Sources:  City of Boulder5  and Douglas County, Colorado)


Miranda-Moreno and others have created a statistically-based method for grouping bicycling patterns using data from 40 North American locations.  They classified these locations into four groups:  utilitarian, mixed utilitarian, recreational and mixed recreational. 

In 2013, Colorado Department of Transportation released a report detailing the process of inventorying, grouping, and computing seasonal adjustment factors for the state. Locations are classified into three groups:  mountain non-commute, urban plains non-commute and commute.

A 2012 Colorado Department of Transportation report discusses the topic of grouping sites and more generally recommends three basic groups: commute, non-commute and mixed. 

Cluster analysis offers a more statistically based option to grouping locations based on daily and monthly factors.  For motor vehicles, the Traffic Monitoring Guide Section 3.2 discusses the pros and cons of cluster analysis compared to other methods for grouping stations, and Appendix G gives an example of cluster analysis applied to North Carolina motor vehicle data. For non-motorized traffic, the 2013 report for the Colorado Department of Transportation (page 95) includes an example of how cluster analysis was applied to bicycle and pedestrian count data.




In the previous step (Step 3), existing non-motorized data was used to determine the traffic patterns that are to be monitored. In Step 4, this information is used to establish unique traffic pattern groups that will be used as the foundation for the monitoring program.

 In some cases, non-motorized count data may not be available in Step 3 to determine the most likely traffic pattern groups. In these cases, previous analyses of non-motorized data from previous studies or of similar locations should be used as a starting point. Once more non-motorized data is gathered in your region, these traffic pattern groups can be refined based on your local data.

Previous (but limited) research indicates that non-motorized traffic patterns can be classified into one of these three categories (each with their own unique time-of-day and DOW patterns):

  • Commuter and work/school-based trips – typically have the highest peaks in the morning and          evening;
  • Recreation/utilitarian – may peak only once daily, or be evenly distributed throughout the day;
  • Mixed trip purposes (both commuter and recreation/utilitarian) – has varying levels of these two different trip purposes, or may include other miscellaneous trip purposes.


1Turner, S., Qu, T., & Lasley, P. (2012). Strategic Plan for Non-Motorized Traffic Monitoring in Colorado (pp. 99). College Station, TX: Texas Transportation Institute.

2Miranda-Moreno, L. F., Nosal, T., Schneider, R. J., & Proulx, F. (2013). Classification of bicycle traffic patterns in five North American Cities. Paper presented at the Annual Meeting of the Transportation Research Board, Washington, DC.

3Federal Highway Administration. (2013). Traffic Monitoring Guide. Washington, DC: U.S. Department of Transportation.

4Nordback, K., Marshall, W. E., & Janson, B. N. (2013). Development of Estimation Methodology for Bicycle and Pedestrian Volumes Based on Existing Counts (pp. 157). Denver, CO: Colorado Department of Transportation (CDOT).

5Nordback, K. L. (2012). Estimating Annual Average Daily Bicyclists and Analyzing Cyclist Safety at Urban Intersections. Ph.D., University of Colorado Denver, Denver.