APPENDIX C

Accounting for CAVs in Airport Landside Traffic Flow–Simulations

Considerations on VISSIM

The metropolitan airport–wide model used for the simulations was created using VISSIM microsimulation software (version 22-06), developed by the German company Planung Transport Verkehr AG (PTV). Microsimulation is a recognized approach for evaluating the operations of complex traffic movements such as those found at airport frontages, including taxi operations, passenger walkways, for-hire-vehicle drop-off/pick up, dwell times, and recirculation. VISSIM is a tool that models driving behavior, such as close-following and lane changing, and it allows other drivers within the model to interact with one another. VISSIM also allows for the customized distributions of driving behaviors and vehicle attributes, which makes it the appropriate tool to use for this research.

VISSIM can provide traditional traffic engineering Measures of Effectiveness (MOEs) that are comparable to the Highway Capacity Manual: A Guide for Multimodal Mobility Analysis (HCM7 2022) for airport curbside roadways. Level of service (LOS), queue length (in feet), and vehicular throughput at the departure and arrivals frontages (in vehicles per hour) were used as MOEs for this analysis.

Behavior Adjustments

The baseline model developed for this project used normal conditions outlined by HCM7 and PTV. These conditions were adjusted according to the engineer’s judgment per PTV recommendations, as the parameters outlined for normal conditions produced an unreasonable output. Table 42 and Table 43 outline the driving behavior adjustments made to HCM7 parameters and the changes made to maximum acceleration and deceleration times, respectively.

Acceleration and deceleration rates between vehicles can be adjusted to have a narrower variability. These parameter adjustments make it possible for self-operated vehicles to move into lanes with long platoons and allow for platoon leaders to make a reasonable lane change without waiting for an unreasonably long gap.

When assessing the impact of CAV penetration, the following criteria should be analyzed:

• Queue length
• Platooning on freeway segments and at frontages
• Recirculation
• Approach roadways to arrivals and departures frontages
• Frontage performance

Table 42. HCM7 Driving Behavior Parameter Adjustments

Table 43. PTV Parameter Adjustments

A platoon of connected and automated vehicles (CAVs) is defined as “a group of CAVs that exchange information, so that they can drive in a coordinated way, allowing very small spacings and, still, travelling safely at relatively high-speeds” (Martínez-Díaz et al. 2021).

Development of the Baseline Model

The initial step for developing the baseline model of this project was to analyze arrivals and departures frontage operations assuming 2019 (pre-COVID-19) base conditions with the existing vehicle mix and no CAVs at all. CAV market penetration was then increased to 30, 60, and 90 percent market penetration. The resulting frontage performance was compared to the base condition to determine the effects of various CAV market penetration levels on frontage performance as well as operating conditions at critical roadway segments, especially those considered bottlenecks today. Under this scenario, the arrivals and departures frontages operate the same way as the existing conditions, but with varying CAV market penetration rates.

The existing 2019 base conditions created in VISSIM required the input of traffic volumes, signal timings, and roadway geometry, including curbside parking area at the arrival and departure frontages. In addition to traffic data, VISSIM model parameters and inputs were required. The parameters and inputs used in the VISSIM models are summarized in Table 44.

Table 44. Summary of 2019 Base Conditions: VISSIM Inputs and Parameters

Recommendations for Further Research

The following gaps in policy or practice were identified during the course of the project, suggesting further research efforts:

• CAV-driver behavior parameters: The CAV parameters recommended by HCM7 and PTV were adjusted for airport operations to avoid unreasonable model behavior. Further research is recommended to optimize CAV-driver behavior parameters under different operating conditions and highway geometry as well as at airport frontages.
• Platooning: Platooning behavior was found to affect the operation of weaving segments, merges, frontages, and roadway segments where lane changes are required. Modifications were made to the gap variability and gap acceptance parameters to reduce the effects of platooning behavior where lane changes are necessary in the network. Further research is recommended to better understand platooning behavior of CAVs and to determine the optimum maximum platoon size under various operating conditions and roadway geometry.
• Frontage performance: Further research is recommended to determine if changes in frontage operations and frontage geometry are needed to optimize CAV operations.