dcsimg
 

What Control Chart Should I Use?

Control charts are simple but very powerful tools that can help you determine whether a process is in control (meaning it has only random, normal variation) or out of control (meaning it shows unusual variation, probably due to a "special cause").

In an earlier post, I wrote about the common elements that all control charts share: upper and lower control limits, an expected variation region, and an unexpected (or special cause) variation region.  But there are many different types of control charts:  P charts, U charts, I-MR charts...how can you know which one is right? 

Which Control Chart Matches Your Data Type? 

The first step in choosing an appropriate control chart is to determine whether you have continuous or attribute data.

Continuous data usually involve measurements, and often include fractions or decimals. Weight, height, width, time, and similar measurements are all continuous data. If you're looking at measurement data for individuals, you would use an I-MR chart. If your data are being collected in subgroups, you would use an Xbar-R chart if the subgroups have a size of 8 or less, or an Xbar-S chart if the subgroup size is larger than 8.  
  

U-Chart
A U-chart for attribute data plots the number of defects per unit.

 

If you have attribute data, you need to determine if you're looking at proportions or counts. If it's proportions, you'll typically be counting the number of defective items in a group, thus coming up with a "pass-fail" percentage. In this case, you would want to use a P chart.  If you're measuring the number of defects per unit, you have count data, which you would display using a U chart.
  
Of course, we're just scratching the surface here -- there's a lot more to finding the right control chart for each individual situation than we can fit in a simple blog post.
   
But if you're using Minitab Statistical Software, you can choose Assistant > Control Charts... and get step-by-step guidance through the process of creating a control chart, from determining what type of data you have, to making sure that your data meets necessary assumptions, to interpreting the results of your chart. 
 

If you're not using it yet, you can download Minitab and try it for 30 days free.  In addition to guidance for control charts, the new Assistant menu also can guide you through Regression, Hypothesis Tests, Measurement Systems Analysis, and more. As a person who needs to use statistics but isn't naturally inclined toward numbers and math, I find it pretty cool to be able to get that guidance right from the software. 

Fearless Data Analysis

Minitab 17 gives you the confidence you need to improve quality.

Download the Free Trial


 

Comments

Name: Mike O'Connell • Monday, September 26, 2011

Eston, excellent post, I am using this specifically at Kraft Wilkes-Barre site to explain a point I was trying to make in a meeting. The team was reporting count data w/out regard to ongoing process variation. So I'm able to send this link to team to help explain. I like the other posts too, didn't realize you folks were doing this.

Thanks and Regards,
Mike O'Connell


Name: Eston Martz • Monday, September 26, 2011

Thanks for reading, Mike! I'm glad you found the post helpful, and I hope the team does as well. We just started blogging about six weeks ago, so the Minitab Blog is still a little bit of a hidden resource until we can get the word out. Please feel free to share news about the blog with anyone you think could benefit from it!
Thanks for the comment, and best regards!
Eston


Name: AMIT SHARMA • Sunday, November 13, 2011

Hello: Excellent details and sincere thanks to Minitab, you, and team members for willing to take time and post details through blog. Simply briiliant. I self is learning and I am sure others shall also benefit. U guys simply rock !!!

Pls keep up great work !! :-)


Name: Tim DuPuis • Thursday, January 3, 2013

I'm trying to use a chart in Minitab that I can put Upper & Lower Control limits I can put in, and simply chart the data. Which chart should I use in Minitab. Remember I want to be able to put the Upper & Lower Control limits I choose to use.

Thanks,


Name: Eston • Friday, March 28, 2014

Hi Tim - It's hard to direct you to just the right chart, because that will depend on what type of data you have, whether you're using subgroups, etc. However, you can customize your control limits, though it requires some calculations -- you can learn more about it at the link below.

However, your best and by far easiest option if you're using Minitab 16 or 17 is to use the Assistant for control charts (Assistant > Control Charts). Using this tool, you can choose to either have Minitab determine control limits and center line from the data, or use known values as you wish to do. Good luck!

http://www.minitab.com/en-us/Support/Tutorials/Optimize-the-Performance-of-Your-Control-Charts/


Name: Doreen • Tuesday, August 12, 2014

I have attribute data. Product fail or pass the quality inspection.
What chart should I use


Name: Eston • Tuesday, August 12, 2014

Hi Doreen, you can use a P chart, a Laney P' chart, or an NP chart if you're interested in the proportion of defectives per subgroup. You can learn more at this web page or contact our support team for assistance: http://support.minitab.com/minitab/17/topic-library/quality-tools/control-charts/understanding-attributes-control-charts/attributes-control-charts-in-minitab/

Hope this helps!
Eston


blog comments powered by Disqus