9.6 Tools: Failure Mode and Effects Analysis—FMEA

As we discussed earlier, after your team has generated a list of potential causes during the divergent phase of the problem-solving process, you need to narrow your focus by screening likely Xs. You want to identify critical factors that have the most significant influence on the relevant CTQ. You can use Pareto Analysis and the Five Whys for that purpose. Another tool that you can use when you have many Xs is the Failure Modes and Effects Analysis (FMEA).

FMEA is a technique that you can use to study all possible causes in a process and their potential impact. You do not use this tool to identify the Xs that cause the Ys and eliminate the Xs. You use FMEA for identifying likely Xs and prioritizing them to determine the order in which your team should examine them. It is a powerful instrument for challenging and getting your team members thinking about the process in innovative ways. It helps you drive change through priorities.

To conduct an FMEA analysis, you can use the sample FMEA table in Table 9.2 and follow the general steps described below:

1. Identify the critical parameters (Xs) and their potential failure modes.

2. Identify the potential effects of each failure (consequences).

3. Rate the severity of each potential effect of each failure (column 4).

4. Identify causes of the effects.

5. Rate each cause’s likelihood of occurrence (column 6).

6. Identify the current controls for detecting each failure mode.

7. Rate the organization’s ability to detect each failure mode (column 8).

8. Calculate the risk priority number (RPN) for each failure mode (column 9).

Where:

Potential failure mode: This item is how your process could fail to meet customer requirements.

Potential failure effect: The possible effects of the failure mode on the process or customers.

SEV (Severity): The seriousness of the failure mode (1–10).

Potential causes of failure: How the failure could have occurred.

OCC (Occurrence): The probability that the failure mode will occur (1–10).

DET (Detection): The ability to detect the failure mode.

RPN (Risk Priority Number): A rating you obtain by calculating SEV × OCC × DET.

Let’s take a look at a loan application process. Upon completing and leveraging your training for a higher paying position, you may want to apply for a housing loan. Imagine that such a process from the financial entity involves you (the customer) filling out and submitting the application.

The institution then enters the data from the application form into a database and issues you a check. A cross-functional improvement team pinpoints “application filled incorrectly” and “data entered incorrectly” as two failure modes. As shown in the table, the potential effect or severity of these failure modes range from 4 to 8. In a real application, you would have more lines.

Note that there are two potential causes for the frequency of the possible causes, ranging from 4 to 6. The ability to detect the potential causes also ranges from 2 to 10. The failure mode “data entered incorrectly” with a probable cause of “data entry error within a single field” has the highest risk priority number (RPN). You would likely want to examine it further since the team discovered that no controls are in place (column 7). Besides, the team assigned a detectability (DET) score of 10. The failure mode for “application filled out incorrectly” has a lower RPN of 96, but you may also want to investigate further since the team rated the severity (SEV) high, at 8.