I rode the new system on the Red Line with my brother to see the fireworks on July 4, 1976. I was fascinated how beautiful and modern the subway system looked and still does. However, last week was the worst DC Metrorail accident in its 30+ year history. 9 people died and many were injured when a subway car rammed into another sitting on the tracks.
How could this happen? When this system started in 1976, it was deemed Fail-Safe and rail cars would be seperated from other cars when the system was run by computers in automatic mode. Last week the accident occured when the system was in the automatic mode.
I read an intersting article in the Washington Post today "Metrorail Crash May Exemplify Automation Paradox" http://www.washingtonpost.com/...
The article talks about the
"The better you make the automation, the more difficult it is to guard against these catastrophic failures in the future, because the automation becomes more and more powerful, and you rely on it more and more."
But this essentually leads to more questions on what exactly is Fail-Safe and what should be the role of humans as well as computers in an automated environment..
Metro has been on manual mode ever since the accident occurred last week. Being on manual mode means reduced speeds of 35 mph compared to maximum speeds of 59 mph in automatic mode.
The NTSB found a defect in the Electronic Control System where the accident occured, however local DC news reported that Metro was considering going fully automatic in a couple of weeks.. However, the defect may be much more serious than a broken sensor. (in my opinion)!! Important questions need to be addressed. Why was the broken sensor allowed to go unnoticed by Metro Maintenance personnel? Are broken sensors able to signal Maintenance or Control personnel that they are not working properly? If the system is truly fail-safe, why isn't there a backup sensor next to all sensors along the entire length of the system? Why is the operational computer running the Metrorail system unaware that a 4 rail car subway train has vanished and not been picked up by other sensors and why did the computer not protect for the place it should be?
Computers are far more advanced today than they were 30 years ago. Metrorail should modernize and upgrade the entire length with a new computer and sensor system that will protect space for not only the location of known railcars, but also locations of where railcars should be but are not detected.
Metrorail should not hurry its way back to automatic mode when these questions remain unanswered leaving a system in place that is questionable at best..