Capability analysis is the comparison of the output of your in-control process with the customer’s expectations, specifications or requirements. Let’s learn a little more about how to do capability analysis.
Overview: What is capability analysis?
Capability analysis is the quantifiable comparison of the Voice of the Customer (specs, requirements, or expectations of your customer) and the Voice of the Process (control limits). The goal of capability analysis is to come up with a quantifiable measure of how well you are meeting your customer expectations. The comparison is made by using the ratio of the difference between the process specifications to the variation of the process values, as measured by 6 process standard deviations.
There are a number of different metrics for doing a process capability analysis. One of the most common is Cpk or Capability Index. Other calculated metrics include; Cp, Pp and Ppk. The graphic below illustrates the concept of capability analysis and the formulas for the two most common capability metrics.
Cp and Cpk are considered the short term capability of your process. Pp and Ppk can also be used for doing your capability analysis. They are considered to be measures of long term capability. If you look at the formulas for Cpk and Ppk for an assumed normal distribution, you can see they are nearly identical:
The only difference is in the calculation of the standard deviation in the denominator. Cpk is calculated using the within standard deviation based on your control chart, while Ppk uses the overall standard deviation. Here are a few differences between Cpk and Ppk.
- Only accounts for the variation within the subgroups
- Does not account for the shift and drift between subgroups
- Is sometimes referred to as the potential capability because it represents the potential your process has at producing parts within spec, presuming there is no variation between subgroups (i.e. over time)
- Accounts for the overall variation of all measurements taken
- Theoretically includes both the variation within subgroups and also the shift and drift between them
- Customer is interested in the long term capability
3 benefits of capability analysis
Capability analysis allows you to measure how well you are meeting your customers’ expectations, specifications and requirements.
1. Simple calculations
The calculations for Pp, Ppk, Cp and Cpk are simple and often done by your statistical software.
2. Generic evaluation
Since the capability metrics are a reflection of a specific process’s ability to meet its specs, they can be used to compare disparate functions to determine who is doing better or worse compared to their specific expectations.
3. Can be used for both continuous variable and discrete attribute data
Capability analysis can be used for both quantitative as well as qualitative data. While the formulas and analysis are different, the concept is the same.
Why is capability analysis important to understand?
Capability analysis is a forward looking measure of your process and attempts to predict the short and long term ability of your process to meet your customer specs.
Capability analysis is forward looking
Capability analysis uses historical data to predict the future capability of your process. You already know your current capability since you have the historical data and know how well you are currently meeting your customers’ expectations.
Can be used to see if there has been improvement
If you understand the concepts of capability metrics, you can use them to see if you have made improvements in your process relative to customer specs.
Capability analysis tells you about dispersion and centering of your process relative to customer specs. This will allow you to focus on either shifting the process center left or right or working to reduce the process variation.
An industry example of capability analysis
As a result of increased complaints about quality, Jill, the company Black Belt (BB), was reviewing the control charts and process capability of the two main suppliers of laboratory reagents. Below are the control charts monitored by the quality department. As you can see, Supplier A shows their process is in-control.
On the other hand, Supplier B is showing considerable special cause variation.
Since Supplier B is not in-control, capability analysis can’t be computed. A decision was made to stop ordering from Supplier B until the company is able to figure out how to stabilize their process. Supplier A’s process was in-control and capable so they got the additional business and quality complaints disappeared.
3 best practices when thinking about capability analysis
Here are a few tips for effectively doing your capability analysis.
1. Do a MSA
Given the importance of doing capability analysis, you will want to be confident in the validity of your data. It is recommended you do a Measurement System Analysis (MSA) before doing your capability analysis.
2. Make sure the process is stable
Since capability analysis is forward looking, you need the process to be in statistical control and only exhibiting common cause variation. You will use the appropriate control chart to do that.
3. Can be used for one or two sided specifications
Your challenge will be to determine which of the capability metrics is best suited for your situation. Should you use short or long term capability? Do you have a one or two sided spec? What are your customers’ expected or required levels of capability?
Frequently Asked Questions (FAQ) about capability analysis
Must my process be in-control before doing process capability?
Yes. Process capability is a forward looking projection of how well you will be able to meet your customer’s requirements. Unless a process is in-control and just showing common cause variation, the future is hard to predict and your process capability assumptions may not be accurate.
What is the difference between Cp and Cpk capability calculations?
Cp is calculated on the assumption that the mean of your data is centered within your upper and lower specifications. Cpk allows for the non-centering of your data and is calculated based on the specification closest to your data mean.
Does process capability assume normality of the data?
Most capability calculations have an underlying assumption of normality of your data. If your data is not normal, you can either transform the data (Box-Cox or Johnson) or use special capability calculations based on the specific distribution of your data.
Are you ready to do some capability analysis on your processes?
Process Capability is the future ability of your process to meet your customer specifications. The capability analysis determines how the product specifications compare with the natural variability in a process. The natural variability of the process is the natural variation due to common causes. The other source of process variability is due to special or assignable causes.
The standard deviation that describes the process variation is a critical part of process capability analysis. In general, the standard deviation is not known and must be estimated from the process data. The estimated standard deviation used in process capability calculations may address short‐term or long‐term variability.