The History of Geology Forum: "Lessons Learned from Failures," will be held from 1:30-3:30 p.m. Sunday, April 18, right before the opening session of te AAPG Annual Meeting in Dallas.

The forum will be presented in Room C146 of the Dallas Convention Center.

Hold the Railings or Get Off the Rig

Disasters Happen; Why Invite Them?

"No force on earth can get everything to stay in balance all the time. To insist on perfection is to shut the whole thing off."

-- James R. Chiles, from "Inviting Disaster"

By DAVID BROWN
EXPLORER Correspondent

Things go wrong.

And they always will, so mistakes can be crucial lessons.

This year, the History of Petroleum Geology Forum at the AAPG Annual Meeting in Dallas will explore "Lessons Learned from Failure."

The forum packs a one-two wallop, starting with a panel discussion of "Bypassed Pays: Opportunities for Large Reserve Additions."

That presentation includes Sidney Powers medalist and AAPG honorary member Robert M. Sneider, with Larry D. Meckel of L.D. Meckel and Co. in Denver, and David G. Smith from Burlington Resources in Calgary.

Then James R. Chiles will describe "Human Factors and System Fractures: Lessons from Oil & Gas Industrial Disasters."

Chiles is author of the book "Inviting Disaster: Lessons from the Edge of Technology," published in 2001.

Robert Ginsburg and Marlan Downey will preside at the forum. Its failure topic followed naturally from a previous session, said Ginsburg, professor of marine geology at the University of Miami.

"In that one, we had done lessons from successes," he said. "We were casting around for an idea and we thought, 'What about the mistakes?'"

Recipes for Disaster

Ginsburg sees company culture as an important part of the failure story.

"People might say, 'That kind of sand is always tight,' or 'Never look in chalk,'" he noted.

He said Chiles can bring an abundance of detail and a unique perspective to that aspect of lessons-from-mistakes.

"In 'Inviting Disaster,' he included a whole series of examples that were attributed to judgment or culture, which I thought was the most interesting part of the book," Ginsburg said.

Chiles investigated dozens and dozens of disasters, large and small, in his research.

All mistakes offer lessons, but with true disasters "the learning is deepest," he said. "That's one requirement -- visibility, memorability."

He still maintains a Web site of updates on disasters and subsequent investigations, encouraging additions and comments at www.invitingdisaster.com.

Chiles' book opens on the morning of Feb. 14, 1982, on the offshore drilling rig Ocean Ranger.

There's no surprise ending to the story. The $100 million Ocean Ranger sank in a heavy storm in the North Atlantic on that Valentine's Day, killing its crew.

"I've heard it described as equivalent in Canada to our loss of the (space shuttle) Challenger," Chiles noted.

Later in the book, he also examines the fate of the Piper Alpha rig in the North Sea, the world's worst offshore disaster.

"Part of my reason for opening the book with (the Ocean Ranger) is that Americans didn't get the full message," he explained -- mainly because the United States never had an offshore rig collapse of similar scale.

"Fortunately for the workers in the offshore drilling industry, a lot of the Canadian and UK directives did filter through the whole industry worldwide," he added.

Chiles explores both the human and mechanical problems that resulted in the failure of the Ocean Ranger.

"One specific concern that came up again and again was the problem of split authority on the rig and onshore.

"At a minimum, you had three bosses," he said.

Also, "at least twice, the Ocean Ranger had imbalance problems," the latest only eight days before the rig's collapse, he observed.

Disasters would be even worse if their lessons go unheeded, a point Chiles makes in his book.

"You have to be willing to learn, and you have to have a willingness to change. I think the deepsea drilling industry has been willing to change," he said.

Elmworth: What Went Wrong?

Sneider will draw on specific examples to illustrate possibilities from bypassed pays.

The best known example might be the Elmworth Field in Canada's Alberta Basin, where 61 dry or uneconomic wells were drilled into Lower Cretaceous clastics.

