Long-Term Process Capability Studies
Not knowing the long-term capability of your processes is like driving your car at night during a snowstorm. Visibility is poor and driving is hazardous. However, knowing the long-term performance of your manufacturing process is like driving a car on a clear day. You can easily see the road and drive confidently between the road lines.
Capability studies statistically compare real-life process variation to product specifications. Long-term process capability is measured by Pp and Ppk. As the name indicates, long-term studies include process performance data that is collected over a long period of time, such as a month. It’s common to collect hundreds or even thousands of data points for use in long-term studies.
Process capability studies require the collection of continuous data types. Continuous data is measured to a decimal point level and opposed to countable-type whole number data. To decide what product features to study, start by looking at customer complaints, defect rates and competitive products. You want to study and improve process variation with the intent of reducing costs and improving the customer experience.
The data that you collect should include all the typical input variables that impact your ongoing process such as raw material lot changes, operator adjustments and environmental fluctuations. It is the process variation caused by these things that you will want to characterize and improve.
You may already have study data available in the form of ongoing quality records. The only other information that you need are the product specs. After data collection, the next step is to enter it into a statistical software package such as Minitab or Excel or a free on-line calculator and generate the study results.
Now to interrupt the study results follow these steps:
First, let’s understand Pp because this reflects our process variation as compared to product specs. You can see in the chart below that the greater the Pp, the greater the process sigma level and the lower the defect rates.
|
Pp |
Std Dev (Sigma) |
DPMO |
Defect Free |
|
0.33 |
1 |
691,462 |
30.85% |
|
0.66 |
2 |
308,538 |
69.146% |
|
1.0 |
3 |
66,807 |
93.319% |
|
1.33 |
4 |
6,210 |
99.379% |
|
1.66 |
5 |
233 |
99.977% |
|
2.0 |
6 |
3.4 |
99.99966% |
If Pp is less than 1.33, then you are at risk of sliding off the process road into defect land. Ideally, your process is performing with a Pp of 2.0 because then you would have a six-sigma process and six-sigma is the gold standard of process management.
Second, let’s compare Pp results to Ppk results. We do this because Pp doesn’t change if the process isn’t centered but Ppk does change. So, if the Ppk is lower than Pp, then we know that our process is running off-center. An off-centered process yields more defects than a centered process. So, center the process to correct this issue.
Third, using the results of the long-term study, compare Pp to Cp. If there is a difference between these measures then run a time-series graph and look for process shifts in the data. Maybe different operators are adjusting the process differently, or maybe raw material lot changes are affecting process performance. Whenever you see a difference between Pp and Cp it means that your process is shifting and you want to get to the root cause of these process shifts.
Fourth, compare the results of your long-term process capability study to your short-term study results. The difference between these two studies indicates your process opportunity. The concept is that if your process can perform at a high-level under ideal short-term conditions, then why can’t it be done every day!
If you work in manufacturing and you are interested in learning more about manufacturing leadership, process management and six-sigma, then consider enrolling in CML300. Tools for the Trenches makes manufacturing experts!
Stay connected with news and updates!
If you want some weekly T4T wisdom coming straight to your inbox for your reading pleasure - look no further! Join our mailing list to receive the latest blogs and updates.
Don't worry, your information will not be shared.
We hate SPAM. We will never sell your information, for any reason.
