Drilling advances

	Vol. 227 No. 9
LES SKINNER, PE CONTRIBUTING EDITOR

Post-event analysis. Let’s face it – things do go wrong in the oilpatch. Murphy, or someone with the same last name, developed his first “Law” years ago at NASA in Houston. We all know it: If anything can go wrong, it will.

Murphy was right.

A lesser known law developed by Murphy says: Nature always sides with the hidden flaw. Unfortunately, the hidden flaw has an annoying way of developing into serious problems on occasion.

The drilling rig is a place where the unexpected can occur at any time. When this leads to an event that results in a dangerous situation, an injury or fatality, loss of property or significant loss of time, the first question is “How did it happen?” Sadly, the answer is often “I don’t know” or mere speculation is offered without in-depth forensic analysis of the event.

Post-event analyses offer a method for dissecting events to lay bare the incident mechanism, the critical event string leading to the event and, ultimately, the primary cause and contributing factors. Many supervisors, operators and drilling companies do not take the time to go through the process. This is especially true when rig rates are high and there are contracts waiting. As an old friend said, “It doesn’t matter who shot the calf, it’s dead!”

Post-event analysis of an unwanted incident is the basis for an entire body of forensic science. Engineers, architects, highway safety specialists, contractors and regulatory agencies participate in these analyses when there is a large insurance claim, the threat of litigation or a fatality.

Exactly what is a post-event analysis? Perhaps it is best to start by stating what it is not. It is not a witch hunt. Finding the culprit and taking revenge upon the guilty party is as antiquated as wooden plugs in car batteries. In the past, this was done with regularity and impunity. Drop an expensive tool on the floor and break it? You were fired on the spot. Generally, the fumble-fingered rig hand had to walk back to town or wait on the slowest crew boat in the fleet. No defense, no appeal, no discussion. You were simply dressed down in front of everyone on the rig and terminated with prejudice (at least, that’s what happened to me once).

The motive behind this was simple: frighten everyone with job loss (or worse, depending on the size of the “stud duck”), so the event would not be repeated. This created fear, animosity and resentment among crews, particularly if the unlucky crewmember was innocent.

Post-event analysis is not an attempt to sweep the event under the rug. A cursory review of the event, ignoring part of the information and jumping to a quick-and-dirty conclusion often leaves the actual cause undiscovered. So, the event repeats itself. A true post-event analysis involves several crucial steps.

First, everyone that was a witness or a potential witness should be interviewed individually by a single investigator (not a tribunal or committee) to determine what he or she saw, smelled, felt, heard or tasted. This fact-finding interview should not delve into the individual’s “idea” about what happened. Like Sergeant Joe Friday used to say: “Only the facts, ma’am, only the facts.”

Second, there should be detailed analysis of any tangible item involved. This is where forensic science becomes important. Is there damage on the piece that would lead to a conclusion about what happened? Is the piece worn, bent, mashed, poorly maintained or incorrectly designed? Were there adequate engineering controls or shutdowns in place?

Third, the crewmembers themselves must be analyzed. Did they fully understand the job at hand and their role in that job? Were they properly trained? Were they fatigued, hungry, sick, angry, upset, or uncooperative at the time the incident occurred? Was there a language barrier that prevented good communication?

Finally, what did Mother Nature contribute? Was there pressure on the well? Was the well producing hydrogen sulfide? Hot fluids? Mud? Oil? Water? Rocks?

Once all the facts are gathered, the specific event string must be determined. Experience shows that in almost every incident, there is a critical event string. Each event in the string must take place in precisely the right sequence and at precisely the right moment for the incident to occur. If the critical event sequence is broken, the incident never happens. Why did it happen this time?

Then, the investigator begins to layout scenarios and evaluate each one against the facts to determine what lay behind the incident. All possible scenarios that could have led to the event should be outlined and mapped, regardless of how unlikely it might be. Then, each scenario is weighed against the facts. Those that could not have occurred are eliminated. Those that could are rigorously checked. Finally, the most likely scenario floats to the surface along with the primary cause of the incident and all contributing causes.

Most investigators then request peer review from a disinterested third party. This may be another investigator an engineer or supervisor within or outside the company, another toolpusher or some other individual who has no axe to grind. This objective review provides the opportunity for the investigator to review the facts and conclusions, before the final step. If there is a reasoning flaw, it usually becomes evident.

The final step is documentation. To borrow a line from a Tom Clancy book, “If it isn’t written down, it never happened.” The report should contain all the facts, as well as they are known, a listing of all unknowns, results, interview forensic analysis on equipment, scenario discussion and a summation of the most likely incident scenario. Of special importance is the assessment of the probable and contributing causes along with recommendations for avoiding a similar future incident.

Sadly, many operators don’t want to know what happened. It might be embarrassing to them, or it might damage corporate reputation. My experience is that owning up to deficiencies and making sure they are corrected actually enhances corporate reputation, but I’m sure there are others who have been blistered by an unfavorable post-event analysis.

One thing is certain, unless we learn from our mistakes; we are destined to repeat them. This is certainly true if we do nothing to interrupt the critical event path that led to the incident in the first place.

Once again, Murphy was right. We all have the obligation to prevent these incidents in the future. To do less is simply unethical.