Case study: Frankfurt International Airport
Frankfurt International Airport in Germany presents a convincing example of the benefits of intelligent dynamic signage on crowd movement in a large, complex occupancy. Serving one of the world’s leading financial cities, Frankfurt Airport is a major international airport with more than 140,000 passengers per day and 80 aircraft movements per hour. From 1980 to 2010, the annual passenger volume tripled from 17 million to 53 million.
Fraport AG, the owner and operator of the airport, led many expansion projects to keep up with this increased passenger demand, including additional runways, a new terminal building, and new operational procedures. These construction projects and changing security requirements resulted in limiting access to check-in counters, gates, and stands for significant periods. As wait times increased and the passenger experience at the airport deteriorated, an ambitious method of redirecting passenger traffic was needed to maximize flow efficiency and avoid operational delay for departing flights.
Developers of the new strategy wanted a system that would be proactive to actual conditions in the airport, rather than reactive. A proactive approach required an accurate forecast of passenger distribution across terminals and security checkpoints and redirection to buffer areas that held extra capacity in the event of overcrowding. Criteria was established for the development of the new product. Each checkpoint in the airport was given a capacity number, which would then be compared against the predicted passenger flows. Where capacity was not sufficient to produce an acceptable passenger experience, the goal of the system was to both actively and passively redirect passengers to other routes.
A basic concept was outlined around three disciplines that were identified as important to the effectiveness of the system. The first was the measurement of actual passenger flows in the terminal, or a current snapshot, at a given time. The second was the forecasting function of the system, which needed to predict how the conditions would change over the next 30 to 120 minutes. The final component was the control function, which needed the ability to change or influence passenger flows through the terminal. The interface between each of the functions is illustrated in Figure 3.
Fraport and system developers used a combined approach to create an accurate snapshot of current conditions in the airport by using each of the following four technologies:
- Boarding passes: Frankfurt Airport uses automated boarding passes at checkpoints, with passengers scanning a barcode as they pass. Each scan of a boarding pass has a time stamp, giving the system an accurate head count of people in the area while also allowing for a calculation of the approximate number of passengers that will be arriving in the private, airside section of the terminal.
- CCTV camera system: Cameras count occupants as they cross virtual boundary lines and transmit data on passenger flows to the main system, providing another data point on current passenger conditions.
- Bluetooth technology: A sensor at the security checkpoint gathers data on passenger movement using the activated Bluetooth devices in mobile phones or tablets. The sensor registers the time at which the Bluetooth signal is first detected, or when the passenger enters the security line, and the time at which the signal is no longer detectable when the passenger leaves the security area. These two data points allow for a measurement of total wait time at the security checkpoint.
- Automated metal-detection gates: Passengers are counted as they pass through automated metal-detection gates, which are installed at all security checkpoints in the airport.
Simulation technology is used to forecast conditions at several security checkpoints, boarding gates, stairs, elevators, tunnels, and other egress elements using three pedestrian models:
- Simple mathematical model: Uses simple correlations that treat occupant movements as a series of discrete events.
- Trajectory model: Provides a general representation of the complexity of occupant movement at different points but ignores the individual properties of each person.
- Social force model: Focuses on individual occupant behavior and evaluates the interaction between people as they move from one point to the next.
Separate simulations for each of the three models are performed using both the data from the measurement function and historical data on the number of passengers per flight and passenger flight-transfer data. The results of each model provide a different set of data and are used to provide an overall picture of the terminal conditions at a given time. Before the system was implemented, a quality assurance program evaluated each calculation method against measured data. Where necessary, input parameters were adjusted until weaknesses in the simulation forecasts were corrected.
Each pedestrian model can perform simulations for the entire airport in 10 minutes or less, allowing for multiple simulations throughout the day. In a typical day, each simulation is performed almost 300 times, producing the necessary information needed to submit to the control function of the system.
Results of the simulations are presented on a graphical user interface in the airport’s Terminal Control Centre, where trained airport staff can take corrective actions to improve flow efficiency. The impact of the corrective actions on passenger movement in the terminal can be fed into the system such that the level of improvement can be analyzed. Once an optimal solution to an issue is found, airport staff can update dynamic signage, as shown in Figure 4, to direct occupants toward more efficient flow paths.
The Frankfurt Airport intelligent dynamic signage system presents an excellent model for other airports and complex occupancies to follow. Since implementation, Fraport’s internal customer-satisfaction surveys have shown improvement in relation to waiting times and movement through the terminal, a notable feat when considering the extensive construction projects being undertaken at the airport.
Andrew Biery is a senior fire engineer with Arup, where he has been a key contributor to the fire and life safety design for multiple projects involving large, public spaces including the New International Mexico City Airport and the Finch West Light Rail Transit project in Toronto.