The Cincinnati/Northern Kentucky International Airport (CVG) is expanding its long-standing relationship with TaskWatch to help automate manual processes and gain insight into its complex operations.
Adding TaskWatch’s computer vision platform that works in conjunction with AWS Panorama, the Airport sees, adds, and explores distinct use cases for advanced artificial intelligence (AI). TaskWatch is leading the way for continuous improvement in capacity management and productivity, which are vital contributors to the overall customer experience.
Brian Cobb, CIO at CVG, explains: “CVG Airport is committed to providing a world-class traveler experience through continuous innovation and strategic partnerships. Using TaskWatch’s application on AWS Panorama, we can bring computer vision to our existing IP cameras to automatically monitor congestion for over 70,000 square feet of airport traffic lanes.
“Once TaskWatch detects an issue, such as a disabled vehicle, TaskWatch sends real-time alerts to airport staff so they can provide assistance, keep the traffic flowing, and reduce delays for our passengers,” says Cobb.
By deconstructing the customer journey and identifying customer pain points, CVG can inform its workforce and positively enhance flyers’ and visitors’ transit experiences through evolving technology.
By coupling detection of events using computer vision with automated work processes, the Airport’s workforce can address passenger experience, safety, and security issues previously not known or addressed in a timely fashion.
For administrators and facility managers both in and outside the industry, this post will show how innovating with computer vision and connecting to workers via wearables can help overcome capacity challenges and accelerate improved outcomes at a low cost.
Solution Overview
TaskWatch is a tech startup delivering solutions to connect the workforce with robots, autonomous material handling vehicles, Internet of Things (IoT) sensors, and software systems.
TaskWatch’s computer vision platform utilizes AWS Panorama and introduces capabilities to quickly and easily deploy custom AI solutions, and trigger automated work processes. TaskWatch brings the power of a standardized cloud platform for custom computer vision solutions that facilitates training, deployment, and management of AI models and workflows.
With the rollout of computer vision, CVG has been able to facilitate actions for numerous events or anomalies occurring on a single video stream that can cover large physical spaces.
AWS Panorama is a machine learning (ML) appliance and software development kit (SDK) that allows organizations to bring computer vision to on-premises cameras to make predictions locally with high accuracy and low latency. TaskWatch utilizes AWS Panorama to implement custom computer vision solutions to improve operations across a variety of industries, such as retail, hospitality, and industrial.
The underlying orchestration of the AWS Panorama edge appliance and TaskWatch cloud running on AWS is as follows:
Figure 1 – Computer vision inference and workflow process with TaskWatch and AWS Panorama.
Existing network video stream is accessed by the AWS Panorama appliance.
Operations business users work on tagging images and building the CV model using the raw video feed and automated tools available in TaskWatch.
Custom workflows and analysis graphical interfaces are configured within TaskWatch for users to respond to events observed by the CV model on an ongoing basis.
To illustrate everything detailed prior, let’s run through scenarios deployed at CVG.
Traffic Flow on the Airport Curbside with Computer Vision
One deployment of TaskWatch aids the Airport in the management of security concerns and vehicle flow on the terminal curbside.
The goal is to ensure passenger and commercial vehicles do not remain stationary for extended periods of time. Should vehicles become stationary longer than anticipated or acceptable, traffic flow slows dramatically, and capacity concerns escalate. This can result in frustrated drivers and customers wanting to be dropped off or picked up.
With targeted information, including relayed images on wearable device, the limited number of curbside attendants can prevent bottlenecks and disruptions to traffic flow in a more efficient timely manner. A steady-state flow provides for a meaningful reduction in the consumer anxiety factor associated with air travel.
When the Airport was first designed, more than 80% of passengers were connecting out of CVG. The remaining 20% of passengers came through the terminal, or the Airport’s “front door.”
Translated in hard numbers, 20% was equivalent to 4.4 million passengers per year traveling through CVG’s front door. Now, approximately 95% of passengers start or end their travel through that same front door without any structural modifications.
Similarly translated using CVG’s 2019 year-end passenger volume of 9.1 million, this means capacity through the same terminal building, curbsides, and roadways handle over 8.6 million passengers, nearly double the peak activity in the Airport’s previous connection-hub days.
