Root cause analysis (RCA) is the process of identifying the systemic cause of a failure incident. This is critical in maintenance for finding and resolving the true cause of mechanical problems, instead of just fixing the symptoms.
Failures identified by a root cause analysis can be classified into one of three categories:
There are a number of different methodologies available for conducting a root cause analysis to find the reason for a failure. We’ve included a few of them below:
Mechanical problems are often complex and have a number of underlying causes, even though they might not be obvious at the surface level.
The core principle of the 5 whys method is to repeatedly ask why a problem occurred multiple times, going deeper into the failure’s specifics each time until the true root cause is uncovered.
Here is an example of the 5 whys method for root cause analysis:
Once you have arrived at a root cause using the 5 whys method, you can identify a solution. In the example above, you can set the proper weight limits of the mixing load by consulting OEM guidelines. You can then note weight limits on all applicable assets and work documents.
Also known as the eight disciplines of problem-solving, 8D problem solving is a popular RCA methodology in the manufacturing, assembly, and services industries. It takes a team-oriented approach to problem-solving and relies on a heavily structured and disciplined process.
8D problem solving goes beyond simple cause analysis and identifies actions that need to be taken to contain the problem as well as corrective action to prevent future incidents.
The A3 problem solving method refers to the paper size usually used to generate these reports (originated by Toyota). The A3 process follows a series of eight steps:
These eight steps are very similar to the PDCA cycle and follow the same principles.
Selecting a root cause analysis methodology is usually done at the management level, well in advance of any actual failures. Once a failure occurs, you should follow the steps laid out by your methodology to analyze the problem and establish a remedial course of action.
There are some best practices that should be observed when performing an RCA — regardless of the methodology.
Most of the RCA methodologies mentioned above require information to be collected, analyzed, and documented in some way over the course of the investigation, and for a good reason. Information is a maintenance team’s best ally and the more information you have about an issue the better. When you can refer to previous failures and how a root cause analysis was conducted, you can begin to see patterns over time, quantify the impact of your actions, and improve reliability at your organization over time.
A computerized maintenance management system (CMMS) is a way to gather all the information and data needed to do a thorough root cause analysis in one place. It also allows you to document this information in a way that’s easy to sort, filter, and use specifically for an RCA. For example, some CMMS have a field on work orders specifically for failure codes and reports that help you compile a list of common failure causes. All this adds up to a quicker, more informed RCA process. Your maintenance department can identify trends that could help you prevent future incidents and follow up on any findings by scheduling corrective or preventive maintenance on the affected systems.
RCA should never be conducted by a single person, but rather by a qualified team of experts. The team should have a well-rounded variety of disciplines and have collective knowledge of the entire production process so that every perspective is considered. Having a diverse team of individuals conduct a root cause analysis also prevents bias from clouding your search for a root cause. If only one or two people contribute to the RCA, it’s likely that you will miss a possible cause and only investigate a narrow set of possibilities.
We can use a hypothetical situation as an example of how to do a root cause analysis.
Let’s say there is an automated dough rolling machine that is chronically offline because the rollers become blocked by the food going through them. On the surface, this blockage simply needs to be cleared and the rollers can function normally again.
However, it keeps happening, which interrupts production and increases work for the maintenance team. When the problem is investigated, it’s presumed that there is either a mechanical failure with the machine or a defect in the quality of the dough.
When the team drills deeper, it discovers that the machine is being washed down by the operations personnel after every shift. The leftover flour is getting wet, solidifying the rollers, and seizing the machine.
What the team initially thought was a mechanical failure is actually both a human and a systemic failure. New procedures can now be created that describe proper cleaning guidelines for maintenance and operations.
Had the team not done a thorough root cause analysis, the operations crew would’ve continued to wash down the machine and caused further seizures, which may have eventually resulted in a more serious breakage.
Leverage the cloud to work together, better in the new connected age of maintenance and asset management.