Inspiration can strike anywhere, anytime - if you're paying attention.
An idea for improving customer service (and possibly increasing concession revenue) came to Joe Lopano, chief executive officer of Tampa International Airport (TPA), while eating lunch with his son.
The two were dining on TPA's main transfer level before catching a flight, when Lopano began wondering if they had time for a slice of pie. He knew how long to allow for the shuttle ride to their gate; but the amount of time needed to clear security was not nearly as clear.
To be safe, they skipped dessert. And Lopano later consulted Doug Wycoff, the airport's manager of innovation and infrastructure support, about developing a system to quantify the unknown, but highly relevant, detail of checkpoint wait times.
Because TPA's four airside areas are separated from its main terminal by a train ride, passengers don't get any visual clues about the length or pace of TSA screening lines until well after they've left the landside terminal. But that will soon be irrelevant. Later this year, the airport plans to begin posting estimated checkpoint wait times on flight information displays throughout the terminal.
"When our customers are shopping or eating in our main terminal, our desire is to get them as much information as possible," says Wycoff. It's important for passengers to know whether they need to get going immediately or they have time to spare, he explains: "We are trying to keep them in our main terminal as long as possible while still making them comfortable."
A multi-tiered system that uses thermal detection, barcode tracking and Bluetooth technology will calculate the checkpoint wait times.
Developing the Solution
The challenge of estimating how long it will take passengers to clear security initially prompted the airport to use Bluetooth technology. Com-Net Software and SITA, two of the airport's existing contractors, were able to glean real-time queue data by tracking passengers with Bluetooth-enabled devices such as cellphones and iPads.
SITA Senior Product Manager Kevin Peterson compares the approach to using an electronic stopwatch. The stopwatch starts when a passenger carrying a Bluetooth-enabled device passes through the landside checkpoint; it stops when the same passenger exits the security area. The difference is calculated, and an average wait time is determined with the help of additional bits of intelligence, explains Peterson.
The address of each mobile device is translated into a unique tracking number to keep customer information truly anonymous, he adds.
While the method has its merits, TPA personnel soon realized it wasn't providing enough information. "Bluetooth in itself doesn't give you a high enough hit rate," Wycoff explains, noting that only 5% of passengers enable their devices' Bluetooth capability at the airport. "Also, we didn't know what the wait time was until after the passenger had left the checkpoint."
To provide meaningful predictive information, TPA needed to measure the volume of passengers approaching its checkpoints as well as the amount of time it took to pass through them.
Wycoff again conferred with Com-Net and SITA, and the airport ultimately entered into a two-year pilot using customized versions of SITA's proprietary core products: Airport iQueue, Airport iTrack and Airport iValidate. Together, the products measure and predict checkpoint wait times by combining Bluetooth and barcode tracking with separate validation technologies.
Tracking passengers via barcodes on their boarding passes is a key element in the trio of integrated technologies. Because every passenger carries a boarding pass pre- and post-security, TPA's data yield increased exponentially.
"We needed more consistency on the hit rate, so we added (technology) that counts the number of passengers from the time they enter our shuttle lobby to the time they exit the back side of the TSA scanning equipment," Wycoff explains. "When you take that in conjunction with the Bluetooth-measured data to validate the predicted wait times, it allows you to manage high volumes of traffic and still provide accurate timeframes, because we know they are coming."
Project: Predicting Security Checkpoint Wait Times
Location: Tampa (FL) Int'l Airport
Technologies Used: Bluetooth & barcode tracking; thermal detection
Vendors: Com-Net Software; SITA
Testing: 2 yrs; 2,500+ performance checks
Expected Public Launch: 3rd quarter 2014
Delivery: Estimated wait times will be communicated to passengers on flight information displays
Data is captured at multiple points between the train's boarding platform and the area where passengers put their shoes back on after clearing security, notes Com-Net Sales Manager Jeff Collins.
"Tampa is a unique situation, because the current wait times through the security checkpoints are highly dependent on the number of passengers on the shuttles, not just the travel time on the shuttle, thereby rendering the use of historical wait time data somewhat irrelevant," Collins explains. "To accurately estimate current wait times, an alternative passenger volume-based approach was required."
Generating information that would predict rather than record wait times requires a multi-step, multi-technology approach. In addition to counting passengers by scanning the barcodes on their boarding passes at document inspection stations, TPA also uses thermal "people counters" that detect passengers' body heat. After cross-referencing passenger counts collected by the different technologies, algorithms predict wait times based on how many people are in and approaching a checkpoint at any given time.
"The primary technology in determining wait times is people counts, but since thermal people counter technology is (only) 95% to 98% accurate, Bluetooth comes in as a re-sync mechanism," explains SITA's Peterson.
Last year, the airport ran more than 2,500 tests to ensure the integrity and performance of the system's results, reports Wycoff.
"We didn't want to expose ourselves to error," he relates. "We have worked on it for a couple years now, and we feel like we are reaching a point where we are very comfortable."
Currently, the system is monitored through TPA's network operation center, and officials are satisfied with its performance, Wycoff notes. Upon full deployment, the airport plans to pass wait time information from the system to Com-Net for dissemination to customers on flight information display system monitors.