Few things can get the heart racing and the stomach gurgling like waiting in a long airport security line and freaking out over whether you will make your flight.
Pittsburgh International Airport has found a way to ease your mind — or confirm your worst fears.
Airport officials have teamed with Oakland-based software firm Zensors to tell travelers how long they will be waiting to clear security, give or take a couple of minutes.
The information is available on monitors in the airport’s Landside Building that keep track of arrivals and departures, as well as on its website, FlyPittsburgh.com.
“We know security can be a frustration for travelers and having accurate wait estimates can help set expectations and aid in planning trips,” said Christina Cassotis, CEO of the Allegheny County Airport Authority, which operates Pittsburgh International.
Zensors technology uses closed-circuit cameras and a specially trained neural network — a type of machine learning that mimics the way the human brain works — to observe and estimate wait times, weighing such factors as the time of day and the number of TSA agents on duty.
The system is able to distinguish between those standing in line and those who are just milling around, sitting on benches, or using the nearby escalators.
Under the system, wait times are updated every minute, with arrows showing whether they are increasing or dropping.
Pittsburgh International is the first airport in the country to deploy the Zensors technology, which is not that much different than that used in autonomous vehicles. It has been operating since Friday.
So far, the technology has lived up to its hype.
Its estimates have been on average within two minutes of the actual time it has taken a traveler to get from the start of the line to TSA X-ray machines, said Katherine Karolick, the authority’s senior vice president of information technology.
The most it has been off is six minutes, she noted. “We’re very happy with where it’s at now,” she said.
Ninety-five percent of the time, the system should be accurate to within two minutes, added Chip Homer, Zensors head of marketing.
The system should become even more precise over time as it, in essence, gains knowledge, Ms. Karolick said. The airport also is planning to install the system at the alternate security checkpoint on the ticketing level of the Landside Building in the near future.
Anuraag Jain, head of product at Carnegie Mellon University spinoff Zensors, said the project is the start of efforts to use artificial intelligence to help travelers in the often stress-filled airport environment.
“We’re applying deep learning in a way that can really become a game changer for passengers and airport operations,” he said in a statement. “We’re excited to help turn Pittsburgh into the world’s smartest airport.”
Working with Zensors is just one of the ways the airport has been using cutting-edge technology in a bid to help travelers. In May, it teamed with CMU to implement NavCog, a smartphone-based navigation system that helps people with visual impairments find their way through the terminal.
That project is part of a partnership Pittsburgh International has with CMU to develop technologies to enhance the traveling experience.
Together, the goal is to eventually develop the means to take travelers from the parking lot to the gate in the most efficient manner. Officials even hope to go a step beyond Zensors to eventually give flyers an exact time to arrive at the security checkpoint.
They also plan to work with Zensors on other projects designed to “improve the passenger journey.” That could include information on how long the waits are at everything from ticket counters to Starbucks.
“The eventual goal is to be able to provide this holistic passenger experience,” Mr. Homer said.
Airport officials also hope to incorporate the various technologies into the new terminal that is being developed as part of a $1.1 billion modernization to be completed in 2023.
Mark Belko: mbelko@post-gazette.com or 412-263-1262.
Updated at 5:31 p.m. on July 31, 2019
First Published: July 31, 2019, 4:49 p.m.
