Thursday, April 17, 2014

Dear NTSB . . . . .

Written by 
The National Transportation Safety Board recently released an open letter to MTA Metro-North Railroad containing its initial safety recommendations derived from the ongoing investigation into the Dec. 1, 2013 derailment of train 8808, which killed four passengers and injured 59.

The recommendations are that Metro-North systematically install “approach permanent speed restriction” signs, and presumably “permanent speed restriction” signs and “resume” signs along its right-of-way. Although the NTSB does not mention specifically the latter two fixed indicators, those who are conversant with and responsible for rail operations know we don’t, we can’t, do one without doing the others.

There is nothing wrong, and only good, in installing such fixed indicators, particularly along a railroad that has no fixed automatic signals save those at interlockings. It’s good to have physical reference points. The templates for many railroads’ operating rules—GCOR, NORAC, the old Consolidated Code—have provisions for the installation and use of such signs. There’s nothing wrong, and nothing new, in that. It costs little, it’s helpful, and of course, to be effective operating a railroad with permanent fixed indications, requires the exact same thing(s) that operating a railroad without these fixed indicators requires: proper training and supervision of employees, and proper enforcement in incidents of non-compliance.

The NTSB recommendations do not address those three core values that of course are the same core values for any, and every, function on any railroad.

Since that recommendation, helpful and innocuous as it is, doesn’t address the root values, the basis for preserving the vitality of train operations, it’s not likely to have any real significance for the safety of employees and passengers.

In making its recommendation, NTSB gets some of the details wrong—stating for example that “As a result of the accident, Metro-North installed approach permanent speed restrictions signs to aid operating crews at the derailment location,” when in actuality Metro-North installed a code in the rail triggering a 30-mph speed restriction enforced by the on-board ATC system.

I know it’s just a detail, but those who are conversant with and responsible for safe train operations know that the devil—and the vitality of the railroad—is in the details.

NTSB also fails to mention the impact of FRA’s Emergency Order 29, specifically requiring Metro-North to provide additional protection in areas where decelerations of 20 mph or greater are required, and the mandatory adaptation of the train control system to enforce such decelerations.

Just another detail.

So much for the first recommendation: nice, helpful, good, innocuous—and superficial.

The second recommendation isn’t new either, but unlike the first, there isn’t a long history of the use of the “inward- and outward- facing audio and image recorders to assist in accidents investigations and with management and regulatory oversight of rules compliance.”

There’s some history, particularly with outward facing video cameras. These cameras have been quite valuable in recording incidents at grade-crossings, and signal violations. As a matter of fact, the Union Pacific freight train lead locomotive that was struck by the Metrolink commuter train in Chatsworth, Calif., was equipped with an outward facing video camera. I’ve seen the video, and if you haven’t, believe me, you don’t want to.

The point, of course, is that while the outward video camera can be helpful after an incident for reviewing the circumstances, and for confirming in all probability what we already know to have occurred from other evidence, it does nothing to prevent the specific incident.

Short version: Outward video cameras are of some benefit—a benefit limited to post-accident analysis.

I’m pretty sure that NTSB does not seriously intend for railroads to equip locomotives with “outward-facing” audio recorders. Just another detail I guess. But maybe not. So what about inward audio recordings?

Here we confront not just a recommendation of extremely limited value, but a recommendation that is pointless. First, commuter train operations are overwhelmingly single-person cab operations. Exactly what do we expect to hear from the internal audio recorder? Snoring? MP3 players? Radio transmissions?

Those of us conversant with and responsible for railroad operations know that railroad radio transmissions involving main line operations are already recorded and available for playback. So what role would an internal audio recorder play in either the investigation of an accident or a review of employee performance in order to identify failures and weaknesses, and thereby prevent a future accident?

The application of internal video recorders may or may not be pointless, but this is one case where I might apply a cost-benefit analysis, given the extremely paltry return on investment derived from the use of the data available from such recorders.

Let me explain: If the video record after an incident shows the locomotive engineer utilizing a cell phone, exactly of what significance is that to the current investigation, or future accident prevention, when the data is derived after the fact? We already have regulations against the use of such devices, and more important, we already know a certain number of employees are going to violate those regulations. So we should already know that the obligation of those conversant with and responsible for train operations is to oversee, confirm, and enforce, by and through our repeated and persistent physical presence, the employees’ compliance with the rules. You can’t do that “virtually.” So exactly what are we going to learn that we don’t already know?

Short version: You can’t supervise the railroad “virtually.” You actually have to be there. You actually have to know what you’re doing, what you’re supposed to do. You have to know the details.

NTSB does not recommend real-time monitoring of video recorder data—the transmission of images to a human observer. Nor does NTSB even acknowledge, much less recommend, that new locomotives and MU locomotives are capable of real-time transmission as to location, speed, braking effort, etc. It would make a least some sense to monitor those operating parameters in real time, even if the monitoring would present an unmanageable burden to the railroads, given the person-hours required to provide that monitoring. Such monitoring might actually prevent an accident. Indeed such “proactive” monitoring is the basis for the algorithms that will drive PTC systems. PTC is where we need to be, or rather, a PTC that defines the limits to one train’s movement authority as the rear end of the train ahead, not “restricted speed” at a signal location.

There’s another aspect of NTSB orientation that makes this recommendation superficial, and ineffective, at its best. That aspect is NTSB’s reluctance to acknowledge, support, and endorse a critical component of enforcement—the application of discipline.

