Process Safety and Railroads
One of the themes of these Safety Moments is that different industries have much in common. Therefore, a safety program that works for one industry is likely to be useful and relevant in another. Certainly, each industry has its own specific issues that are not shared with others (some of these are described in Safety Moment #8: “But We’re Different, You Know”). But, as a generalization, different industries have much in common.
The principles that lie behind process safety management (PSM) are used in related business areas, such as pipelines and power generation. What may be surprising, however, is the degree to which these principles apply to industries that have no chemical processing or energy management component. One example is the railroad industry.
One of the central safety problems that railroaders face is the fact that trains and automobiles potentially meet one another at one of a multitude of railroad crossings. This problem is not one that the process or energy industries face to any significant degree. Most process plants are contained within a fence, and members of the public can only reach offshore facilities by boat. Therefore, the potential for an accident involving an automobile driven by a civilian is minimal.
Yet, the principles of risk management and process safety management can be used to analyze the chronic problem of train/automobile collisions. The accident shown in the YouTube video is representative of the problems to do with railroad crossings. The sequence of events was as follows:
- A car was turning left at a busy railroad crossing.
- It got stuck between the tracks and could not escape.
- The people in the car got away from the scene.
- The police arrived. Nevertheless, there was not sufficient time for the oncoming train to stop.
- The train hit the car.
The natural response to such events is to talk about “driver error”. Those conducting the incident investigation will look at possibilities such as:
- Simple carelessness;
- The driver may have been inebriated;
- It was not clear to the driver that there were two tracks.
- The driver was in a rush to make an appointment;
- The driver was using a mobile phone while the car was moving;
- and so on.
Obviously, the driver was in error. So, there is a call for more visible yellow lines on the road, more warning signs and a general push toward making drivers become “more careful”. But, by jumping immediately to the conclusion that the issue is one of “driver error” the number of potential recommendations is constrained. Thus, we may be missing the opportunity to fundamentally improve the safety of the crossings. A formal risk analysis may provide more useful insights.
Formal Risk Analysis
Those who work within the discipline of process safety understand that risk consists of four elements:
- A hazard — a situation that has the potential to cause harm. The hazard in this case consists of automobiles and trains being in the same place, potentially at the same time. Removal of the hazard is the only way of reducing risk to zero.
- The consequences of that hazard. In this case the consequences are injury or death of the car driver, complete loss of the vehicle and damage to the train’s locomotive.
- The anticipated frequency with which the consequences occur.
- The safeguards that help either prevent the event from occurring, or that reduce the consequences and/or frequency if it does occur. In this situation safeguards include flashing red lights, warning bells, yellow lines on the road and driver training. (These safeguards are all directed at reducing the frequency of the event; there is not much that can be done to reduce the consequences of a train hitting a car.)
Given this background, a formal risk analysis could work as follows.
The only way to totally remove the hazard is to design a system in which trains and automobiles do not occupy the same space at the same time. Unless this is done, some collisions are inevitable — risk can never be zero. Trevor Kletz is famous for saying, “If a tank’s not there it can’t leak”. A paraphrase of this statement would be, “If a train’s not there it can’t hit a car”.
Therefore, the best solution is to separate the road from the tracks, say by building a bridge or a tunnel. Such an action will reduce the frequency of a train/car collision to zero. (In this context the Virginia legislature is to be complimented — State law states that new crossings cannot be at grade.)
(Related to “remove the hazard” is “remove the people” from the scene. This is what happened here. Recognizing that they were stuck, the people in the car moved away from their vehicle — this was just the right thing to do.)
The second item to look at is the consequence term. On process facilities reducing the consequences of an event can be done through actions such as installing a system that diverts released gas to a flare or blowdown system. Unfortunately, it is difficult to think of a means of reducing the consequences to do with a train/car collision. Trains are so much bigger and heavier than cars that the consequence of a collision is always going to be one-sided.
The Frequency of Events
The frequency of events can be reduced through driver training programs — some of which have already been discussed.
Finally, more safeguards can be provided. These include more visible barriers, more flashing lights, louder horns at the intersection, installing floodlights for night-time operations, and installing full gates at the crossing, not just a one-arm barrier.
The risk analysis techniques that are now so integral to process safety management have a broad application to a wide range of industries, such as the chronic problem of collisions at railroad crossings. And always, the best response to a risky situation is to remove the hazard. That is the only way in which risk can be reduced to zero.
Public awareness to do with these events has increased enormously due to the fact that a video camera has been installed at the critical junction by a railway enthusiast group: Virtual Railfan. Similarly we are seeing more and more actual events at process facilities because security cameras are becoming so pervasive.
Copyright © Ian Sutton. 2018. All Rights Reserved.