Congestion Management System

Chapter 2: Performance Measure Selection

2.1: Background and Federal Requirements
Federal regulations require that the CMS include performance measures that “will provide a measure of the extent of congestion and permit the evaluation of the effectiveness of congestion reduction and mobility enhancement strategies for the movement of people and goods”. In addition, the performance measures have to measure against what is the locally defined standard of congestion.

Performance measures need to be common values so that everyone who works with them (engineers, planners, politicians) will understand what they are and can communicate them with the general public. It is also ideal to select performance measures that are easily calculated using data that is already being collected because data collection is a time and resource consuming activity.

Based on these reasons, the following performance measures were chosen to evaluate the transportation system:

(1) Volume to Capacity Ratio (present and future)
(2) Roadway Corridor & Segment Travel Time (peak versus off-peak)
(3) Stopped Delay at Intersections (from Travel Time Study)

The above performance measures are the ones currently used for this report as data was the most readily available for them, but it does not preclude the addition of other measures in the future.

2.2: Definition of Local Congestion
Congestion is defined by the CMS regulations as, “The level at which transportation system performance is no longer acceptable due to traffic interference.” Levels of acceptable service vary according to factors, which include geographic location, population of area and classification of roadway.

Therefore the congestion ratings that make the most sense for the Knoxville Area CMS are to follow the guidelines in the Highway Capacity Manual to develop a Level-of-Service (LOS) based standard to measure congestion. More specific designations will be given for each performance measure, but in general the following LOS criteria will be used to define congestion for the CMS facilities and intersections analyzed:

(1) LOS D – Marginal Congestion
(2) LOS E – Moderate Congestion
(3) LOS F – Serious Congestion

It should be noted that typically a LOS D is considered acceptable for an urban area, since designing to achieve a higher LOS will usually result in too great an expense in infrastructure given right-of-way constraints. Therefore, LOS D is used here to identify conditions that are on the verge of becoming unacceptable, and merit closer future observation.

Additionally, it is widely accepted that congested conditions will likely always exist to some extent in an urbanized area due to the tendency towards an equilibrium state of people’s tolerance to travel delays to available capacity of the system. In other words, usually if capacity is significantly improved, the road will tend to become saturated once again as people shift their commuting habits, which include traveling the improved route rather than a “back” route they previously had taken, changing the time they leave home or work, or even making trips that would have been avoided altogether under previous conditions.

2.3: Performance Measure #1 – Volume to Capacity Ratio– Present and Future
The first performance measure relates congestion to the level of traffic volume versus the total capacity of the facility, otherwise known as volume-to-capacity ratio or v/c ratio. The v/c ratio provides a good indication whether the facility is congested by relating whether there is “excess” capacity available, or saturated conditions exist. A v/c ratio equal to 1.0 or greater indicates that the demand volume is exceeding the available capacity of the roadway and forced flow conditions will inevitably result, this is LOS F operation. The other categories vary slightly depending on the particular methodology from the Highway Capacity Manual that is being employed, but in general the following v/c ratios and their corresponding LOS are as follows:

(1) v/c < 0.65 = LOS A,B,C (Not Congested)
(2) 0.65 < v/c < 0.85 = LOS D (Marginal Congestion)
(3) 0.85 < v/c < 1.00 = LOS E (Moderate Congestion)
(4) v/c > 1.00 = LOS F (Serious Congestion)

The methodology used to determine the roadway segment v/c ratio and subsequent LOS is a software program that was developed by the University of Tennessee, Knoxville for use by the Tennessee Department of Transportation (TDOT). The software program is known as “EVE”, which stands for Evaluation of Efficient Roadways. The EVE software(2) incorporates the Highway Capacity Manual methodologies to automate analysis of roadway and traffic data to obtain an estimate of LOS.

The EVE program uses actual roadway data such as number of lanes and other geometric characteristics that are stored in a comprehensive database known as the Tennessee Roadway Information Management System (TRIMS) to generate specific capacities for segments of each state classified roadways. Traffic count data are also contained in TRIMS, and modifications are made to calculate passenger car equivalents based on the numbers of trucks on each segment in order to generate the total volume of vehicles using the roadway.

EVE is also able to project traffic volumes into the future using historical traffic growth trends and analyze future v/c ratios assuming no transportation improvements are made. The future analysis will be incorporated into the CMS as well to further identify facilities that may need future attention. Future years of 2010 and 2020 will be added to the monitoring of congestion under this criterion.

The following three maps (Maps 2,3, and 4) depict the deficient roadways by their level-of-service for existing conditions (2000), 2010, and 2020 under Performance Measure #1.

