![maxresdefault[1].jpg](https://static.wixstatic.com/media/cef8ab_8d4e0ddce355463e8edabd7c8937b6f6~mv2_d_2500_1406_s_2.jpg/v1/fill/w_124,h_70,al_c,q_80,usm_0.66_1.00_0.01,enc_auto/maxresdefault%5B1%5D.jpg)
Uncontrollable Factors In Railroad Safety
What make trains safe, unsafe, and downright dangerous.
by Charles A. Turek
May 10, 2024----How safe you feel aboard any form of transportation depends on several factors, which we will discuss below; but how safe you are depends on design, engineering and manufacture of the transportation conveyance, I.e., the science of that particular mode's structure and infrastructure. How safe you are also depends on people, and railroading has always been a people intensive business.
How safe you feel and how safe you are do not necessarily coincide.
Let's look at the first half of my thesis, the emotional part, if you will, before getting into the science and the people.
How safe you feel depends on experience. I will cede to the argument that experience can be both emotional and scientific, but hear me out. If your parents took you on a train ride as a toddler, and you liked it, you probably feel safe on trains. If, however, your first memory of a train ride was traumatic, whatever the reason, you are likely to avoid travel by rail. This holds true for any form of transportation, but particularly for train and air travel. I couldn't locate any studies to put this in a scientific context, but I think it is the bigness of trains. Airlines used to have bigness, too, but now have bigness under glass. Gone are the days when you could get as close to a moving airliner as you can to a moving train. There are still those urban airports and places around the globe where a landing aircraft comes over the fence at a hundred feet, but that's nothing like standing with your toes to the yellow line as the daily limited goes through your station at seventy-per, or even at fifty-per.
Naturally, if you survived riding a train that derailed or crashed, you may feel unsafe on a train for the rest of your life. The same is true of other forms of transportation. The feeling is emotional, and has nothing to do with why the train you were on derailed or crashed. In fact, knowing why could make your aversion to trains even worse.
How safe you feel depends on communication. Let's start here with word of mouth, which today means not only spoken communication but also social media. A friend or relative has had a bad experience on a train and tells you about it. Conversely, the friend gives a glowing account of a wonderful train trip through the mountains. How safe you feel on a train will depend on these communications, particularly if they slant one way or the other. As a small child, I heard the gory details of how a maternal uncle was crushed on the open platform of a rapid transit car years before I was born. I have never been able to find a media account of this accident, but even without documentation of the event, I still hesitate to walk between cars on any train. And I am someone who likes riding trains.
Print and broadcast media seem to have a more subtle effect on how train riders feel about safety. This may have something to do with sensationalism in their news departments. It may also have something to do with the liberal media's reluctance to provide photographic evidence of failures in safety. Consequences of wrecked vehicles, be they trains or trucks or aircraft, are shown as property damage. Rarely are severe injuries or corpses shown in photographs. I realize this is partially out of respect for the victims, but I also find it somewhat hypocritical in a society that watches the same carnage for entertainment, whether fictionalized or true crime. The result may be that such accounts serve to make us more complacent and feeling safer than we would or should.
How safe you are depends on experience of others. What? Didn't I just say experience makes you fell safe or unsafe? I did. Here, I'm speaking of the experience of literally hundreds of unknown designers and inventors, tradesmen and ordinary folks who have built and operated railroads since the early seventeenth century. Again, this is true for other modes of transportation for times from the present (air and space) back to thousands of years BC (water craft and animal borne conveyances.) Safety wasn't a science, or even a practical discipline, when those inventions were put on rails and hardy experimenters chose to ride them. But none, I'm sure, wanted to be maimed or killed by their inventions. Each failure or accident resulted in knowledge, the derivation of tricks of the trade, and certain precautionary measures that became recommendations for future builders long before they became codified in science or engineering disciplines.
In other words, the first steam boiler that blew up prompted its builder to tell the next one to watch the water level and the pressure. The next one after that may have found a way to regulate the water level, or a guideline as to how much water to start with and when to refill the boiler. Now we're starting to move into the realm of science, although very tentatively.
How safe you are depends on design. Once inventors and the trades needed to make railroads work started to write down instructions and teach others how to make a train and run a railroad, a school of design came into existence. From the standpoint of an aesthetic, people who rode trains or shipped their goods on the railroad would acquire an emotional feeling of safety from knowing that "the railroad" looked like "a railroad" should look. Designers knew how the parts fit together and how track and trains should look, and the conservative designers would design new trains that looked pretty much like the ones that were tried and true. This was good for actual safety, as well.
