Thursday, February 19, 2015
DC Metro tunnel smoke disaster: software, IT, and employee or contractor work quality at issue?
The recent fiasco over the Washington DC Metro accident January 12, 2015, concerning the wrong direction in which the tunnel fans directed the smoke from a third-rail insulation fire, has led, after an NTSB report, to questions about Metro Transit software and the fact that the system used by train controllers dates back to 2002. There were plans to replace it, despite numerous upgrades, and apparent awareness that it was difficult to use and could lead to operator errors.
The Washington Post story February 16 is by Mary Pat Flaherty, Paul Duggan, and Mary Aratani, link here.
It sounded as if the system did not present the information in a way that operators could rapidly use it. This sounded like the old systems-analysis problem where user requirements aren’t clearly stated in time when a system is developed. I have a relative who runs a company in Ohio that sells process-control software; I wondered if he knows anything about all this.
I had that experience in the early 1980s with a Blue Cross consortium, but turf protection among the clients was a problem that got in the way. That wouldn’t be a problem for a transit system, hopefully.
There’s also a question of the kind of staff and contractors hired. I would like to think that had I been working at Metro on that system, I would have known of the dangers and been meticulous enough to make sure this would not happen. Does this come down to the care of just one employee on his job?
We call this “attention to detail”. I’m reminded of an event in 1991 when working at USLICO (the foreshadow of ReliaStar) in Arlington, when another employee in production control found a major flaw in the elevation procedure that no one had noticed for over a year since getting a new source management system. That’s what having good employees means.