NTSB has been explicit in its criticism of railroads’ “punishment-based” culture regarding employee failure. The Board has spoken against the “unjust culture” prevalent on railroads where a “good employee” can, for a first, but serious, violation of the operating rules, face mandatory suspension without pay.

NTSB isn’t alone in its uncharitable assessment of railroad disciplinary processes. Associates of the Volpe Institute have made similar evaluations. Apparently, FRA has also joined in the chorus as it intends, in its next IMOU for C3RS reporting, to include decertifiable offenses (like stop signal violations and exceeding the limits to a movement authorization) under the protective umbrella of C3RS.

Those of conversant with and responsible for safe train operations need to clear the air about this, and refute these characterizations whenever and wherever they arise. Let’s get this out of the way (this one time) in a hurry. The fact that a “good employee” can, after years of good service, suffer serious consequences for an operating rule violation is not an indicator of an “unjust culture.” It is an indicator that, indeed, the consequences for an operating rule violation in and of itself can be so much more severe than even the most severe disciplinary rendering. The consequences can be death, and death to more than just the employee violating the rule.

That’s how those of us conversant with and responsible for safe train operations judge discipline; that’s how we determine “justice” in our culture. We are actually doing the employee a favor. We are being lenient when we assess a 30-day or 60-day penalty for a serious rule violation. We are taking a real risk: that our decision to only suspend, rather than dismiss in all capacities, the employee will not come back to haunt us.

This risk is not confined to simply that individual employee returning to work and violating the same or similar rule with the “penalty” being assessed by the laws of physics against him/her and others who committed no such violation. The risk is that the employees collectively will disregard not the disciplinary consequences of a rule violation, but those consequences of the laws of physics if and when those of us conversant with and responsible for safe train operations act in anything other than the most serious, determined, and responsible manner.

Our use of progressive discipline is an attempt at leniency, and it is a conscious acceptance of risk and responsibility, something that some individuals—those not conversant with and responsible for safe train operations—may never understand.

Now that’s OK, if they don’t understand it. Just don’t them get in your way. Don’t let them stop you from meeting your responsibility in removing from service, permanently if conditions so warrant, an employee who represents a threat to the safety of himself/herself and/or others.

So the next time someone who has conducted studies on C3RS says to you, as was said to me, when you question the functioning and the benefits of such a program, “From your questions, you sound like that old-school railroader with that attitude we’re trying to overcome,” smile and just say “thank you,” because I’ll tell you a not-so-secret secret—the problems some of the commuter railroads are now experiencing are the results of years of trying to “get rid of” that “old-school railroader attitude.”

I have to tell you, if I were running a railroad today, Class I or passenger, facing the mandatory implementation of PTC (which I support) at the same time as I am being subjected to such conflicting and nonsensical pressures regarding enforcement, I really would look to technology rather than supervision for a solution. And I would look at all this news about drones, drones for deliveries, drones for fire departments, drones for everything, and I would put all that droning together and think how wonderful it would be if I could “drone” my trains—operate all of them no matter where they are remotely, with a joystick, from a single location. Let the track database, the train handling algorithms, the outward facing video cameras (giving your drone controllers a view of the right-of-way) actually give you a return on the investment, in this case the return being “no bang for the buck.” The controller is there only in case an event arises that the algorithm cannot handle—say a pedestrian trespasser walking along the right-of-way, not paying attention.

Imagine the “just culture” we could create. Imagine how little fatigue would be a factor in employee performance. Imagine how much we could save on “away from home” terminal status; on initial terminal delay; on crew calls; on arbitraries; on training and certification.

I know that’s years away and I’m a dreamer. But I’m a practical dreamer, and I want to make a practical contribution to accident elimination, so I propose the following. Since the MOU for C3SR programs specifically exempts “real-time observations” of rule violations from its protective mandate, because examination of real-time data can include immediate transmission from a moving train, because the amount of real-time data transmitted from moving trains presently exceeds the capacity of immediate railroad response, and because only 60,000 miles of the general railway network will be equipped with PTC systems, any review of data from any source—locomotive event recorder, digital archive and replay of control center’s record of train movements, or review of wireless train data transmissions, etc.—that provides evidence of an operating rule violation will qualify as “real-time observation” and will not be subject to C3SR protections.

That’s what I would do, thoroughly modern old-school railroader dreamer that I am, with a thoroughly modern old-school railroader dreamer attitude for which I’m grateful. I’d do my level best to put the NTSB out of business. And that’s an attitude the NTSB should welcome.

David Schanoes

David Schanoes is Principal of Ten90 Solutions LLC, a consulting firm he established upon retiring from MTA Metro-North Railroad in 2013. David began his railroad career in 1972 with the Chicago & North Western, as a brakeman in Chicago. He came to New York 1977, working for Conrail’s New Jersey Division. David joined Metro-North in 1985. He has spent his entire career in the operating division, working his way up from brakeman to conductor, block operator, dispatcher, supervisor of train operations, trainmaster, superintendent, and deputy chief of field operations. “Better railroading is ten percent planning plus ninety percent execution,” he says. “It’s simple math. Yet, we also know, or should know, that technology is no substitute for supervision, and supervision that doesn’t utilize technology isn’t going to do the job. That's not so simple.”