You have identified a problem, and you have made an immediate correction to fix it for now. How do you make sure the same problem doesn’t happen again? The goal of implementing corrective action is to identify and to address the root cause of problems so they don’t reoccur. Therefore, root cause analysis is at the heart of an effective corrective action program.
Root cause seems to be an appropriate term for several reasons. First of all, it focuses on the source of the problem. Next, it implies that the source or true cause of the problem is hidden and out of view, which is why a root cause analysis is needed to improve processes.
Simply addressing the symptoms without addressing root cause sometimes creates more problems in the long run. A classic example is expedited orders. A customer calls because an order is late. So, to expedite it, someone takes time and effort to move the order more quickly through the process. Of course, in giving this order priority and moving it ahead of other orders in the pipeline, more orders are going to be late. Now more customers will be calling, more orders will be expedited, etc. The process of expediting this order, in effect, causes a chain reaction and negative spiral.
Does that mean that when a customer calls because an order is late that we shouldn’t expedite it? Of course not. The point is that expediting the order shouldn’t be all we do. We should investigate why the order was late and collect some basic information. How often do orders ship late? How often do customers call and complain about orders being late? What is the apparent reason this order was late?
At this point, we are just trying to collect some initial data to determine if this warrants corrective action. Not every problem or nonconformance requires corrective action. How to determine if a corrective action is needed varies by organization. It may be left up to individual employees or process owners, some might have a CAPA (Corrective Action Preventive Action) committee, and others may leave it to management. The need for corrective action should be evaluated along several dimensions. How often does it happen? How important or critical is the issue? How are customers impacted? What is the cost of taking action versus not taking action? The answers determine if there is need for corrective action.
In our example, if this was a rare event due to a special case with minimal customer impact, then perhaps we perform the correction (expedite the order) and do nothing more. If orders are regularly late, or if the customer was severely impacted, then it is a candidate for corrective action, which starts with root cause analysis.
Once you decide that corrective action is appropriate, there are a multitude of root cause analysis tools to employ. The tool you select should depend on the skill and training of those employing root cause analysis, and on the complexity of the issue you are trying to solve.
While the complexity of corrective action tools range widely from fairly straightforward (i.e. Five Whys) to fairly complex (Current Reality Tree), most are found to be fairly effective at reaching potential root causes if properly carried out. Perhaps the most popular tool, and in the middle in terms of complexity, is the Fishbone Diagram (also called the Cause and Effect Diagram or Taguchi Diagram) as shown in Figure 1. More experienced and skilled teams seem to be more effective at using complex tools.
However, the largest factor determining the effectiveness of root cause seems to be, according to the research, how effective the corrective action team works together. Critical areas seem to be a knowledgeable facilitator, how comfortable the team is challenging assumptions and critically evaluating the direction of the analysis. Another important factor is including corrective team members from outside the area of focus. Someone not locked into a certain way of thinking can bring a fresh perspective to the root cause analysis, which in turns leads to thinking outside of established patterns.
Besides the experience of the team, the type of problem or nonconformance should also play a role in selecting a tool for corrective action. For example, using a simple tool like Five Whys might seem appropriate for determining why a customer’s order was shipped late, but where issues of health, safety or other high risks are involved, a more comprehensive tool like Current Reality Tree, Cause and Effect Diagram, or Interrelationship Diagram would be more appropriate
There are many root cause analysis tools, and they all seem to have their supporters and detractors. You should choose which ever tool fits your organizational needs the best, or even mix and blend methods if you feel it is effective for your purposes. Then training should communicate established root cause analysis practices throughout the organization. Which ever method you choose to use in your organization, there are some common areas of importance:
While the many tools available for conducting root cause analysis may all be effective, their effectiveness is very dependent on how they are implemented: data collection, involvement and training of team members, a communicative and qualified facilitator, how questions are asked and answered, and follow-up verification. They are all critical to the success of the root cause analysis process, upon which corrective action relies.
