TRAVEL-TIME-BASED PERFORMANCE MEASURES:
HOW THE DATA IS COLLECTED

Travel Time Data Collection Method
Travel time data are collected using a probe vehicle that travels with the flow of traffic according to the “floating car” technique. Each probe vehicle is equipped with a global positioning system (GPS) and with a data collection device (laptop or palmtop computer) that records travel times and distances at one-second intervals. For each segment, a valid sample size of travel time runs is obtained in order to calculate a significant average peak-period measurement. A segment usually begins immediately after a significant intersection and ends immediately after the next significant intersection.

The roadway monitoring captures typical traffic conditions during commute times. Roadways are monitored during weekday morning and evening peak commute periods, primarily between 6:30 AM and 9:00 AM and between 3:30 PM and 6:30 PM. Monitoring does not occur on weekends, Monday mornings, or Friday evenings; nor does monitoring occur during the peak period following or preceding a local, state, or national holiday. Monitoring is conducted during the public school year, in the spring and the fall seasons.

Processing of GPS-based travel speed data using geographic information systems (GIS) allows large databases to be manipulated easily. Furthermore, the integration of the collected information with additional geographic content allows for the production of many of the maps found in this report.

The roadway performance measures of average travel speed and delay are calculated as part of the processing of the individual samples of observed GPS data. These measures are explained next.

Roadway Travel Time Measures
In order to apply performance measures and congestion thresholds, the CMS-monitored roadways are grouped into three general categories: arterial roadways, partially limited-access roadways (which have characteristics of limited-access control, but tend to have midsegment curb cuts and slower speed limits and design speeds), and limited-access highways.

Identifying congested areas or locations presenting mobility concerns must be accomplished using slightly different thresholds of level of service (LOS) for the different categories of roadway.

Listed in Table 3.1 are the performance measures and congestion thresholds used for each type of roadway.

In general, the CMS identifies congestion on monitored roadway segments by using a combination of the following travel-time-based measures: average travel speed, speed index, and delay. These performance measures are calculated from travel time data collected at peak commute times in typical traffic conditions.

Table 3.1. Roadway Performance Measures and Congestion Thresholds

Roadway System Performance Measure: Average Observed Travel Speeds
Travel speed is a typical measure of performance for a roadway segment; for example, the Highway Capacity Manual (HCM) defines level of service on urban streets (arterial roadways) and freeways in terms of average travel speeds. The LOS corresponding to the average speeds varies, depending on the roadway classification (see Table 3.2).

Table 3.2. Level of Service Based on Average Travel Speed: Arterial Roadways

Source: 2000 Highway Capacity Manual, p. 15-3.

In order to keep the roadway classifications general and simple, for the CMS analysis the higher-speed arterials (those with some degree of limited access) and partially limited-access highways are classified as Urban Street Class I/II, whereas the remainder of the arterial roadway network is classified as Urban Street Class III.
Figure 3.2 depicts the recently monitored roadway network and indicates the CMS roadway classification. Or for a particular roadway’s designation, please refer to Table B.1.

For limited-access highways, LOS is described in the Highway Capacity Manual in terms of the flow of traffic relative to free-flow speeds.

Levels of service A, B, and C essentially describe conditions equal to or greater than free-flow speeds. LOS D describes conditions where speeds are beginning to decrease, and LOS E describes travel conditions at capacity, but with vehicle speeds that still exceed 49 miles per hour. LOS F describes conditions where traffic flow is congested.

Table 3.3 gives the specific speed ranges assigned to each LOS category, based on the HCM descriptions of LOS.

Level of Service Based on Average Travel Speed: Limited-Access Highways

Based on the 2000 Highway Capacity Manual, pp. 13-8–13-11.

Roadway System Performance Measure: Travel Speed Index
The posted speed limit is one of the factors that influence travel speeds on roadways. Thus, in order to complement the average observed travel speeds, a speed index is used to account for the speed limit factor. The speed index is simply a ratio that is calculated by dividing the average observed travel speed by the posted speed limit for that roadway segment. The index helps to determine whether a slow observed speed is caused by congested conditions or simply by a lower posted speed limit.

Roadway System Performance Measure: Delay
For purposes of CMS monitoring, delay is defined as the time a vehicle travels below 5 mph on a roadway segment (including time that the vehicle is stopped, as long as the speed has been lower than 5 mph for at least three consecutive seconds). The observed delay is closely related to “control delay” (for arterial roadways), which is the delay that occurs when a vehicle moves forward in a queue, a slow stop-and-go process.

Along most segments, delay can be attributed entirely to intersection controls (for example, traffic signals) at a segment endpoint.

Using widely accepted industry practices, an intersection with an average control delay per vehicle of more than 55 seconds is considered to be operating at LOS E. The HCM does not include a definition of a delay threshold for freeways.

The 2000 HCM strongly recommends that any analysis of signalized intersections include both a capacity analysis and an LOS analysis in order to obtain a complete picture of existing intersection operations. In other words, the CMS analysis should be viewed as a cursory assessment of signalized intersections; further data would need to be collected in order to determine the severity of problems for a specific traffic signal operation.