Greetings and salutations! Welcome to another fun edition in the land of safety. As I mentioned in the previous edition, this month I’m going to talk about Root Cause. Root cause analysis and its process is always one of the most fun topics in the oilfield, said no one ever.
The first and most important step in any process to determine the root cause of an accident is a rigorous accident investigation. Some companies merely ask the supervisor to investigate the accident to find out what went wrong. Over the years I have found that this is the area where most companies have a tendency to fall short. The process is usually as follows: 1. Establish a timeline in the sequence of events leading up to the accident itself, 2. Interview all witnesses, 3. Take all pertinent pictures that capture the scene. This in and of itself is an area that requires proper training. The stark reality is that most companies do not spend an adequate amount of time on the accident investigation itself. By the time most accidents are investigated, the scenes are contaminated. There seems to be a premature assumption of what the cause is after brief interviews with those involved and those required to write their respective statements. Most responses are “Well, he had his head up somewhere…” Well, even though it’s anatomically impossible, it still wouldn’t be a legitimate answer. Human nature kicks in and hardly anyone wants to admit fault or culpability. Root cause results, based on inaccurate or incomplete information, are not a root cause.
Here are the basic principles: Command (policies); Apply (procedures); Monitor (Inspect); and Enforce (reward and discipline). In other words, C.A.M.E. These principles do not stray too far from parenting, but that’s another story in and of itself. Before we start to break this down, I’d like to preface by saying that safety in general has a propensity or proclivity to make things confusing and convoluted to the intended audience in the field.
I’ve realized that the people in the field are not interested in “how to build a watch.” They just want to know “what time it is.” The best advice that I can give for training are the words of Henry David Thoreau: “Simplify, Simplify, Simplify.”
Policies are the written rules meant to keep one in compliance with all of the regulatory entities. Procedures are how to perform the policies within the policy confines. Inspecting what you expect (monitoring) is the process to ensure quality compliance. The last, but perhaps most important phase, is either reward or discipline in a wat that is fair and consistent. With all of this established, we go to the root cause of accidents. If it’s a policy or procedure that needs attention, that is relatively easy to arrive at. More times than not, most accidents are a manifestation of lack of monitoring or enforcing individual performance.
There are three main types of individual failures. Failure occurs when an individual is either (1) unable, (2) untrained, or (3) unwilling to comply with company policies. Seldom is the failure a flaw in equipment or a procedure, if you have sound policies and procedures. So when you have an accident, you can usually trace the root cause to where an individual took exception to a policy or procedure. Which is usually a breakdown of monitoring or enforcement.
An example of being unable comes when a person or persons were not provided with the necessary tools to perform any given task. This is a situation when a person or group is forced to improvise—a step also known as taking shortcuts. An example of this comes when a crew member is trying to get a bite with a 36” but the jaws and heels are worn out. They try to make it work or they improvise by taking a shortcut. In the investigation, it is determined they previously asked for new jaws and heels for the pipe wrench that slipped off the pipe, but none were provided. They were unable to perform the task with improper tools for the task at hand. This is a blocked performance. The individual(s) will say they had asked for proper tools. The supervisor usually responds with, “That’s the first I’ve heard of this.”
The condition of being untrained affords another reason why individuals fail. We as an industry run most new-hires through a maze of training at the beginning of employment. This is usually like drinking water out of a fire hose. To the experienced hand, the training is a grueling and boring repetition of the previously canned training that they’ve had before. It’s not consistent with the “real world.” The rookies in the industry have no concept of how to apply what they have learned. Training has to be applied in the field as well as with mentoring and on-the-job training. Lastly, when we come to the unwilling, we can arrive at an easy fix. If the employees have been properly trained and properly shown and they still take exception to the policies and procedures, then, after documentation, they must be released. The oil and gas industry was founded on risk. However, the risk has to be studied, calculated, monitored, and reduced.
So when conducting an accident investigation, effort into the investigation is a direct correlation to the effectiveness of the mitigation of the risk. Any root cause analysis that is done with insufficient data, incorrect data, or false witness accounts is nothing more than a root cause of speculation.
When doing an accident investigation, here are some tips: 1. Separate all witnesses and get their individual statements. 2. Afterwards, ask at least 20 questions about the sequence of events, in order to try to establish a timeline. 3. Ask additional questions about any holes in their account of what happened. Ask what could have been done differently to prevent the accident. More times than not, the employees already know what could have been done differently.
So, in summation, the root cause of a poor Root Cause Analysis is an inadequate investigation. Spend the time necessary for a meaningful qualitative RCA. Regardless of whether you you use TapRoot, Five Whys, or the Bobby Jones RCA, Behavior Justification (my personal favorite), if you do an inadequate investigation you will not be led to a true root cause. Again, be safe and let’s get home to our family.
Dusty Roach is a safety professional based in Midland. He is also a public speaker on subjects of leadership and safety, and he maintains a personal website at dustyroach.com.