Hasn't anybody seen me lately, figuratively speaking, since most of you who read this never see me; others have seen me. What I should say is "hasn't nobody read much from me lately" but that's not in the lyrics to Randy Newman's song. So it's "hasn't anybody seen me lately."
I'll tell you why. I've been working. I've been working in New York where things went pretty well. And I've been working in Kazakhstan where things didn't go well at all. Tried to do the right thing in both cases for the clients, of course. And had great success in New York and no success in Kazakhstan. But that's just the way it is. You do your best, and if it's not your best that's wanted: Walk away. Don't wave, don't say "it's been fun." Just walk.
In the immortal words of The Contours, "Now I'm back to let you know, I can really shake 'em down."
So where are we? We're sixteen months from the implementation date for PTC and we're about two weeks beyond August 22, when FRA released its "Final Rule," actually, the second final rule, FRA having issued its first final rule in 2010, but some rules are more final than others regarding 49 CFR 236, Positive Train Control.
This final rule, in response to a petition from AAR expands the lines that will be made exempt from the requirements of the RSIA 2008, and expands the service that will not require PTC equipped locomotives on PTC equipped lines.
Why would FRA expand these exemptions? Is it because FRA provides evidence that the technological platform on which PTC is based is not reliable? Although FRA refers to such a concern, FRA introduces no data for such a concern, and FRA is, or is supposed to be, "data driven." You can look that up on the FRA's website.
Is it because FRA hopes that by expanding exemptions, the Class 1s might achieve compliance with the mandated PTC operational date of December 31, 2015? No. FRA has expanded its exemptions to the law based upon AAR's petition, which petition made no mention of reaching compliance, or substantial compliance by that date. FRA's discussion to the final rule makes no mention of achieving compliance.
FRA expanded its exemptions on the basis of, and only on the basis of, its logic or lack thereof in establishing de minimis exceptions to the application of the law.
Did I just intimate that there is a lack of logic to the FRA's concept and application of de minimis exception? Am I suggesting that a "data driven" organization like FRA doesn't have any data to support its de minimis exceptions? Yeah, I am, and I'm not alone. AAR's petition which has precipitated this review and expansion of de minimis states, that "there is no underlying logic" to FRA's de minimis exception for tracks based on density of service, that is traffic less than 15 millions of gross ton-miles per year; nor is there any logic for the exception based on class of track (classes 1 & 2 with authorized speeds not exceeding 30 mph). Not only does the petition say that, it proves it.
So if there's no underlying logic to the de minimis restrictions as they were originally composed, then what's the point of expanding the de minimis restrictions? What's the logic there?
Let's be clear, FRA brought this upon itself by attempting to reintroduce via regulation, what Congress had already deemed irrelevant by legislation: that is to say a cost-benefit analysis.
The RSIA requires the Secretary of Transportation (or designee) to do many things. It does not require the Secretary to establish a cost-benefit matrix to justify, reduce, isolate, or minimize the tracks upon which PTC must be installed. That is defined in the law as main line tracks on which an intercity or commuter passenger service is operated or PIH or TIH materials are transported. Main line is defined as a rail line handling more than 5 million gross ton-miles annually. Cost-benefit analysis does not apply. Cost-benefit analysts need not apply.
Simple, right? Ah yes, but it's the simple things that give us the most problem. Back in the day, back in the switch position awareness days, it was the railroads who were pumping for a de minimis exemption based on lower speeds of some territories from the cumbersome, tedious, and time-consuming requirements of switch position awareness. And back in that day, FRA was rejecting, and categorically, the argument that lower speeds ipso facto could be considered a satisfactory substitute for enhancing operating safety procedures in the face of a trend in accidents. What a long time ago that was. Hell...8 or 9 years at least.
AAR, acting in its self interest however, is right this time. There is absolutely, positively no underlying logic to FRA's de minimis exceptions. Then. There is absolutely, positively no underlying logic to FRA's enhanced de minimis exceptions. Now. No matter how vindicated AAR may feel.
The Congress defined PTC by function: an interoperable system that will prevent train-to-train collision, overspeeding, unauthorized entry into a work zone, operation through an improperly lined switch. The system must be designed to fulfill these functions 100% of the time; not 99%, not 98%. The design must be for 100% operating functionality. That's why we call it Positive Train Control. The system positively prevents these accidents.
In railroad operations "positive" has a specific meaning and that meaning is.... absolute. We speak of positive or absolute block, meaning the section of track between two designated controlling points is known to be clear of all trains, engines, and equipment and its use is now reserved exclusively for one train, or one locomotive, or one set of MOW equipment.
We speak of a "positive stop" signal, an absolute stop signal, as in "you, the train crew, must absolutely, positively stop to the rear of this signal or face dismissal in all capacities. You will stop and stay until hell freezes over and beyond and even then you may not proceed without the permission of whomever has the best long underwear in the operations control center." There are no exceptions.
That's how positive train control is supposed to differ from automatic train control. It enforces the positive, absolute stop. It absolutely enforces all speed requirements. It absolutely enforces exclusions from work zones. It positively prevents the operation of a switch improperly lined. In all cases but the "overspeed" condition, it enforces a zero velocity requirement until and unless authority has been transmitted releasing the zero velocity requirement.
