March 15, 2018

Analyzing A Week of Catastrophic Aviation Failures

Almar Sheikh by Almar Sheikh

The Real-Time Knowledge Engine has now been up and running for about a month and we’re starting to see some amazing progress in not only detecting events, but in understanding the entities involved in each incident. A great example of when this type of analysis becomes crucial is during the detection of “crash” event-types. While it’s important to monitor and analyze all types of crashes, often times a crash involving a passenger aircraft will likely be a higher crisis priority than a single-car collision. Isolating “entities” involved in the crash (like a car, train, or airplane) will become fundamental for the personalization of alerts and in understanding the severity of each event. We call this type of analysis Smart Tagging and last weekend we were able to see the impact of this new kind of detection in action.

The most severe aviation accidents often involve passenger planes, like the recent US-Bangla plane crash (make sure you’re logged in the view the alert) at Kathmandu’s airport on March 12th. These types of crash incidents typically make international news, as the ripple effects are felt by many. The airports, airlines, aviation bodies, and news organizations using SAM were alerted to this crash at 08:45 GMT, 14 minutes before the BBC and 1 hour and 23 minutes ahead of CNN. With a time advantage over traditional information sources, users are able to respond faster and more intelligently, and with Smart Tagging they are able to prioritize by perceived severity.

In a similar situation on March 11th, a Turkish plane hit the Zargos Mountains, crashing in southwestern Iran. SAM was again able to identify an airplane was involved and alerted users to the incident at 15:35 GMT, 10 minutes before the AP and 3 hours and 9 minutes before Euronews.While these are examples of larger-scale crashes, there are also many smaller-scale crashes that occur daily, never making major news cycles. By relying only on traditional sources of information, there is not only a time delay, but a subjectivity to each news organizations willingness to cover the story. Because SAM’s Knowledge Engine scans all public social chatter, we’re able to provide a faster and more balanced view of what is happening on the ground.

In just the past few days we’ve detected a U.S. Military jet crash in Key West, a private plane crash landing on a high-traffic boulevard in Kissimmee, Florida, a light aircraft going down just west of Winsford, England, and a tour helicopter capsizing in the East River in New York City. While these crashes received less overall coverage, there are major implications for different organizations like aviation authorities, aircraft manufactures, local officials, and tourism bodies. By Smart Tagging each event with the type of entity involved, users are able to gain quick, accurate context and react if the incident is relevant to them.Without human intervention, SAM’s AI is able to understand both the meaning and the types of entities involved in a particular incident. Providing users with as many data points as possible, to assess the personal importance of the situation. You can think of SAM’s Real-Time Knowledge Engine as your very own AI Analystinstantly distilling events occurring globally into a credible stream of information. In all the cases above, Smart Tags were able to help SAM users identify which crash types were most important and leverage the speed of the alerts to get ahead of the crisis situations affecting their organizations.

By continuing to use this site you consent to the use of cookies in accordance with our cookie policy. Learn more