HCM 7th LOS Criteria in Micro Simulation

The objective of this article is to provide a brief overview on how to compare the results of micro simulation against Highway Capacity Manual 7th (HCM 7th) Level Of Service (LOS) for motorized vehicles. It should be noted that the results of the HCM are valid within the limitations of the methodologies stated in HCM.

The methodologies for calculating the LOS of other modes (e.g. pedestrians and bicycles) are also provided in the HCM and should be considered in traffic analysis. These methodologies are not reviewed in this article.

Intersections

In the HCM, the delay (s/veh) is the main quantitative measure or criterion for determining the LOS of the intersections. The thresholds for all types of unsignalized intersections (Exhibits 20-2 & 21-8) and roundabouts (Exhibit 22-8) are the same. The thresholds are higher for the signalized intersections (Exhibit 19-8), since it is already accepted by the drivers that the signalized intersections have longer delays compared to unsignalized ones. The analysis time period is usually the peak 15 min within the peak hour (please refer to the HCM for other options).

As an example of the intersection methodologies, for signalized intersections the delay is first calculated for the lane groups (please refer to the HCM for the definition). Then for other objects (e.g. the approach), it is calculated through averaging the delay of associated lane groups weighted by their volumes. Similarly, the delay of turning movements can be calculated. Just to add, since the volumes are input by turning movements, the number of vehicles using each lane group is calculated through a process explained in Chapter 31 (in case there is any shared lane). It is an iterative process that balances the volumes on lanes to achieve the minimum service time for the vehicles having the choice of lanes.

In VISSIM, the Delay is considered as the difference between the actual travel time and the travel time if it was possible to travel with the desired speed (based on desired speed decision, curve speed and reduced speed area) without encountering other vehicles or obstacles like traffic signals. Planned stops (parking dwell time or public transit dwell time) are not included in the delay, but the deceleration and acceleration before and after them are included. Stop Delay is part of the Delay, in that the vehicle is fully stopped.

In VISSIM, the Delay at the intersections for each turning movement, can be measured using Evaluation Nodes. The extent of the Link considered in the Delay is not unlimited and is set in the Evaluation Configuration. Additionally, the latent vehicles (not yet entered the network due to long queue in simulation) will not be included in this Delay measurement and are measured separately.

Based on the details of the HCM methodologies, the concept of delay is the same as VISSIM and so the two results can be compared directly for the analysis time period (please refer to the HCM for the definition).

As mentioned before, the results of the HCM are valid within the limitations of the methodologies stated in HCM. Therefore, the comparison is valid within these limitations, including no demand starvation and no queue overflow or spillback. As an example, the limitations of signalized intersections methodology are provided on page 19-20.

Urban street segments

Urban street segments methodology of the HCM is developed to determine the frictional impact of the driveways and other elements along a segment of up to 2 miles long between two intersections. Limitations of the methodology are provided on page 18-13. The duration of the analysis time period should be so that the traffic condition is steady throughout this period. As the result, the analysis time period can be 15 min or 1 hr.

Through vehicle travel speed (space mean speed or in other words, length divided by the travel time) is the main criterion for determining motorized vehicle LOS of urban street segments (Exhibit 18-1).

In VISSIM, travel speed along a segment can be calculated based on the output of Vehicle Travel Time objects. Apparently, this travel time includes the delay caused by the downstream intersection (e.g., signal delay). The output reflects the average travel time of vehicles that traverse both the start and end cross-sections regardless of their paths in between. Furthermore, results from short measurement segments are not reliable due to the stochastic nature of the simulation.

Freeway segments

Freeway segments methodologies of the HCM are developed to analyze basic, weaving and merge/diverge segments (please refer to the HCM for the definitions). For the merge/diverge segment methodology, the study area or the ramp “influence area is generally limited to the two rightmost freeway lanes and any acceleration or deceleration lanes present, for a distance of 1500 ft downstream of the merge point or upstream of the diverge point”. For the weaving segment methodology, it consists of all lanes from 500 ft upstream to 500ft downstream of the contact points of on and off ramps with the freeway.

The main criterion for determining the LOS of such segments is density (Exhibits 12-15, 13-6 and 14-3). The analysis time period is the peak 15 min within the peak hour. The density (pc/mi/ln) is calculated as follows:

(Demand rate / Number of lanes) / Speed

Where:

• Demand rate (pcu/hr) is the entering passenger car units to the segment within the 15 min analysis time period times 4

• Speed (mi/hr) is the space mean speed or in other words, length divided by the travel time

To compare VISSIM results with the HCM, it may be more accurate to measure the density in VISSIM with the same method as the HCM. PCE factors provided in Exhibit 12-25 should be applied.

The density that is directly output by VISSIM is collected with a different method. Please refer to the manual for the details. To compare this density with the HCM, it should be collected Per Link Segment and make sure it is divided by the Link Number of Lanes (Density is the total of all lanes and Density Per Lane is the average). Moreover, this output refers to the vehicles (veh/mi and veh/mi/ln) rather than the pcu and so it needs to be transformed for the comparison.

Two-lane highway

The two-lane highway methodology is developed to analyze a two-way corridor with one lane in each direction that is longer than 2 miles (its signalized intersections are more than 2 miles apart). The corridor is divided into homogeneous segments for analysis (please refer to the HCM for more details). The analysis time period is recommended to be 15 min.

Follower density (Exhibit 15-6) is the main LOS criterion for two-lane highways. “Follower density is the number of vehicles in a follower state per mile per lane (i.e., percent followers multiplied by density).” However, “segment performance values are reported for the end of the segment, rather than corresponding to aggregated values along the full segment length” (Page 15-10). The threshold for being considered as a follower is the headway of 2.5 sec with the front vehicle (front bumper to front bumper).

As explained, follower density (followers/mi/ln) is the share of followers passed through a section multiplied by the density. For a section, the density is flow rate (veh/hr) divided by point speed (mi/hr). Since the analysis time period is 15 min, the flow rate is counted volume times 4.

To compare VISSIM results with the HCM, it is more accurate to measure the density in VISSIM with the same method as the HCM. It should be noted that the density that is directly output by VISSIM is collected with a different method and is associated with a segment rather than a section. Please refer to the manual for the details.

In VISSIM, the details of Vehicles such as Driving state, Clearance, Following distance, etc. are available as attributes of Vehicles In Network and can be extracted with several different methods.