2.4: Performance Measure #2 – Roadway Corridor & Segment Travel Time
The second performance measure that will be used to identify congestion is a comparison between the off-peak travel time of a facility to its peak period travel time. A separate report entitled “2000-2001 Travel Time Study” documents the procedures that were used in collecting travel time data for incorporation into the CMS. Basically, a Global Position System (GPS) unit capable of collecting real-time position and speed data was placed on a probe vehicle that would actually be driven along with the regular traffic stream on each of the identified roadways to measure travel time and amounts of delay.

The probe vehicles were driven during both the morning and afternoon peak hours, as well as an “off-peak” hour that was usually chosen as some time between 9:00 a.m. and 11:00 a.m. The travel time data was stored electronically in a Geographic Information System (GIS) which enables graphic representation of speeds through digital thematic mapping.

The performance measure of off-peak versus peak hour travel time was established by the Knoxville TPO in its initial CMS plan developed in the mid-1990’s. The latest travel time study is an update to one that was performed between 1996 and 1997 with results documented in the “Travel Time Studies July 1997” report. A different methodology for collecting the travel time data was used in the 1997 report where the probe vehicles used a more manual technique using stopwatches, which makes it difficult to readily make comparisons between it and the latest study performed using GPS units. A future consideration involves the development of a transportation database similar to TRIMS that would be able to incorporate the travel time data from both reports and make comparisons between the two.

The following ratios of peak hour travel speed to off-peak travel speed were selected as congestion measures for the Knoxville Urban Area CMS, which mirrors the criteria from earlier CMS efforts by the Knoxville TPO:

Peak Hour Speed/Off Peak Hour Speed:
(1) > 0.75 = No Congestion (LOS A – C)
(2) > 0.667 and < 0.75 = Marginal Congestion (LOS D)
(3) > 0.5 and < 0.667 = Moderate Congestion (LOS E)
(4) < 0.5 = Serious Congestion (LOS F)

In simpler terms, serious congestion is defined as when speed in the peak hour is less than one-half of what it is during the off-peak time. Since speed and time are directly proportional to one another, another way to interpret serious congestion is when it takes you more than twice as long to traverse a roadway segment or corridor in the peak hour as it does during the off-peak times.

The roadways are broken into two different analysis categories – corridors and segments. A corridor represents the entire length of the roadway, unless the roadway traverses through the downtown core. For example, the Interstate 40 corridor was broken up into two different “super-segments” (east and west) for analysis purposes because the AM peak movement is assumed to be towards downtown, and the PM peak movement is assumed to be away from downtown. Therefore, the Interstate 40 east leg AM peak movement is westbound whereas the AM peak movement is eastbound on the west leg. The segments were defined as pieces of the roadway between major intersections (usually signalized) on an urban arterial, or between interchanges on the interstate.

The two levels of analysis – corridors and segments – are important because an entire corridor that shows up as being deficient will usually take precedent over one that has only a few segments along it that are deficient. Map 5 depicts the roadways that meet the congestion criteria under Performance Measure #2 for the AM peak hour, and Map 6 shows the congested segments during the PM peak hour.

2.5: Performance Measure #3 – Stopped Delay at Intersections
The final performance measure that will be incorporated into the CMS is that of Stopped Delay at intersections. The stopped-time delay was computed using the GPS units as part of the same Travel Time Study mentioned earlier. The units measured the amount of stopped delay that was incurred while traveling between intersections, which most often was the time that was spent stopped at a traffic signal. The 1994 Highway Capacity Manual (HCM) definitions for LOS based on stopped delay at a traffic signal will be used to define congestion in this report, since the more current editions of the HCM use a different methodology. The following are the 1994 HCM criteria for LOS based on traffic signal stopped delay:

(1) LOS A, B,C < 25.0 seconds of stopped delay
(2) LOS D > 25.0 sec < 40.0 sec (Marginal Congestion)
(3) LOS E > 40.0 sec < 60.0 sec (Moderate Congestion)
(4) LOS F > 60.0 sec (Serious Congestion)

It is important to note that the intersection delay that was collected from the Travel Time Study does not necessarily represent the total delay at a particular intersection because the trip was made straight through on the corridor and therefore delays that are being incurred for turning movements or on the side streets are unaccounted for. It will still provide, however, a method for identifying areas that need further investigation such as prioritizing intersections for turning movement counts and signal re-timings.

 

Footnotes:
2 More information about the EVE software program is available in the Report on Roadway Deficiency Analysis Software Development prepared by the University of Tennessee Transportation Center, April 1998.