When things go awry, it’s usually because of something in the details. It would be nice if that weren’t so, but as you know, in safety little things play big. As to why that’s so, for openers, there are just so many of them; none are unimportant; attending to every single one is hardly the path to getting things done quickly.
Compounding the problem, there’s the communication process. Tell someone all the details when assigning them work – exactly where to tie the identification tags, exactly where to open the line, exactly what PPE to wear, exactly how to wear it – they’ll probably accuse you of micromanaging.
Though, to point out the obvious, something akin to that happens every time someone authors an energy control plan or issues a work permit. At least they seldom stand around, looking over the shoulders of those assigned to carry it out, making sure every single thing is done exactly as planned.
Perhaps they should.
As to how often things do not go exactly as planned, that would make for an interesting metric, wouldn’t it? There’s bound to be some kind of a performance gap; so how big is it?
Most of the time when something’s not done quite right, something else will pick up the slack: hence the principle of cross checks, independent layers of protection, margin of safety. The requirement of two pilots in the cockpit for a commercial airline flight is the perfect example. Confine measurement to the end-result – did anybody get hurt? – performance usually looks pretty darn good.
However, on those occasions where a series of things that aren’t done the way they’re supposed to, time to cross your fingers and hope for the best. The odds are normally in your favor.
But then, so are the odds of a major failure being nothing more than the unfortunate series of smaller ones. Exactly what happened recently: identification tags placed not exactly where they were supposed to; confusion as to where to break into a line; PPE worn not quite right; a hazard detection requirement waived as it was seen as redundant. When those details went awry, it led to the release of 27,000 pounds of hydrogen sulfide gas, causing two fatalities.
Case Studies
In our practice, we consider ourselves to be management consultants but most of the work we do is nothing more than good, old-fashioned teaching, in classrooms real and virtual. Case studies are an essential feature in the design of our courses; cases are drawn from real life experience, not all of which are good. Case studies can be a very effective way to learn.
In his best-selling book, Thinking Fast And Slow, the Nobel Prize winning professor of psychology, Daniel Kahneman repeatedly makes the point, “People are very critical of others, but not nearly so about themselves.” Is that ever right! Seems like every time we run a case study, out comes “They didn’t do this” and “They should have done that”.
As if to suggest, “We would never be so careless or foolish to do something like.”
I suggest that you read the preliminary investigation report of this incident, published by the Chemical Safety Board, even if your operation has no hazards even remotely like H2S, or piping systems that need to be broken into from time to time. This case has important implications as to the management and execution of the details: how and where things go awry. When you read the report, think “How could something like this happen here?”
Thankfully, their report does not include the distraction of “root cause.” Eventually someone will hang a convenient label on this tragedy, thinking that in some way it serves the purpose of the investigation. Better for you to simply focus attention on the lessons to be found in the details that played so large.
I’ll offer three.
Lesson One
Of the many factors keeping people safe, knowledge of the hazards and how to work safely are the most important. This is a case where it appears everyone involved understood the hazard perfectly well. Moreover, robust systems and processes were in place to do the work safely. They just weren’t executed all that well.
The first lesson is to appreciate how commonplace that is. When was the last time you read a significant event investigation report that concluded, “Nobody had a clue….” or “We had no process to safely handle…..”?
On the other hand, at the point where preventive and corrective actions are laid out, how often do you see “Retrain those involved” or “Revise the procedure”? All the time.
Stikes me as a disconnect. Not that teaching people how to do the things they don’t know how to do, or aren’t good at doing – aka, knowledge and skill – isn’t a wise investment. As is improving procedures difficult to execute.
Another example of where the problems are found in the details.
Lesson Two
As to the details in this case, long story short, two people assigned to break into an out-of-service line to remove a blind flange – temporarily installed as a safety precaution – instead broke into an adjacent line. They were exposed to a hazard everyone knew all about; so much so, the two were outfitted with respiratory protection and supplied air. Another precaution.
Details cited in the report are revealing:
- The work took place in a pipe rack, twenty-five feet above ground level.
- The location of the blind flange was shown to their supervisor, who, in turn pointed that out to the two.
- That work instruction took place on the ground.
- Since they were protected by supplied air, wearing a personal H2S monitor was considered redundant, and therefore not required.
- The identification tag that was supposed to be attached to the flange to be opened was tied to a nearby handrail.
- For the blind flanges in fourteen other piping sections that were also part of the task, tags were attached to the flange.
- A red temporary flange locking clamp was attached to a flange on an identical piping section five feet away. That line was in service, containing H2S.
In the CSB’s view, the two “…believed that…was the intended location of the work.”
Probably true. But to believe so required a lot of contrary information to be ignored. A case of relying on assumptions? Victims of a Cognitive Bias like Confirmation Bias? Never gave it a thought? Hazard recognized, but not to the point of stopping the job?
Those details are important parts of the process of recognizing the hazard.
Lesson Three
Viewed with the perfect vision of hindsight, there were huge warning signs in full view; the kind you would like to think would have caused one or both to climb down from the pipe rack, find the person who commissioned the work, and ask for some clarification. “Exactly which line are we supposed to open?”
Otherwisde known as stopping the job.
Stop work is one of those those vitally important management safety practices leaders talk about all the time; had it been done – executed – this situation would have been just one more non-event, all in a day’s work. Did they miss the sign something was awry? Or did they see it, and just do it, anyway? Perhaps believing, “Wearing all that PPE, nothing could possibly go wrong”?
Of course, stopping the job takes time. The event happened near the end of a turnaround, around 4 PM. You know what they say: time is money.
Their questions to answer. The ones you ought to be asking yourself are about your good followers: would they see warning signs like these? If they did, would they stop the job?
About Those Details
Most of the tough safety challenges leaders face – awareness, behavior, complacency, compliance, distractions, equipment, pressure from production, cost, schedule and peers, training – are nothing more than various facets of execution. It’s the fundamental challenge in managing safety.
Get down to the details, execution is a matter of human performance. It takes leadership to fix the human component.
Yes, human performance draws on other factors touching humans: equipment, materials, technology, design, process. Given the unpredictable nature of human behavior, it’s tempting to focus on non-human solutions to execution: automation, engineering controls, design, human factors. Those can be fixed with money. Spend all you want, but you can never completely remove humans from every process.
Some organizations are far better at executing the details. The secret of their success isn’t found in piling on ever more levels of procedures to catch the errors that keep finding their way through all the preceding layers. Their path to great execution comes by getting people to just play better. Their leaders make that happen.
You might be one of those leaders who relished the challenge of managing the details, and are good at it. Great if that’s you, but what if it’s not? Those details are not going take care of themselves.
You might want to start looking around for those who are the masters of execution, and pay attention to what they do, and how they do that.
It’s all about those details.
Paul Balmert
April 2025