This is the second in a three-part series on how COVID-19 could revolutionize airport checkpoints. Read Part One here.
Accessible property screening – TSA’s official term for inspecting passenger’s carry-on items with an X-ray scanner – has long been a critical aspect of aviation security. In response to evolving threats, TSA adjusts its processes and procedures to meet the evolving threat and to achieve the highest levels of transportation security. Pre-9/11, passengers could put practically anything short of a loaded gun into a carry-on bag. Now, TSA may screen personal electronic devices separately, as well as foods, powders, liquids, and materials that obstruct clear images on the X-ray machine. This has improved security dramatically – but at a cost. Accessible property screening often involves lots of interaction and physical contact between the Transportation Security Officers (TSOs) and the traveling public, not to mention the delays that occur when passengers forget to take prohibited items out of their bags.
| Predictions
· Two Years: Remote Imaging Analysis occurring in the nation’s largest airports · Five Years: Use of Regional Operational Imaging Analysis Centers becomes the norm · Ten Years: Self-Service Screening becomes a reality |
But thanks to several converging developments in security technology, we believe TSA is the cusp of the most far-reaching changes to accessible property screening since the start of the TSA Pre-Check program a decade ago.
One element of this transformation is well underway, with the deployment of computed tomography (CT) systems at aviation checkpoints. These systems use the same technology that has been screening checked baggage for years. Given certain threat conditions, these CT systems will allow passengers to leave items, such as electronics and foods, in their carry-on bags. Unlike traditional X-ray scanners, checkpoint CT systems produce a 3-D image which a TSO can manipulate to get a better view of a bag’s contents. As a result, a checkpoint CT system often allows a TSO to clear items without having to open the carry-on bag, thus reducing a touchpoint (and a delay) in the process. TSA’s procurement of checkpoint CT is well underway, and its ultimate goal is to deploy a CT system at every checkpoint screening lane in the country.
TSA has also been installing Automated Screening Lanes (ASLs) across the aviation security enterprise. To date, TSA has overseen installation of 231 automated screening lanes at 18 airports nationwide. The ASLs offer additional security enhancements, provide faster processing times, and increase the customer experience. They allow multiple passengers to place their items in trays simultaneously and submit them for screening without waiting on the first person to complete the process. Beyond the ability for multiple guests to fill their trays at the same time, the ASLs offer additional benefits for travelers, including an automated tray return system, a separate conveyer system for bags that require additional screening, larger trays that can handle more carry-on capacity, and the use of Radio Frequency Identification (RFID) tags on each tray for better tracking and accountability of a traveler’s items. Overall, the combination of checkpoint CT systems and ASLs will significantly increase passenger throughput and improve operational efficiency.
2 years: Remote Screening and Increased Use of Artificial Intelligence (AI)
Today, at most TSA checkpoints, the TSOs who operate the X-ray machines sit or stand next to the X-ray machine, evaluating the X-ray image being displayed. But that could change with remote screening or, more accurately, “remote image analysis.”
Inherent to ASL systems and as an adjunct to existing X-ray/CT systems, there exists a capability to remotely screen passengers’ assessable property. With remote image analysis, images are delivered to a remote location via a secure network to a TSA operator for analysis. The TSO is no longer stationed next to the X-ray machine but is now working alongside other TSOs in a remote “image analysis center.” This approach is already used for checked baggage screening, and it provides several benefits for security and passengers. Operators no longer have to wait for images to appear on their screen – they are delivered as soon as the operator is ready, which improves TSO productivity and checkpoint throughput. Remote image analysis also eliminates a touchpoint between the TSOs and the traveling public and is expected to increase operator efficiency due to the removal of operator distractions that occur at a checkpoint. TSA is piloting this capability at several airports and we would expect to see an increased level of remote screening over the next few years.
Another capability that will become more mainstream in aviation security is Artificial Intelligence and Machine Learning.
Today, AI capabilities are used in the automated threat detection algorithms used to screen checked baggage. Known as ‘Image on Alarm,’ AI helps detect suspicious items, and only those checked bags that have or possibly have a threat object get displayed to the TSO for a decision. The result is that the majority of checked bags are automatically cleared allowing the TSOs to focus their efforts on a very small percentage of potential threats and those gray-area instances.
Although the situation is a little different with accessible property screening (TSA is not only screening carry-on bags for explosives, but also for prohibited items such as knives and guns), the ‘Image on Alarm’ capability, given the improvements in AI, will likely soon be used at the checkpoint, via AI-enabled checkpoint CT systems. It is expected that this capability would dramatically improve checkpoint efficiency given that only a small percentage of carry-on bags would be flagged for TSO intervention.
5 years: Building on the Foundation of Remote Imaging Analysis
In five years, it is possible that TSA could conduct remote image analysis for an entire airport. Digital images from every CT/X-ray system from every checkpoint screening lane throughout an airport could be sent to a remote screening center located on airport property. And by using an AI-enabled “image on alarm” approach, only a small subset of images would have to be sent to the remote screening/remote image analysis center thus reducing the bandwidth and latency requirements needed for this capability.
Taking the concept further, TSA could establish a remote imaging analysis center to support the accessible property screening at multiple airports in a given region, e.g. Washington Dulles, Washington Ronald Reagan, and Baltimore Washington International airports. If implemented at that scale, the operational benefits – improved throughput, better use of TSO resources, better use of airport real estate – for both TSA and airports could be quite significant.
10 years: Self-Service for Expedited Travelers
Alongside the DHS Science and Technology (DHS S&T) Directorate, TSA is conducting research and talking to industry about the feasibility of bringing self-service concepts to the passenger screening process. The goal is to enable a self-sufficient checkpoint security screening process that improves the expedited passenger experience. It is envisioned that passengers permitted in the expedited screening process would submit their accessible property for screening and receive a pass/fail indication. If a failed indication were provided, the passenger would be shown what is causing the alarm and be allowed to self-resolve that alarm by removing the item. The accessible property would be resubmitted for screening. The expected potential is for a reduction in bag searches (secondary screening), which decreases the amount of time a passenger is in the screening process.
Although this capability is in the very early stages of the systems engineering process, it is foreseeable, given successful laboratory tests, that prototype technologies could be seen at airports in the next few years.
https://www.dev.hstoday.us/subject-matter-areas/airport-aviation-security/three-ways-covid-19-could-revolutionize-tsa-checkpoints/
