To bust the asymptote, industry may have to fundamentally redefine safety

By Linda Hsieh, Editor & Publisher

In a 2008 head-on collision between a commuter train and a freight train in California that resulted in mass casualties, the ensuing NTSB investigation blamed the incident on the commuter train’s operator. This individual caused the collision, it was concluded, because he had been distracted by text messages and missed a red signal warning him from entering a section of single track where the freight train had been given the right of way.

That type of blame on the worker is exactly what renowned organizational safety expert Todd Conklin preached against in his talk at the recent IADC HSE&T Conference in Houston. It was an eye-opening speech that shed a lot of light on the next steps that the upstream oil and gas industry must take in order to “bust that asymptote,” he said, quoting a former boss at Los Alamos National Laboratory. 

The industry has put in significant efforts over the past few decades to improve safety, resulting in a dramatic decline in injuries and fatalities. Yet, our safety curve has plateaued. “This is your story of safety. You’re in a classic asymptotic relationship,” Dr Conklin told the conference attendees. “And doing more of the same is not getting you better results.”

To move to the next stage of its safety journey, companies will need to think about safety differently. First, they must accept that every accident is not preventable. “This is something the automotive industry figured out, and you guys haven’t made the jump yet,” he said. Accidents are unintentional deviations from an expected outcome, which means they’re hard to predict and, therefore, hard to prevent.

“I don’t know how we drifted over time into this perfection model,” he said, where companies believe “there’s a perfect work environment and, if the worker is obedient enough, problems will go away. That’s attractive, but it’s just wrong.” That type of thinking also leads companies to “focus on investigating how we failed to prevent the accident,” rather than investigating the accident itself. This doesn’t mean prevention isn’t important, but it’s not enough. To make the next step change in safety, he said, companies need to start defining safety not as the absence of accidents but as the presence of controls, or safeguards. 

The primary safety function of seat belts in a car, for example, is to position the humans in “survivable space,” where all the different safety systems of the car can function more effectively in case of a crash – and the automotive industry designs their cars assuming a 100% chance they will someday get into a crash. Similarly, the drilling industry must adopt that kind of mindset so that “when the system inevitably fails,” there are multiple layers of safeguards built in so it can fail safely.

“The most profound message I can give you is never have a worker one safeguard away from a failure. You have to have multiple controls in the system.”

Dr Conklin also encouraged companies to adopt a “deliberate strategy” to improve. When an incident happens, “you can choose to either blame and punish, or learn and improve, but you don’t get to do both,”  he said. “If you choose to blame, you’re going to shut down learning. Absolutely, I promise you. And if you choose to learn, then you really can’t punish.”

When investigating incidents or any kind of operational upset, start by asking about the “what” instead of the “who,” he urged. Stop trying to seek behavior that can be labeled as somehow deficient and the cause of the problem.

“Workers are really smart, and we have to stop seeing them as the problem and start seeing them as the solution to draw from,” he said. “Don’t go out and look for places where workers deviate, because you will always find deviation in your industry. Go out and look for places where controls are effective, and repeat that every chance you get.”  DC 

Click here for a link to watch DC’s interview with Dr Conklin.