Does that sound to you like a system that Congress intended to be subjected to de minimis cost-benefit analysis? Doesn't sound that way to me. We may argue about the wisdom of Congress in passing this legislation. We can argue about the necessity for such legislation. We cannot argue about what the legislation requires and where it requires it. We cannot subject PTC systems to cost benefit analysis, when the cost-benefit analysis has been found to be an inadequate basis for evaluation by an act of Congress.
I'm kind of on the side of Congress here, in all its wisdom or lack thereof. I think it's painfully clear to the most casual observer that Congress wanted to absolutely, positively prevent repetitions of the accidents at Chatsworth, California and Greenville, South Carolina. I think Congress intended that absolute, positive control eliminate such accidents wherever they might occur, not simply from track where they were more likely to occur. The "P" does not stand for "probable;" it does not stand for "permissive."
But all that....all that's just academic at this point. So what do we get with our new improved expanded, without logic de minimis final rule? We get: 236.1005 (b) (4) (iii)(A)-- a railroad may request a review for an exemption of any track not used for commuter or passenger rail service based on a minimal quantity of PIH materials traffic. and (B) FRA will approve a request for exemption for rail segments that (1) carry less than 15 million gross tons annually (2) that hoes not have a heavy grade as defined in 232.407 (3) where the railroad adopts and complies with an operating rule requiring the crew of any train approaching working limits extablished uner part 214 to notify the roadway worker in charge of the of the train's approach at least 2 miles in advance, or......as soon as practical. Let's stop right there for a bit (I'm losing track of the ( )s and the ( )s).
Wait a minute, adopt an operating rule? PTC is supposed to prevent human error. Adopting an operating rule instead means relying on human compliance, and thus does nothing to positively, absolutely eliminate the very reason the legislation was enacted-- human error. Adopt a rule? Well, let's see. Metrolink in California had a rule that said when the interlocking signal displays "stop," the train must stop before any part passes the signal. And let's see, Greenville, South Carolina-- there were operating rules there too-- restoring a main line switch to the position for movement of traffic on the main line. Checking the switch to make sure it is lined and locked properly. Not reporting "clear" of the main line before the switch has been properly lined and locked.....we got bushels of rules. And they're all good rules. Every one of them. I love them all. Saved my life, a few times.
But....but once upon a time I was in Egypt working on their railroad, pointing out the vulnerabilities of the British token block staff, and tokenless tyre systems when the Chief Engineer of Signals said to me: "But David, if everyone obeyed the rules, the system is safe." And I said: "If everyone obeyed the rules, you wouldn't have needed to hire me."
Get it? He did. So why, after human performance, rule compliance, has been identified as inadequate to the public safety, do we now suppose that human performance, rule compliance, is an adequate substitute for PTC? Beats the hell out of me, comrades.
Wait, I'm almost being unfair. FRA insists on more protections than just a new rule. Sure thing. The line must carry less than 100 loaded cars of PIH annually (excluding cars that are "empty," carrying only residue); no more than 2 trains per day can carry any quantity of PIH; and where speeds do not exceed 40 mph; and where any train transporting any quantity of PIH materials is operated with a vacant block ahead of and behind the train. Hmmh...........OK, how do we enforce any of this? Well railroads will most generally comply with the 100 car annual/2 train daily limit, because FRA can conduct an audit and fine the railroads, or remove the exemption. But we cannot prevent railroads from making mistakes, and misclassifying a car. Hold on, brings me to another story.
Once upon a time, way way back when I was just a trainmaster at a hump yard, we pulled out of the bowl a track of rehandles. Yes we stuck in our thumb and pulled out a "candy striper"-- that is a tank car of hydrocyanic acid, a material so toxic, that if you smell it, it means you're already dead and so is everbody else. These cars are designed to withstand tremendous heat, pressure, and impact. The shell of the tank is completely surrounded in foam and painted with red stripes, hence the name. You don't hump these cars, you don't hump against these cars, and here we were pulling a cut of rehandles out of the bowl, and the candy striper was 15 deep out of 30 cars.
While the clerks on the level below the hump tower ran for their automobiles; while the retarder operator crossed himself; and whle the conductor on the job asked me what I wanted to do with "this?" I entered the car number into the system computers to extract the waybill information. You know what came up? The car had been entered as an empty tank car last containing liquified petroleum gas and was waybilled to Bayway, New Jersey, where it definitely did not belong. Scary, huh? Mistakes happen. Human error happens. Somebody entered the wrong data.
Establishing an absolute block ahead of and behind trains carrying the PIH? Who does that? The train dispatcher. How does the train dispatcher do that? In CTC territory, the dispatcher can hold the PIH train at a stop signal until all trains ahead are a "block ahead." And the dispatcher can hold all following trains at the signal until the PIH train clears the block. Compliance with the stop signals however, absent PTC, will not be enforced, and once again we're back to human compliance.
In "dark territory," the situation is a bit more fragile, as the train dispatcher must receive reports from the train crews themselves that they are clear of a point designated as the end of one block and the beginning of another. A dispatcher then issues a written directive to a following, or opposing, train to enter that block. There are a couple sources for error here. The dispatcher can issue the dreaded "lap order," in which trains are given authorities that overlap each other, allowing two trains to be in the same place at the same time. The crew can make a mistake in executing its authority, overrunning its end point, entering another block without authority. Once again the so-called de minimis exception is made for an operating condition dependent upon the very weakness PTC was designed to eliminate.
Don't go away. This is only part 1. Much more to come.
September 4, 2014
Someone very dear to me has made another clerical error
Randy Newman, "Harps and Angels," 2008