4 ms·

Saw this interesting Twitter thread with some data yesterday. In fact it doesn't seem derailments are unusually high compared with history. It seems they may ev

by erentz 4y ago

Saw this interesting Twitter thread with some data yesterday. In fact it doesn't seem derailments are unusually high compared with history. It seems they may even be decreasing. https://twitter.com/mike10010100/status/1627344918220767232 https://twitter.com/mike10010100/status/1627344918220767232 The complaints about precision scheduled railroading are valid though. Every foamer has been talking about it. And again as your link shows this derailment (like many) had absolutely nothing to do with the overturning of the breaking rule that has frustratingly grabbed everyone's attention. (I guess everyone loves a simplistic explanation that blames government officials even if it's wrong.)

rhaway84773 4y ago
If it’s true that rail companies are running fewer, but longer, trains then wouldn’t absolute derailments decrease even if per ride derailments have increased? And longer trains derailing means that the impact will be more spectacular, which would explain why we are seeing more news coverage of these fewer absolute derailments.
- erentz 4y ago
  Maybe not? If what we care about isn't derailed trains but the actually derailed cars, like say lost packets on a network. Here's a graph of total derailed cars. He has a lot of graphs I haven't reviewed all of them. But might be some other interesting ones in there. https://twitter.com/mike10010100/status/1628072318491668500?s=20 https://twitter.com/mike10010100/status/1628072318491668500?...
  - macinjosh 4y ago
    Unlike a lost network packet a single derailed car can slow down the whole train and any trains behind it on either side. By using longer trains they are putting more eggs in one basket. On top of that longer periods of stress on a particular part of rail due to the longer trains is probably uncovering new failures in problem spots especially if heating is involved in the problem. Longer train = more time for a piece of rail to heat up. I think the metric of concern would be total consecutive time the rail is loaded or something like that.

by erentz 4y ago

rhaway84773 4y ago

If it’s true that rail companies are running fewer, but longer, trains then wouldn’t absolute derailments decrease even if per ride derailments have increased? And longer trains derailing means that the impact will be more spectacular, which would explain why we are seeing more news coverage of these fewer absolute derailments.

erentz 4y ago
Maybe not? If what we care about isn't derailed trains but the actually derailed cars, like say lost packets on a network. Here's a graph of total derailed cars. He has a lot of graphs I haven't reviewed all of them. But might be some other interesting ones in there. https://twitter.com/mike10010100/status/1628072318491668500?s=20 https://twitter.com/mike10010100/status/1628072318491668500?...
- macinjosh 4y ago
  Unlike a lost network packet a single derailed car can slow down the whole train and any trains behind it on either side. By using longer trains they are putting more eggs in one basket. On top of that longer periods of stress on a particular part of rail due to the longer trains is probably uncovering new failures in problem spots especially if heating is involved in the problem. Longer train = more time for a piece of rail to heat up. I think the metric of concern would be total consecutive time the rail is loaded or something like that.