As it turned out, those attempts missed 10 principal reservoir units that now have more than 2,000 producing wells.

What went wrong?

Not a lack of information.

"When you look at the data, all the data's there to make the right decision," Sneider said.

Part of the problem stems from the nature of department jurisdictions, he noted.

"People usually don't put all the pieces together," Sneider said.

"They work in a large enough organization -- and it doesn't have to be very large -- that they don't tie together all the pieces, which include both engineering and petrophysics," he added.

Another mistake comes from failure to examine the entire exploration picture, beyond geology.

"We're dealing with rocks and fluids," he said. "We do a good job with the rocks, but we don't take a very good look at the fluids."

Part of the solution comes from involving the right technical expertise.

And involvement of managers?

Not a help, according to Sneider.

"The more managers on an evaluation team, the more data doesn't get integrated into the wildcat wells," he said.

In discussing overlooked pays, Sneider will talk less about the pays, and more about the "overlooking."

"It's not a complicated problem," he said. "It's an organizational problem."

Costly, But Crucial

When many people think of disasters from internal organization problems, they think of NASA.

Chiles looked into several of the space agency's failures.

"NASA really was safety conscious leading up to the Columbia disaster," he said, "but it had put itself in a narrow view of what to worry about."

Because NASA engineers saw themselves as dealing with constrained resources, they concentrated on the "big things" that might go wrong, according to Chiles.

"The downside of that was that things that didn't fall into the red zone were managed by downgrading the potential risk," he said.

In evaluating disasters, Chiles could usually find precursors that indicated the later failure.

The usefulness of those precursors, however, is open to debate, he admitted.

"People say, 'It was just a precursor afterward.' The criticism is, it was just a precursor in retrospect. If it was one of a thousand things that happened, it might just be noise," Chiles said.

Now he favors building up benchmarks to compare incidents.

Armed with a concrete set of readings, tolerances and specific requirements, a safety auditor can point out any deviations, he said.

Chiles can identify a number of barriers to safety and reliability, including the attitude that "testing is such a bother."

Testing and training should reflect the real stresses of on-the-job performance, he said.

Sometimes that doesn't happen because it's harder to justify a failure that occurs during testing, or to defend an accident that occurs during training.

"I try to make the point in the book that life-and-death training can have a terrible price to it, but it's necessary to do it," he said.

Be Very Careful

Industrial insurers don't know how to give credit for training programs and prefer to reward companies for installing protective hardware, Chiles noted.

But training is so important that insurers need to find a way to reward a company for properly training its personnel, he said.

Avoiding failure always involves, and often begins with, the human factor.

"A lot of it is empowered and alert workers -- 'alert' in the sense that they know when something needs to be done," Chiles said.

"An aware crew is a lot safer than one that is blissfully ignorant," he added. "They know when they are pushing the envelope."

He'd like to see workers trained to stop disasters in their tracks, to become what he calls "crack-stoppers."

"It is really the notion of 'system fracture.' I've been told that this concept is a helpful addition to what's already a rich literature," he said.

Systems fail "in a step-by-step way analogous to how metal cracks under stress," according to Chiles.

"If there's a crack-stopping barrier, it will stop short of that culminating event," he said.

In almost any disaster, Chiles observed, "the question remains until the end: Will somebody stop it in time?"

In some cases, preventative measures are ignored even when they're readily available -- people don't take time to grab a hard hat, or they don't want to look stupid wearing a pair of safety goggles when they mow the lawn.

Chiles recalled touring an offshore platform, where he was instructed in safety measures as soon as he arrived.

During the tour, he failed to hold a handrail on a steep flight of steps. He was told to hold the railings, or leave the rig.

Chiles said he thought about the warning.

He realized the only thing that would look more foolish than clutching a handrail would be falling downstairs and breaking his neck.

"That would be a stupid mistake to make," Chiles said.

"Ever since then on a flight of stairs, I always hold the handrail."