Design meant that rails were designed to be a certain distance apart, ties had to be so many to the mile, and the material holding it all down had to be heavy enough to satisfy the heaviness that designers designed into the trains. Locomotives had to have big wheels. Cars carrying goods had to have sturdy walls, doors, and things to keep out the weather. Goods cars with valuable merchandise had to have locks. Cars for people had to have seating, preferably facing in the direction of travel, and something to keep the weather out as well.
These were all practical considerations, before engineering science took over. Everything had to be designed to look heavy enough to accomplish the task, which was moving people and freight over the rails without the whole damn contraption falling apart on the way. You were safe as a practical matter, and you are safe today, because current designs draw on the designers of the past. Somewhere back in history, if a design didn't look heavy enough, then it was deemed to be inadequate no matter if the engineers said otherwise. You were safer because of this, even though some designs looked safer than they actually were. (See: Tay Bridge Disaster of 1879)
How safe you are depends on engineering. Today, design is more of a sub-category of engineering. A radical new design no longer has visual features or cues that give confidence. New designs require extensive testing after rigorous engineering calculations are done, and materials sciences are applied, to decide whether metal A or plastic B or carbon fiber C will be an appropriate choice. Up until the late nineteenth century, however, much innovation was still by trial and error. Safety was a factor, but not the only factor in whether an innovative railroad design was successful. College level schools of engineering focusing only on railroads or transportation were few.
The railroads, themselves, however, already had seven or eight decades of trial and error under their collective belt, and the largest of them supported their own design, engineering and testing laboratories. Everything from the quality of fuels, through lubricating oils, to grades of steel, hardiness of weather protection, best materials for ballast, and up to and including the best cooking methods for dining car service could be empirically examined, tested, and compared with practices of competitors. And by the quarter point of the twentieth century, these competitors included automobiles, trucks, buses, and airlines.
I believe it is partly because of these up and coming modes of transportation that railroads supported university college-level schools of railroad engineering such as the one established at my alma mater, the University of Illinois at Champaign-Urbana, in 1906. Schools such as this gave railroad professionals the opportunity to cross matriculate with all the other engineering sciences. Civil engineering makes railroad track, grades, bridges and level crossings safer. Metallurgy makes track, locomotives, cars, interiors, exteriors and crush protection safer. Structural engineering provides for stations and public facilities that are both attractive and convenient, and for heavy lift facilities to get raw materials and goods on and off trains. Mechanical engineering ties the physics of motion and of turning fuel into energy together. Safety engineering ties all the best practices of all the engineering disciplines together to get the most productive railroads with the lowest risk factors. More about risk later.
How safe you are depends on how the component elements of the railroad are manufactured. At one time or another during the history any large American railroad, it had manufacturing facilities where everything from nuts and bolts to entire locomotives could be made. Today, practically all components, from track to train to station to transfer facility, are made by suppliers or contractors. This started primarily during the mid-twentieth century when motive power switched from steam to diesel-electric or straight electric. It also happened in part when custom work became too expensive and off-the shelf components, designed for a wider market, became more profitable for outside manufacturers to produce.
It is virtually inarguable that each and every component of a railroad, in one way or another, affects your safety. Pick any component at random, and I can probably tell you how a failure of that component, or the unavailability of that component, means you are less safe. Think of a part of a railroad far removed from the actual operation of a train on a track, and you will still find a certain amount of criticality that might result in injury to you, even if you don't ship by rail or ride any trains. All physical material things that go into a railroad and railroad operation must be manufactured either to the highest standard or to the lowest standard permissible by law and/or regulation.
Think about that last statement! I'll give you two examples.
Example One: Broken metals parts are common causes of train derailments or wrecks. Assume I'm talking about any metal part, big or small. Federal rules and regulations have raised the floor below which parts could be made with absolute disregard for quality, even though properly engineered. So at the very least, the bolt that's holding your passenger train car is made to the lowest quality permissible. It is never the highest quality possible, for that would require that each bolt cost a fortune to produce, and probably only one out of a batch of hundreds would be used. So we still occasionally, to a degree of acceptable risk, get failures, wrecks and derailments.
Example Two: Railroads used to use tons of paper, and more recently use tons of electronic computer equipment to keep track of everything from signaling to train schedules to car routing to ticketing. There is no lowest quality permissible for such computerization. Either the computers work or they don't. So the technology must be manufactured to the highest standard, and yet, computer glitches happen. A power supply fails, the internet goes down, a cable comes unplugged. So we still, occasionally, to a degree of risk that can't really even be calculated because of the micro-electronic nature of the component, get failures, wrecks and derailments.