Another important factor to consider is when to stop searching for root causes, and instead select action items to address identified root causes. If you keep searching, you can always find more and more potential root causes. Use experience, knowledge, and intuition to know when you have dug deep enough looking for causes, and now it time to take action.
Problems manifest themselves in many ways but to truly solve a problem you must make sure you have found the root cause. A good place to start is by understanding the top ten root causes of business problems you will encounter.
People don’t make mistakes. Systems make mistakes. If you have a system for training, well-written procedures, following-up on procedure usage (i.e. internal auditing, metrics, rewards), developing competent employees for the role they are placed in, updating and innovating methods, attention to detail, disciplined maintenance, quality designs, constant rewards and incentives for good work, and supplier validations, then you would have eliminated 80% of the causes of business problems. The last 20% is left to the individual’s ability to operate the system you have just created. What do we call such a system? A Quality Management System!
For a more in depth example, let’s look at writing articles. In the absence of any policy or process for editing and revising articles, the author might not catch his/her spelling errors or typos (humans err – that’s why newspapers used to have proofreaders). So when customers read the article — multiply the writer’s two eyes by thousands, maybe millions — the likelihood is great that at least one of them will catch the error.
On the other hand, if there is a process in place where a second set of eyes reads the article and necessary corrections are made (or the central idea is validated) before releasing it, the likelihood of mistakes getting out to the reader are vastly minimized. This doesn’t mean that mistakes won’t happen. But here too, the system might have a role to play. Is the individual in the right place? If writing is a requirement for the job, was the individual properly screened or trained?
Mistakes are made due to the lack of proper training. However, it is not the sole reason why things go wrong. Of course, there are situations where the person is clearly not qualified to do the job — here, too, the system comes into play (i.e., is the selection process foolproof?). Does the training cover how to use procedures?
Unused procedures are not effective. Why aren’t they following procedures? Procedures requires the discipline to use them and the management follow through to enforce them. The most important question supervisors can ask — “What does the procedure say?”
If a procedure is unclear, then it will be a lot harder to follow. Even well-written procedures are not perfect. When procedures are used, they are updated to reflect current lessons learned. Poorly written procedures are an indication that your procedures are not used.
This can result in business problems because your employee may not be the right person for the job. Better screening, job descriptions, or testing can help you to place the right person in the right job. Yet even with the right person you could be using poor methods.
This refers to methods that have been outdated and not changed, or at least the changes were poorly communicated. Are your methods captured in your procedures?
This is really about attention to detail, understanding your product, and caring for the output that you are passing on to the next step in the process. Pay attention and take the time to inspect your product. Poor inspection really means you are leaving up to your customers to inspect your product. Do you really want your customers to find your problems?
If you neglect your equipment then it is more likely to malfunction. Lean thinking focuses on preventive maintenance, which means regularly maintaining your equipment to ensure it does not break down in the middle of something important you are doing, like getting an order out on time.
We’re talking about bad design in the first place. Focus on designing in quality by doing it right the first time and you will avoid one of the causes of business problems. If you don’t have time to do it right, then how do you have time to rework it later?
Are you selecting these because the price is right? You get what you pay for… You shouldn’t be using poor quality raw materials.
Does your management appropriate incentives in place engaging employees to do it right the first time? I am not talking about just money. Recognition of good quality or pointing out poor quality performance may be all that is needed to send the message that quality is important and thus preventing many of these root causes in the first place.
Quality standards and quality tools are there precisely to increase the likelihood of positive outcomes. You may be surprised at how a simple process change can result in a big drop in error rates. But like anything else, the way to improve and be successful at corrective action and root cause analysis is to practice it.
While literature and training are great ways to get started, it is through executing the root cause analysis process that you will learn and improve. And successful corrective action and root cause analysis are critical to achieving continual improvement in your organization. Corrective action is all about finding the root cause.