Root cause analysis puts process on track

After a catastrophic event, such as a shackle line going down, what are you going to do to ensure it never reoccurs?

Root cause analysis must be based on data and observed facts, not hunches.
Root cause analysis must be based on data and observed facts, not hunches.

What happens when the wheels fall off and events in your process cause major disruptions? Do you have a systematic response? Once everything is put back together and the processing plant is back up and running, do you take the time to carry out a postmortem and discover the root cause of the event?

Conducting a root cause analysis (RCA) may be your best move to guard against a reoccurrence. In RCA, supervisors are expected to begin explaining why they did not reach their goals, and this is the foundation on which effective change happens.

The term, “root cause analysis”, however, makes the process sound as pleasurable as going to the dentist. And often, what passes for a RCA is asking in a weekly staff meeting, “How did this happen?” followed by the answers: “I think it’s because…” or “Generally, when that happens, it is because….”

Seldom is the answer, “The data determined that…”, or “Our study revealealed”. And, more often than not, those hunches are regarded as facts, and the consensus is that there is no good way to address the problem or somebody would have done so already.

Start with the culture

An effectively conducted RCA entails asking hard questions, and people may be embarrassed by the results. That is why it is crucial to have a company culture where the RCA is not used to assign blame. Instead, it should be used as a tool to learn from mistakes by bringing facts out into the open. Doing so allows effective changes to occur.

No one wants to talk about what they may have done wrong. The natural inclination is to off-load blame elsewhere. “It’s the maintenance department’s fault; sanitation’s fault….”, and so on.

Only when the threat of retribution is taken away will the facts begin to emerge. Think of having this kind of culture as the anesthetic for the process. When you are numb to the blame game, real change can occur.

Ways of defining the problem

The first step is documenting exactly what occurred. Some would call this “defining the problem”. But, getting a clear picture of this can be difficult because this process often doesn’t occur until after the fact.
A good starting point is to begin by asking, “Why” five times. Like the drilling process, you begin to see what caused the problem.

It may seem like you are acting like a four-year-old, but it has been proven that until you ask “Why” at least five times, you will not begin to find the actual cause of problems. Some issues require asking it even more times, depending on the complexity of the issue.

Relentlessly asking why, gets to the true root cause, but it can be embarrassing for the person responsible for answering the questions. It cannot be stressed enough that unless people are unafraid to answer honestly, the truth will be difficult, if not impossible, to discover.

Another useful tool is the “fish-bone” or “cause-effect” diagram. This is particularly useful for problems involving many departments. Draw it out with a center line and five lines extending from that. Label each of the five lines with Manpower, Machinery, Methods, Measurement and Material. Define the problem at the end of the center line (the effect) and begin brainstorming. This works by bringing everyone together in a group or interviewing each individually, by going through each of the 5 Ms and asking, “What about manpower can cause this problem?” “What about machinery can cause this problem?” Remember that, at this stage, it is all about brainstorming.

Getting at the data

Once all the potential causes have been identified, the problem solving process truly begins. Each of the potential causes needs to be studied to see if, in fact, each does have an effect on the problem.

Surprisingly, the best results from studies often come when the studies are conducted by someone with little or no experience in the process. Consider recruiting somebody from your human resources or accounting departments to conduct the study.

Having an inexperienced person count the number of broken wings going into the chiller, for example, will produce clear, accurate data. Because they have no pre-conceived notions on how the process is supposed to work, they will ask questions to understand what they are seeing. This is where it is vital to stick to the facts. Data is the most important factor in determining whether a stated cause has the said effect.

Close the loop with analysis and action

Having said all that, conducting studies is where the RCA process usually ends. Many theories and potential reasons will be elimated, but managers seldom close the loop by conducting analyses to determine the actual causes for the problems. It is tough to shine the spotlight into your own closet sometimes. No one wants to be embarrassed, so people usually skip this step and move on to the following step in the process without the proper data to support their conclusions.

Once the root causes of the problems are lined out, the necessary changes in the process can be implemented to ensure the problem will not repeat itself.

It is important at this step to have clearly defined actionable items. Each action item or change in the process must be assigned to a person who will shepherd the change through to completion. This means they are ultimately responsible for the success of the change by a set date. Again, the tendency is to not set completion dates and to not see changes through to completion.

Don’t leave out monitoring and control

The final step is to set up a process to monitor and control the changes that have been made. For instance, if the issue is broken wings and it is determined that 50 percent of the incidence occurs before the birds arrive at the plant and another 40 percent occurs in the hanging process, there needs to be a monitoring process to measure the incidence at each step in the process. This ensures the changes made in the catching, hauling, cage dumping or hanging remain effective.

So, in short, a properly conducted root cause analysis depends upon several elements. It requires a company culture allowing for honesty throughout the process. It must be based on data and observed facts, not hunches. Finally, any changes in the process as a result of the RCA, need to be monitored for their effectiveness. Once this process becomes routine it will feel less like a visit to the dentist.

Page 1 of 34
Next Page