Embedded Files
Car Following Model Parameters (Freeways – Wiedemann 99)
UDOT uses VISSIM to model complex traffic behavior. VISSIM’s fundamental logic is built on car following models with a series of parameters available for adjustment. It is important to calibrate these models using the correct parameters to produce accurate flow rates. To help identify the flow rates in a VISSIM model, download a VISSIM layx file from the UDOT website at https://udot.utah.gov/connect/business/design/traffic-modeling-guidelines. This file is set up to visualize the links based on the link flow rate.
The Wiedemann 99 car following model should be used for all freeway facilities as included in the UDOT VISSIM template. Typically, calibration of the freeway only requires small adjustments to the CC1 and CC2 values. These adjustments should be used to replicate realistic flow rates, particularly in congested conditions. Field flow rates for Utah freeways are measured by radar and are publicly accessible online through PeMS. If the segment being modeled does not currently experience congestion, then similar freeway segments for other locations in the state should be reviewed. Adjustments to CC1 & CC2 require justification among field observations. Note that UDOT requires additional documented data to support changes to all other CC values as situations that justify these changes are extremely rare.
Table 1. UDOT Template Default Values for Wiedemann 99 Car Following Model
CC1 controls the gap time distribution (in seconds), which is the gap that a following vehicle wants to maintain in addition to the standstill distance (CC0). CC1 is the dominant factor that controls capacity at high volumes. A smaller time gap would increase capacity while a larger time gap would decrease capacity.
Desired safety distance = CC0 + CC1 × speed
Freeway Flow Rates
Below are several examples of the various maximum flow rates achieved in the field along locations with regular congestion. This data was measured using side-fire radar.
Speed vs flow diagrams based on these data are shown in Figures 1 to 6.
The maximum capacity of a corridor is affected by a number of factors including:
Closely spaced interchanges which create a weave condition
Lots of heavy trucks which reduce maneuverability
More than 4 lanes in width
Lower speeds
Steep grades
Curves in the roadway
Narrow lanes
Barrier that is too close or steep drop offs
The Department has found that, in Utah, the optimum capacity for a freeway is generally reached at a width of 4 lanes. As the number of lanes increases, the number of lane changes required to utilize those lanes also increases. As vehicles maneuver into new lanes, other vehicles slow to accommodate them. This can artificially increase the density, meaning there is empty space on the freeway that could have otherwise been utilized, had vehicles not been crossing over so many lanes. This reduces the available space for vehicles to occupy which ultimately reduces capacity.
Other conditions may also affect capacity. Field observations should be performed to identify these and model them appropriately.
Figure 1 shows a low flow rate location that breaks down around 1,500 vehicles per hour per lane (vphpl). Figures 2 and 3 show medium flow rate locations that break down around 1,700 vphpl. Figures 4-6 show high flow rate locations that break down around 1,900 vphpl. The figures are provided as examples to help determine appropriate flow rates.
Figure 1. Northbound I-17 between Park Ln and Shepard Ln
Figure 2. Southbound I-15 between I-215 and 9000 South
Figure 3. Southbound I-15 between 500 East and Pleasant Grove Blvd
Figure 4. Northbound I-15 between 600 North and I-215
Figure 5. Northbound I-15 between 9000 South and I-215
Figure 6. Southbound I-15 between 600 South and Pleasant Grove Blvd
To demonstrate the correlation between flow rates and CC1 values, test model runs were performed for a basic 70 mph freeway segment with 8% heavy vehicles using a range of CC1 values. The results of this analysis are shown in Figure 7, which is provided for use as a reference.
Figure 7. VISSIM Freeway Vehicle Flow Rates
CC1 and CC2 Model Parameters
The recommended starting value for CC1 is 1.3 seconds in the Wiedemann 99 Car Following Model. Note that this is a starting value, not an absolute value. It is critical that accurate freeway flow rates which represent the operations in the field are reflected in the model. It should be noted that although the CC1 value is a time distribution in Vissim, the default is to have no variance in the time (i.e., the standard deviation should be 0). The CC1 can then be adjusted as needed to calibrate the model to a capacity more in line with those observed in the project study area but should be justified with data to support the change (e.g., PeMS peak flow rate). Abrupt changes in the CC1 from one link to another should also be avoided so that artificial congestion is not created at link boundaries.
CC2 controls the following distance oscillation, which is the maximum additional distance beyond the desired safety distance that is accepted by the following vehicle before intentionally moving closer. The default value of 13.12 feet (4.0 meters) results in stable following behavior. The default value should almost always be used and only adjusted in exceptional cases after discussion with UDOT.
Situations may arise where additional factors may need to be adjusted. The consultant should discuss the changes with UDOT and fully document the modifications and reasons why the changes were needed.
Page updated
Report abuse