3 ms·
> The problem is that it's a very time-consuming and expensive fix to repair and/or replace all the signals. There must be some period of time with low traffic
by nisa 8y ago
> The problem is that it's a very time-consuming and expensive fix to repair and/or replace all the signals.
There must be some period of time with low traffic - put a few managers and signal-guys in the train and rock the line at maybe +5% of regular advertised speed. Count where and how often the train stops. Fix that. Do that for all lines.
Honestly if that is the real problem that would probably take them a few weeks at most.
- jrochkind1 8y agoIt sounds like the signals may not really be _fixable_, from rohansingh's comment nearby. They are very old fully mechanical devices, you could send someone around to reset them, but there's no guarantee the reset would actually _work_, they might remain unpredictable. After all the problem is that they _tried_ to set them to a new value, and it didn't work. Perhaps they were working fine in the same configuration they were in for decades, but once they were touched to change the settings, they are just entirely unreliable. I dunno, but ancient mechanical parts can be like that.
- nisa 8y agoWe are talking millions of dollars. There must be someone who can figure that out - even if the failure is kind of random it follows probably a distribution and you can set it to a value that has enough margin to stop the train and not trip on slower trains. But you are right - easy to look and comment from the outside, there is probably a reason it is how it is.
- jrochkind1 8y agoWell, yeah, clearly they GOT to fix/replace them. I have no idea what the budget for that would be, but seems plausibly outside of _easy_ fix.