“Amtrak Crash and America’s Declining Construction Spending” is a story about an American worker ($200,000/year including pension and other benefits?) driving a train at more than double the speed limit for a given length of track (resulting in deaths, injuries, the destruction of a train, the suspension of train service, etc.). From this the New York Times infers that the logical next step is to give this group of workers tens of billions of capital equipment.
Related:
- “High-speed Rail in California versus China” (costs at least 6X as much per mile to build rail here)
- “New York’s new train station” (about an 11-year $4 billion project for one station)
- “Amtrak Bill Continues History of Wasted Subsidies” shows that it is just as hard for a corporation to get off welfare as an individual… and note that the $45 billion in tax dollars consumed over the years, if adjusted for inflation, would be more than double the cost of the 800-mile, 186 mph Beijing-Shanghai railroad (i.e., if Americans don’t have high-speed rail it isn’t because we haven’t paid for it)
I’ve taken Amtrak on the East Coast a number of times recently, and thought the northbound trains were running a little too quickly (and this is not the normal complaint with Amtrak, just the opposite). The accident didn’t really surprise me. But what is the argument here? That there shouldn’t be passenger rail in the Northeast?
I don’t think we know enough about the incident yet to place blame. Perhaps there was a broken signal along the track, or the visibility was poor so the conductor didn’t know a curve was coming up, or a speed limit sign was missing.
The article does a poor job of articulating this, but it’s possible that these things were missing/broken because there is insufficient funding to maintain the hundreds (thousands?) of miles of tracks. And if you think these train workers are so incompetent, let’s take them out of the equation by installing safety devices that limit the speeds of trains at certain parts of the track. Guess what, that requires money.
We are approaching the point where tractor-trailers can be made self-driving, but the technology for self-driving trains has been around for decades. It’s not a difficult problem because a railroad is a closed and controlled system. I’m sure everyone has been to an airport where there is an unmanned shuttle train between the terminals. The chief obstacle is probably union opposition, but the can be bought off.
At the very least, take away the engineer’s ability to control the speed of the train. At each station, the engineer would press a “start” button and then the train would proceed at the programmed speed profile to the next station. In the unlikely event the engineer spots an emergency (e.g. a car on the tracks) before the sensors do, he would press the big red emergency brake button but otherwise he could spend his time dozing and text messaging on his phone as he is already.
Once a couple of decades have passed without the engineer ever having to press the button, the existing engineers could be pensioned off and the position allowed to remain empty. There is precedent for this. Steam trains had a position called “fireman” whose job it was to shovel coal into the boiler of the locomotive and a “brakeman” who would go from car to car (walking on the roof of the train) and apply the brakes. After these functions became automated, the existing employees were retained (doing nothing) until they reached retirement but vacancies were not filled.
Son Dang – I live near Philadelphia. On the night of the crash, the weather was absolutely clear with perfect visibility. Any train engineer would have driven this route hundreds of times and it one of the few curves between New York and Philadelphia, so regardless of any signals, etc. he should have known that this curve was coming up (if he was paying attention which he clearly wasn’t). Although Amtrak has not yet implemented “positive” train control on this stretch, which has the ability to override the engineer (it is scheduled to be installed by the end of this year), it does have warning systems that sound alarm bells and lights in the cab if the engineer is speeding. It is possible that these systems didn’t work but I would be willing to bet that they worked just fine but that the engineer was distracted (on the phone) or semi-asleep. As a start, engineers should be prohibited from carrying phones or electronic devices while working.
Please don’t call a train driver an “engineer”. It is an offence to any engineer.
Most phones have GPS and can easily calculate speed, perhaps not as quickly as a speedometer, but say, an average over 15 seconds. How difficult would it be, to have it report actual speed to a remote location (central monitoring point) every 2 minutes? And cheap – such phones can be bought on Amazon for under $200, unlocked.
Paddy, the new positive train control system relies on GPS but it will costs many millions of $. As Phil can tell you with regard to commercial jets, for anything that involves mass transit you have to add several zeroes to the price. This accident alone is expected to result in hundreds of millions of $ worth of lawsuits (it would be more but Congress capped liability at $200 million per incident). If you were the maker of such a cheap system and it failed, you would get sued too, so the price of the system has to include that risk. So you charge $200 for the electronics and $200,000 for the liability risk.
Implemented positive train control is not a simple matter – it is meant to avoid more that just speeding. It also is supposed to prevent collisions so you have to know where every train is. Often trains from more than one company share the same rails (sometimes freight and passenger travel on the same lines, in the Northeast Corridor the local transit agencies use the Amtrak rails) so they have to intercommunicate. The cell network doesn’t offer 100% coverage, etc.
In PTC, the GPS knows where you are and the map knows the speed limit at that point. If you are over the speed limit, a warning sounds and then a few seconds later the system applies the brakes if you don’t. You don’t really need PTC to do this.
In today’s news, it was reported that on the southbound side of that same curve, automatic train control (a much older system that does the same thing) had been installed years ago, because the Phila. bound trains come into that curve from a 100+ mph straightaway. Northbound, the speed limit ahead of the curve was 70 mph (slowing to 50 in the curve) but even if the train hit the curve at 70 nothing would happen, so they never installed ATC on the northbound stretch. It didn’t occur to them that the engineer would begin speeding well before the curve and hit it at over 100. Contributing to this, the train was being pulled by Amtrak’s most up to date Siemens locomotive, which like any good German vehicle has great pickup.