In both examples, even the highest standards, if physically and economically attainable, will result in some degree of risk, and you are more unsafe as a result.
Uncontrollable factors in railroad safety.
In the last section above, I've given you a taste of the uncontrollable physical factors. I would divide these into mechanical, electrical, and electronic. Paying top dollar for mechanical components engineered by the best and designed through years of experience cannot reduce risk of failure to zero. The same goes for electrical components. Electronic components are even worse when it comes to diagnosing and correcting failures. Partly by virtue of advancing technology, and partly by government regulation, railroads are on the march towards greater reliance on computers and computing, and on the bastard cousin of computing, artificial intelligence.
Positive train control----the ability to know the whereabouts of all trains and control not only speed and separation but also hazard detection and autonomous reaction to imminent hazard----has been developed (in the United States by government mandate) over more than a decade, and yet it is still in its infancy. It hasn't been in operation long enough, by the standards of an industry that measures maturity in centuries, to know what it can and can't do in all situations. Some of the most spectacular wrecks in railroad history have occurred since its implementation. Imagine overlaying the bastard baby AI over it and increasing the uncertainty exponentially. Yet, that is where computerization of railroads is headed.
People run the railroad. People make mistakes. The truth of both statement is irrefutable. One can see where this leads to uncontrollable safety factors by simply looking at the best methods for creating an error-free safe environment for those people. We cannot hypnotize people to be error-free. We cannot program them or order them to be error-free. Eventually, it comes down to training. Safety training is a part of almost every work environment, but it is particularly front-and-center with railroads. But safety-training itself is not without error. Flawed humans designed the training and wrote the rules and teach the seminars and classes.
So, just as we said above that safety awareness and methods of accident prevention are the accumulation of years of experience, so too are unsafe or error-prone practices that have built up in safety training programs over years. The promulgation of these errors is one of the greatest, if not the greatest uncontrollable factor in railroad safety. It's insidious! You can't know what errors are deeply embedded in safety training until they manifest themselves in accidents. Then you can only go back and try to address them, hoping at every turn that more will not be added to the list of factors that are unseen until they happen.
This is why some in railroad management see full computerization and eventually artificial intelligence as the ultimate solution for keeping everyone safe. No human operators, no human making the rules, and no human doing the training. Logically, of course, it can't work exactly that way. Railroads are a service, and that means they must interact with the people or businesses who are buying the service. Investors would have it just that way, if they could train shippers and passengers to interact only with computer screens.
Practically speaking, the best we can expect from artificial intelligence in the safety department is the ability to foresee those hidden flaws in training and all the other aspects we have covered. To be able to evaluate cause and effect from millions of data points multiple steps into the future, and then to notify human safety managers of the outcome of those unseen combinations of flawed data may be AIs greatest strength. At least for railroad safety.
Risk. The management of risk is a discipline more related to statistics than to science or engineering. It can be scientific, if the data is used to develop theories that can be tested without causing harm. (Can we wreck a passenger train to see if anyone gets hurt?) It can be pseudo-scientific, if used to develop theories that are based on unsupported or emotional assumptions. (Did we assume that a train designed to standards used in 1910 can function safely because there were no wrecks in 1910? If it had no wrecks, the personnel must have been better.) It can be strictly mathematical. (If railroad A had fewer wrecks than B last year, it was safer.)
Management of risk means applying a statistical level of risk against expected rewards. The rewards can be either anticipatory (safer outcomes) or economic (higher profits.)
An example of the former case would be the expectation of a better safety record for a certain amount of time spent on safety training. If statistics reveal that a certain amount of training time results in a peak in safe operations, then we weigh the risk of fewer hours resulting in less safe operations but more time on the job for workers. An example of economic rewards against risk would be whether or not to risk higher operating speeds to get more customers to ship by rail. If statistics showed that higher speeds also resulted in less safe operation, is the expected economic benefit of more customers going to outweigh the cost of injury or property damage?
This can all be reduced to a risk versus reward calculation. It is not "the greater the risk the greater the reward," which is a strictly economic myth with little basis in science or mathematics. If it were true, then operating a railroad at unsafe speeds would result in enormous profits. It would be, and is, a callous and simplistic way of viewing all of the factors I have listed above in making you feel safe and be safe on or near a railroad.
And I am Mistertrains.
©2024 – Charles A. Turek – mistertrains@gmail.com
Graphics courtesy of: Microsoft Image Creator and Designer
(Charles A. Turek is a writer and novelist based in Albuquerque, NM. After four decades working in areas of the insurance industry related to transportation, he now writes on all aspects of American railroading. Posts contain opinion as well as factual information and are politically conservative in nature.)
Comments