Faced with limited infrastructure and the reality of not being able to build a new terminal or adding additional roadways, CVG turned to advancing technologies to manage capacity constraints while not sacrificing the integrity of its commitment to customer service and brand reputation.
TaskWatch captures an existing video feed of the curbside traffic and, with the help of AWS Panorama, analyzes that video, looking for event-driven specifics defined by Airport operations leaders. Computer vision algorithms are then run on the edge computing device.
Once TaskWatch identifies an event, such as a vehicle that has remained stationary for too long, the system sends a picture and time stamp to curbside attendants equipped with a TaskWatch wearable device to initiate the appropriate response.
Figure 2 – Computer vision workflow for real-time targeted action.
In this stationary vehicle scenario, the attendant approaches the driver of the car that’s blocking the traffic flow and provides factual information in a non-threatening manner to get them to drive and move ahead. If the analytics identify no driver present, then alternate security protocols are initiated.
Computer vision further provides the Airport with additional reporting details and analysis for further consideration. On average, there are 12.8 such notifications per hour of cars that have remained stopped and not in the mode of loading or unloading. This is about one notification per five minutes.
Backend calculations of the amassed data points support tailored time-of-day scheduling to determine the necessary number of curbside attendants. Additional details come to light, such as peak-hour traffic, number of cars per square foot available, impacts on through-lanes, and anomalies previously not experienced or viewed by a curbside attendant.
The traditional resolution method would be an extensive and costly traffic study that may or may not include customer or worker feedback or the real-time indication of fix or failure. Now, CVG is generating a continuous live traffic study with the added capability of event-based learning that improves and enhances the efficient use of the Airport’s limited curbside capacity and addresses issues on the spot.
As TaskWatch captures more event data points, predictive analytics will minimize or prevent future occurrences. Armed with this resource, the Airport is better prepared with computer vision and AI to improve user experiences while enhancing its imperatives of safety and security.
Efficient and Timely Aircraft Turn on the Tarmac
TaskWatch and AWS Panorama are also used at CVG to observe and analyze service activities conducted on each arriving and departing aircraft gate.
By analyzing the timing of each ground handling event of the airplane’s arrival and departure, the Airport and tenant airline can remotely monitor each flight in real-time or replay. This analysis supports the Airport and airline’s mutual brand commitments to improve response time to missing or delayed work steps, provide for post-event service level performance reporting, and predict and prevent delays with the mass data collection.
Improving overall operational efficiency and timeliness of response is game-changing for enhancing the customer experience for the masses.
To make this technology work, TaskWatch captures an existing video feed observing an aircraft docking or undocking at the prescribed gate. The video feed analyzes such activities at each service event (such as load baggage), notifies appropriate ground personnel of event-driven concerns (bags not yet loaded, for example), and generates a prioritized response to prevent a delayed departure.
Figure 3 – TaskWatch and AWS Panorama analyzing activities on the tarmac and initiating workflow processes.
The traditional work effort relies on visual cues and radio relays among specific team members, leaving others potentially left out or the information misinterpreted. Computer vision automates the same process over multiple and simultaneous aircraft arrivals and departures, something that a widely dispersed work team over multiple aircraft is challenged to do under the best of circumstances.
Tracking to this level of detail ensures service activities run smoothly, and passengers have the best possible experience regarding up-to-date information that pertains to their travel.
Conclusion
Using state-of-the-art technology, user-defined events, and identification of operational exceptions, TaskWatch is redefining the way Cincinnati/Northern Kentucky International Airport (CVG) gains efficiencies through automation.
Demonstration of successes and quick deployment of custom computer vision solutions have opened doors for additional business use cases with TaskWatch.
Additionally, backed with the stability of the AWS Cloud and AWS Panorama hardware, TaskWatch can ensure product delivery is reliable, redundant, and scalable to meet future-defined needs.
At CVG, TaskWatch and AWS Panorama are doing just that to maximize advancements between computer vision and artificial intelligence to minimize operational deficiencies and excel in customer experience.
.