Tuesday, October 27, 2009

Was the Operations Research Crew Scheduling Program Just Too Interesting?

According to CNN.com as well as the New York Times, the reason that the Northwest pilot and first officer were unreachable for about 90 minutes on last week's flight from San Diego to Minneapolis (and overshot their destination by about 150 miles before turning around) was that they were engrossed in the crew scheduling software on their laptops.

Since Delta merged with Northwest, there have been obvious issues regarding the retraining of personnel (I have had long discussions with stewardesses on flights during which I was told how different the philosophies were of these two airlines pre-merger as well as the number of stewardesses that would assist on a flight).

According to news reports, the first officer was assisting the pilot with the crew scheduling software and clearly the subject was of sufficient fascination for both that they did not realize that they were to be piloting an airplane! Those of us who work in operations research are well aware of the underlying mathematical models and, coincidentally, today I was teaching integer programming models in my graduate class at the Isenberg School.

Of course, these two have now lost their pilot licenses. I remember a colleague of mine, Dr. Richard Stone, who is a Lanchester Prize winner and early in his career taught at the Kennedy School at Harvard and then left to join the Operations Research group at Northwest Airlines. You can read about some of his relevant activities here.

It is rather ironic that O.R. (operations research) which focuses on optimization of business processes, including crew scheduling, may have had a role to play in this major human error and only because the software was clearly so interesting that it was an obvious distraction to the flight crew. Why were there no "bells and whistles" in the automatic pilot software when the plane overshot its destination, I wonder?!

As for the merger of Delta and Northwest, locally, we lost our direct flight from Bradley airport (Hartford/Springfield) to Amsterdam, which was a terrific asset while we had it for about a year and a half.

I have done research on mergers and acquisitions in oligopolies, which airlines are, and you can find my latest paper on the subject, which is in press in the journal Computational Management Science here.