Air Traffic Snarled Due to Single Point Failure

Thanksgiving is approaching with its heavy travel via planes, trains, and automobiles and what happens one week prior?! A single failure of a circuit board in the FAA's air transportation communications system snarled air travel yesterday. The full story can be found here. It took hours to fix and flight plans had to be entered manually in the meantime, resulting in numerous cancellations and delays throughout the US.

In the article, Professor Michael Ball of the University of Maryland is quoted. Ball is a long-time member of INFORMS and a fellow transportation colleague with whom I have had the pleasure of associating with through the Transportation Science & Logistics Society of INFORMS. The society's most recent newsletter can be accessed here.

Ball's quote: A good communications system system should have enough redundancy that a failure shouldn't hurt it that badly. Here we go again -- a single point failure with major disruptions in a network. This is another example of supernetworks -- a failure in communications affecting transportation.

The Federal Aviation Administration has struggled for years to modernize air traffic control and its associated systems. Coincidentally, while I was completing my PhD at Brown University with a specialty in operations research, I was interviewing with consulting companies, high tech companies, and academic institutions. I had received, interestingly, an offer from the MITRE Corporation in the Virginia area to work on the FAA's air traffic control system. I did not accept the offer (but it was so tempting) even though I was interviewed by a fellow Brown grad (who is now at IBM) Dr. Igor Frolow. Now there are again calls for more funds to fix the problems with the FAA. It's not a simple issue of increased financial funds, but, rather, of thorough planning and appropriate resilient and robust network design.