December 15, 2007

Focus on O&M (December 2007)

Third, grow a “why tree”

The root causes of all industrial incidents lie in the little things that present themselves on a day-to-day, moment-by-moment basis. To uncover them, we must sincerely ask, Why?

Experts recommend using a unique device called the “why tree.” It’s similar to the fault tree used in design industries, but it’s much easier to use. What you must remember, however, is that the why tree is not probabilistic in nature, and therefore does not predict potential failure rates. Instead, it uses evidence, along with ignorance and expertise, to dig logically into an actual problem.

When incidents happen in our physical world, we can always find a physical problem. But all physical problems are caused by people who do something inappropriate. In other words, people are triggering mechanisms. Unfortunately, the usual reaction is to find the person who triggered the problem and blame him or her. Our eagerness to “point the finger” and discipline the scoundrel implies that the cause lies with people—not what’s in their minds, not the way they make decisions, but the physical person.

Granted, sometimes discipline makes sense. But only when we define specifically what we want it to accomplish. Our intentions should be precise, fair, and appropriate. More important, we must continue down the why tree and understand why people did what they did. The following is a good test of when to stop asking why: “Do I understand the causes well enough that I’m convinced I would do the same thing under the same circumstances?” When you can answer “yes” to that question, you’re at the point where you can identify the root causes and make lasting, constructive improvements.

Make the findings visible

When a root-cause investigation is finally complete, the team lets out a collective sigh of relief. “Finally, we have cracked the shell. We know what caused the failure. Victory is ours.”

“BILGEWATER,” as Marmaduke Surfaceblow might say.

Who cares if the investigating team understands the causes of failure, if nobody else understands? Who cares if the principal investigator has been enlightened, but not the plant manager? What is the point of root-cause analysis if every single individual in the company does not learn the lessons? Read the sage advice of a failure specialist:

  Several years ago, I was investigating a large electrical system failure at an industrial facility. The entire complex had lost power for many hours. When I arrived on the scene, I felt almost overwhelmed by the complexity of the electrical system and its cascading series of failures—I was an electrical novice. Our team eventually completed its work and produced a final report. I knew what was in it because I was part of the team. But I didn’t write it. In fact, I couldn’t have written it because it was too complicated. The report was five inches thick—page after page of detailed, technical writing.

  Because I was contractually bound to contribute my own narrative, I decided to write a different report—one that people could read. I limited it to a few pages, and, because most people prefer looking at visual images, I included numerous drawings. I even hired an artist, which meant I had to translate the causes of a complex, technical catastrophe to a person trained only in fine arts.

  At times, I found myself talking to the artist as I would to a 10-year-old. But in the process, I found areas that even I didn’t fully understand. I realized that I was falling into the same trap as other so-called experts—when you stripped away the technical jargon, there was little substance to what I was saying. It was then that I realized the importance of translating a study’s findings. When my final report was ready, the artist fully understood the causes of the multimillion-dollar, technological failure. So did others with no technical training—including my secretary and my wife.

This specialist’s final report was only 30 pages—15 of text and 15 of supporting graphics. Most of the graphics were cartoons that exaggerated the phenomenon being explained. The ability to exaggerate added immensely to the report’s clarity—it was a way of “raising the volume” to get the reader’s attention.

—POWER editors