I liked, and like, the railroad operating rules. I liked, and like, knowing that there is a right way to do things, and that some things in this life really are black and white, and that not all questions are questions of perception.
I hope you know what I mean.
I have always thought that the logic of the railroad-- that train movements could be generated, authorized, and executed only to the extent that other conflicting or overlapping authorities were precluded, gave railroading a certain rigor; a vitality.
Archimedes had his lever and and if you gave him a place to stand, he thought he could move the universe. Give me a train sheet, a place to sit, a block line, and the rule book, and I'll deliver it safely by rail.
I like that conducting this business requires that we think ahead of the trains, where they will be, what will happen and that we make that determination based on our previous experience, the current conditions, and our knowledge of the rules.
If that sounds like an algorithm to you, it sounds like an algorithm to me too.
So..so I was reading The Wall Street Journal just the other day, Thursday January 9, and there was the page one article "Fiery Oil-Train Accidents Raise Rail Insurance Worries." It's here behind a paywall, but it's certainly worth reading.
At the same time, a friend and colleague, Michael Savchak, loaned me his copy of Beyond Hidden Dangers-- Railway Safety into the 21st Century by Stanley Hall. This book continues the analysis of train accidents in Britain that the author had initiated in a previous volume.
Michael himself has compiled his investigations into the origins of train control systems in the US into a book, History of Train Control, Cab Signal, and Train Stop Systems, copies of which he is willing to provide to interested parties. He can be reached at: email@example.com
Mr. Hall places the examinations of the accidents, at Southall, Ladbroke Grove, Hatfield, Great Heck, and Potters Bar in the context of the break-up, and break-down, of British Rail by the privatization of rail service in Britain, and within the context of his own service to and knowledge of British railroading, having worked his way from stationmaster to divisional operating superintendent to British Railways headquarters as the Signalling and Safety Officer.
During British Rail's last years of existence, 1990-1996, the railway recorded 8 passenger fatalities, the lowest number for any similar span of time in its history. In the six year period immediately following privatization, 52 passenger fatalities occurred. Of course, such simple comparisons are only that, simple comparisons, but remember what I said about the rules, and the algorithms we derive from the requirements of the rules?
Sure you do, and it means that there are reasons for sudden, rapid changes in the trend of train operating safety. It means that somewhere along the line the reservoirs of past experience, the attention to current conditions, the rigor of thinking ahead of the system, and the vitality of rules has been degraded. The algorithm has broken down, not because it's an obsolete formula, but because the inputs necessary for it to guide decisions aimed at resolving conflicts before they become problems have been ignored, if not abandoned.
Considering the 1990-1996 period, Hall states:
British Rail had a very strong operating organisation which ran from top to bottom, from British Railways Board Headquarters to drivers and signalmen and others at the workface, and similarly from bottom to top. It was a very effective chain of command...Each level in the organisation was responsible for day-to-day safety in its own domain. It was a well-tried organisation which in essence dated back to the 19th century, and it worked.
Keep that in mind, the next time you read about how railroads in the US are encumbered with this 19th century quasi-military organizational philosophy complete with a "chain of command" and operating rules that "get in the way of safety."
Mr. Hall also says this, and on the same page:
"Old railway men used to judge the seriousness of an accident by its potential, not actual, consequences. The latter are in the hands of fate. We seem to have lost a measured professional reaction to railway accidents in recent years."
That's almost too much insight for one person to put in writing on a single page.
"Judging by its potential, not actual, consequences" is an essential component of our algorithm. Once the rule violation or unsafe act occurs, we cannot control the consequences.
By the very make-up, the meaning of vitality, we cannot control those circumstances. That's why we have the operating rules; that's why we utilize a vital logic. We have to proscribe, exclude, the act in any manifestaton to protect the operation.
Keep that in mind next time somebody wants you to cut a deal for a train crew that ran a stop signal, arguing "nobody got hurt,"and after all, "he or she won't ever do that again." Or somebody thinks stop signal violations should be covered by close-call confidential reporting.
Actually we shouldn't have to keep that in mind. It should be imprinted upon you before you ever make it to supervision.
Anyway what does all this have to do with DOT 111 tank cars, Lac Megantic, Casselton and other locations? Well, the WSJ article has some interesting information. Like? Like that the cleanup is costing $4 million dollars a week, with the final cost of the cleanup possibly reaching $200 million. That the carrier, the Montreal, Maine, and Atlantic Railway Ltd. carries only $25 million in liability insurance.
The MM&A has filed for bankruptcy all the better to protect its assets and that of its parent, Railworld, from those impacted by the fire and explosion.
Which means? Which means that the Canadian federal and provincial governments are paying for the cleanup. And?
And separate and apart from the cleanup of the city itself, are the damages to over 100 private business and homes, and the loss of 47 lives.
And? And this is important because (a) there isn't enough liability insurance in the world to cover the possible damages from a catastrophic rail event involving hazardous materials. Total liability coverage for the industry is $1.6 billion (b)somebody has to pay, actually sombodies who collectively, as an institution don't have the option of declaring bankruptcy, and those somebodies are the public, known as taxpayers (c) that in the world of cost-benefit analysis, damages and social costs of train accidents are not figured as part of the "costs" since these are either covered by insurance, of which there is not enough, or the parties identified in (b).
Keep that in mind next time someone starts whimpering about an "unfunded mandate" and "cost-benefit."
I bring this up because for 25 years, we've known about the weaknesses of 111 tank cars, and "cost-benefit" arguments continue to be utilized as arguments against mandating improved design standards for these cars.
I bring this up because the cost-benefit analysis directly contradicts the vitality of our algorithm for safe train operations.
The cost-benefit analysis says we don't have to exclude, to act in accordance with excluding the unsafe occurrence.
The cost-benefit analysis presumes we will know the actual consequences of the event; that the seriousness is measured by the actual consequences rather than the potential consequences
While we toss around buzzwords like "robustness" "pro-active, not reactive" we are inclined to use "cost-benefit" as a rationalization for inaction and weakness.
My friend, Michael Savchak says that part of the problem is the enhancements in safety brought about by the already existing systems make it that much harder to justify the expense of additional improvements. He calls it a "classic conundrum." Indeed it is. It's the old law of diminishing returns.
However, I'm not here just to complain about things. I'm here to help the Class 1s overcome the "unfunded mandate" of PTC. And I think the mechanism for funding that mandate might be applied to funding the redesign and construction of new, safer tank cars.
And it goes like this...
Some fun with numbers: (1) companies ship about 150 million tons of hazardous material by rail annually (2) average length of haul for freight on US railroads is about 950 miles per shipment (3) this works out to about 142 billion ton-miles of hazmat haul per year (4) current estimates of the cost of PTC installation on 60,000 miles of track in the US come to $10 billion (5) if railroads assess a 2 cent per ton-mile surcharge on hazmat shipments, the costs of PTC installation will be recaptured within 4 years. (6) if railroads desire, they can reduce the burden on the hazmat shippers by assessing the PTC surcharge to all shippers, just as they now assess a fuel surcharge to all shippers.
Look, somebody's got to pay, right? There's a cost. And the benefit is we preserve the robustness of our algorithm for safe train operations.
January 11, 2014
Safety is of first importance
These rules provide for safety of employees and the public, for efficiency of operation and for protection of the company and its traffic
Obedience to the rules is mandatory
In case of doubt, the safe course must be taken.