TL;DR
This paper introduces a novel two-lane simulation model for airport curbsides, incorporating both human-driven vehicles (HDVs) and connected and autonomous vehicles (CAVs). The study demonstrates that increasing CAV penetration significantly boosts traffic flow efficiency and reduces vehicle delays, offering valuable insights for future airport infrastructure design and traffic management.
What Problem Does This Paper Address?
The rapid growth of air travel, particularly post-pandemic, has intensified pressure on airport infrastructure, leading to increased ground congestion, flight delays, and security concerns. Airport curbsides, crucial for passenger and luggage transfers, are particularly susceptible to congestion due to high demand for pick-up and drop-off services. While existing research has explored curbside capacity and traffic flow, there's a need for more sophisticated models that account for the emerging mixed traffic environment of human-driven vehicles (HDVs) and connected and autonomous vehicles (CAVs). This study aims to fill this gap by analyzing the operational characteristics of such mixed traffic flows at airport terminal curbsides, providing a theoretical basis for improved design and management strategies.
Methods
The researchers developed a two-lane parking simulation model for airport terminal curbsides. This model integrates several key components:
- Intelligent Driver Model (IDM): Used to simulate the behavior of human-driven vehicles (HDVs), accounting for varying driving styles (radical, stable, cautious) based on information acceptance levels and time headways.
- PATH-calibrated Cooperative Adaptive Cruise Control (CACC): Models the behavior of CAVs when communicating with other CAVs.
- Degraded Adaptive Cruise Control (ACC): Models CAV behavior when the lead vehicle is an HDV, where CACC capabilities degrade to ACC.
- Departure Constraints: Implemented to enhance safety when vehicles leave the parking or drop-off zones.
The simulation captures the entire process of vehicle arrival, parking service, and departure, allowing for the analysis of traffic flow dynamics under various scenarios.
Datasets
Not discussed in the paper. The study relies on a simulation model rather than real-world datasets for evaluation.
Key Insights and Ideas
- CAVs are a game-changer for curbside efficiency: The introduction of CAVs, even at moderate penetration rates, dramatically improves traffic flow speed and reduces delays. This suggests that investing in CAV infrastructure and adoption could be a highly effective strategy for alleviating airport congestion.
- Parking space utilization follows predictable patterns: Regardless of CAV presence, parking space occupancy exhibits a consistent three-stage pattern: stable and concentrated in the initial berths, a slow decline in the middle, and a rapid decline towards the end of the curbside. This highlights that congestion is more likely to occur at the front of the curbside.
- Delay and efficiency are inversely related: As expected, higher traffic efficiency (speed) directly correlates with lower vehicle delays. This reinforces the idea that improving flow dynamics has a direct impact on passenger and vehicle experience.
- Parking demand significantly impacts performance: Higher parking demands exacerbate congestion and reduce traffic efficiency, making the benefits of CAVs even more pronounced in these high-demand scenarios.
Key Results
The simulation results highlight several significant findings:
- Increased Average Speed: For a curbside length of 400m and a parking demand of 1300 pcph, increasing CAV penetration from 0% to 100% resulted in average traffic flow speed increases of up to 48.7%.
- Reduced Average Delay Time: The same scenario (400m, 1300 pcph) saw average delay times decrease by up to 20.0% with 100% CAV penetration.
- Consistent Parking Utilization: The study found that the pattern of parking space utilization (stable, transition, rapid decline) remained consistent across different CAV penetration rates.
- Impact of Parking Demand: At a CAV penetration rate of 30%, increasing parking demand from 900 pcph to 1500 pcph led to a decrease in average traffic efficiency from 8.8 m/s to 6.0 m/s and an increase in average vehicle delay from 116.6s to 149.9s.
Limitations and Future Work
The authors acknowledge several limitations:
- Simplified Model Assumptions: The simulation model simplifies real-world traffic by assuming vehicles follow the same rules, ignoring individual driver differences, lateral movements, sudden event handling, and weather impacts.
- No Dynamic Differences: The model does not account for dynamic differences between various vehicle types beyond HDVs and CAVs.
Future research directions suggested include:
- Driver Behavior and Technology Interaction: Investigating the impact of individual driver differences and how driving behavior changes under high CAV penetration scenarios.
- Multi-dimensional Traffic Flow Simulation: Expanding models to include more complex real-world factors like weather, lighting, and traffic incidents to enhance practical applicability.
- Long-term Impact and Management Strategies: Exploring the long-term effects of driver behavior and roadside scenarios on mixed traffic flows and optimizing traffic flow management strategies.
Summary updated