Modelling Dynamic Traffic Management at Luton Airport with PTV Vissim

London Luton Airport (LLA) is one of the UK’s busiest airports, carrying over 16.2 million passengers in 2023 and plays a vital role in connecting the UK to various European and international destinations.

In October 2023, a major fire in Terminal Car Park 2 (TCP2) caused a partial structural collapse, taking both the car park and the existing drop-off area on its ground floor out of operation. This disruption left the airport in need of a temporary solution to maintain operations during the reconstruction of TCP2.

Ramboll supported London Luton Airport Operations Ltd (LLAOL) in developing options for a temporary solution to manage passenger drop-off. This included the creation of a new Temporary Drop-Off Zone (TDOZ) in the Central Terminal Area (CTA), along with Intermediary Drop-Off and Pick-Up Zones (IDOZ and IPUZ) located in the airport’s existing Mid-Stay car park and safe adapted pedestrian facilities in the vicinity of the terminal.

As part of the option development process, we conducted microsimulation modelling using PTV Vissim and Viswalk to accurately assess the impact of the proposed designs on traffic flow and pedestrian interactions within the airport.

The Model

The primary aim of the study was to evaluate the operational efficiency and performance of various layouts for the airport’s drop-off and pick-up zones, as well as traffic management solutions for the network in the vicinity of the terminal. This included key considerations such as safety, vehicle access and egress, dwell times, time delays, and interactions between vehicles and passengers.

Amongst the scenarios tested, the most challenging from a modelling perspective involved implementing a dynamic traffic management system. Traffic was diverted from the TDOZ—the primary drop-off location—to the IDOZ and IPUZ to prevent queues forming at the TDOZ entrance. During temporary closures, drop-off traffic was redirected until congestion eased, enabling the TDOZ to reopen.

Model Extents

The modelled area extends from the Percival Roundabout, continuing past the junction to the Mid-Stay car park, which is currently also serving as an intermediary drop-off and pick-up zone. Beyond the junction lies a tunnel, and after the tunnel, on the right, is Terminal Car Park 1 (TCP1), which remains operational. Adjacent to it is the site where TCP2 once stood, now undergoing reconstruction. The new TDOZ is located to the left of the tunnel exit. Built on a former taxi zone, the TDOZ has been operational since July 2024.

Assumptions

The simulation focused on the morning peak hour, with vehicles at the Drop-Off zones assigned a dwell time of three minutes and those at the Pick-Up zones given ten minutes. Each vehicle was modelled with an average occupancy of three individuals, including a driver and two passengers. A speed limit of 10 mph was applied across the network, reflecting typical operational conditions within the airport premises.

Model Specifications

Traffic Signals Optimization
An external controller was implemented in this model as an adaptive traffic signal control system to replicate the system in place at the Mid-Stay junction. Detectors and signal controllers were employed to assess traffic flows on each arm of the junction, calculating the optimal allocation of green time to maximize throughput based on traffic conditions

Vehicle types and routes
Vehicle types were defined by purpose (drop-off, pick-up, taxi, etc.), and three types of routes were implemented to represent their movements within the airport. Static routes were established for each vehicle type and their designated zones, along with partial routes to handle specific vehicle movements and distributions. Additionally, Parking Lot routes were created to allocate parking spaces for each drop-off and pick-up vehicle, ensuring the model accurately reflects their behavior and specified parking durations.

 Pedestrian types and routes
Two distinct pedestrian types and routes were coded: one for Drop-Off and another for Pick-Up. Dropped-off pedestrians travel from the parking lot where their vehicle has parked to the terminal, while Pick-Up pedestrians make the reverse journey.

The Challenges

The intricate interaction between vehicles and pedestrians, combined with the detailed structure of the model, posed several challenges.

One of the challenges was ensuring that each pedestrian was generated at the right time and location, and that their routes were accurately allocated. Pedestrians in high-traffic areas, such as drop-off and pick-up zones, have the potential to cause delays, which needed to be factored into the assessment.

Another key challenge was implementing a dynamic traffic management system to control congestion at the Temporary Drop-Off Zone. With limited capacity at the TDOZ, traffic had to be diverted to the Intermediary Drop-Off and Pick-Up zones to prevent queues at the TDOZ entrance. The model had to simulate these temporary closures and the redirection of traffic until congestion eased, allowing the TDOZ to reopen.

The Approach

Based Scripts that can be incorporated into Vissim using the Component Object Model (COM) Interface to trigger specific actions or conditions in real-time during simulations.

Pedestrians
The integration of User-Defined Attributes and Scripts was crucial for accurately modelling pedestrian inputs within the simulation. UDAs were used to connect each parking lot to its corresponding pedestrian area, determining the origin or destination for each pedestrian based on whether the parked vehicle is dropping off or picking up passengers.

For each vehicle parked, the script generates the appropriate number of pedestrians at the correct location, defines their destinations, and adjusts the vehicle occupancy attributes. The integration of UDAs and Event-Based Scripts ensured that both vehicle and pedestrian behaviors were dynamically linked, allowing for accurate representation of real-world interactions.

Drop-off Vehicles
Two Static Routes were set up for vehicles classified as drop-off: one leading to the Temporary Drop-Off Zone and another to the Intermediary Drop-Off areas (IDOZ and IPUZ). This dual-route system was designed to simulate dynamic traffic management automatically diverting vehicles to the Mid-Stay Drop-Off zones when the TDOZ closes and redirecting them back to the TDOZ once it reopens.

A route decision point, simulating a variable message sign, is positioned just before the Percival roundabout entrance. This decision point dynamically allocates routes based on the TDOZ’s operational status.

To make this process dynamic, a combination of UDAs and Event-based scripts was employed. The UDAs were set up to reflect the TDOZ condition, establish thresholds for its closure and reopening, and track the number of vehicles entering or exiting the zone. At each simulation time-step, the script evaluates the UDA values to determine the TDOZ’s status. Based on this evaluation, the script either closes or reopens the TDOZ and assigns the appropriate route for drop-off vehicles, ensuring a responsive and adaptive traffic flow management system.

This dynamic traffic management is demonstrated in the video where vehicles are color-coded by type and route:

• Black:        Drop-off vehicles with no route assigned
• Orange:    Drop-off vehicles heading to TDOZ
• Red:          Drop-off vehicles heading to Mid-Stay drop-off zones
• Blue:         Pick-up vehicles
• White:      Other vehicles

The Results

TDOZ model results were instrumental in analyzing and optimizing various layout options, helping Ramboll provide London Luton Airport Operations Ltd with the optimal solution for managing passenger drop-off. By accurately simulating traffic and pedestrian movements, the model identified potential bottlenecks and congestion points, ensuring that the final design effectively accommodated peak hour traffic while minimizing delays. This data-driven approach played a crucial role in delivering a solution that balanced safety, efficiency and operational performance during the airport’s infrastructure redevelopment.

The implemented solution has been monitored, and its performance aligns with the predictions of the developed model. This alignment highlights the model’s reliability in simulating real-world conditions and confirms the effectiveness of the simulations in delivering an optimized and functional solution.

Conclusion

This model highlights some of the challenges of tackling real-world complexities effectively and successfully, illustrating the importance of the advanced tools and flexibility that PTV Vissim brings to microsimulation projects. By leveraging features such as User-Defined Attributes and Event-Based Scripts, we were able to dynamically manage traffic flow and adapt the system in response to real-time conditions. These features enable precise simulations of dynamic scenarios, ultimately delivering insights that shape efficient and robust transportation solutions.