How to Create a Control Chart

Control charts are an efficient way of analyzing performance data to evaluate how a process changes over time. Control charts are also called statistical process control, or SPC, charts, and they have many uses, like checking machinery or evaluating test scores. It might sound complicated at first, but it's a pretty easy process—especially if you're using software like Excel, Power BI, or Tableau. Our guide will show you how to create a control chart to track your own process data!

Steps

Download Article

1
Check to see that your data meets the following criteria:
- Data should usually be normally distributed revolving around a mean (average).
  - In the example below, a bottle company fills their bottles to 16 oz. (mean); they are evaluating if their process is “in-control”. The amount in ounces over 16 oz. is normally distributed around the mean.
- Measurements need to be independent of one another.
  - In the example, the measurements are in subgroups. The data in the subgroups should be independent of the measurement number; each data point will have a subgroup and a measurement number.
2
Find the mean of each subgroup.^{[1]XResearch source}
- To find the mean, add all measurements in the subgroup and divide by the number of measurements in the subgroup.
  - In the example, there are 20 subgroups and in each subgroup there are 4 measurements.
Advertisement
3
Find the mean of all of the means from the previous step (X).^{[2]XResearch source}
- This will give you the overall mean of all the data points.
- The overall mean will be the centerline in the graph (CL), which is 13.75 for our example.
^{[3]XResearch source}
5
Calculate the upper and lower control limits (UCL, LCL) using the following formula:^{[4]XResearch source}
- UCL = CL + 3*S
- LCL = CL – 3*S
- The formula represents 3 standard deviations above and 3 standard deviations below the mean respectively.
7
Draw a line at each deviation.
- In the above example, there is a line drawn at one, two, and three standard deviations (sigma's) away from the mean.
  - Zone C is 1 sigma away from the mean (green).
  - Zone B is 2 sigma away from the mean (yellow).
  - Zone A is 3 sigma away from the mean (red).
^{[5]XResearch source} Your graph should look like something like this:
9
Evaluate the graph to see if the process is out-of-control. The graph is out-of-control if any of the following are true:
- Any point falls beyond the red zone (above or below the 3-sigma line).
- 8 consecutive points fall on one side of the centerline.
- 2 of 3 consecutive points fall within zone A.
- 4 of 5 consecutive points fall within zone A and/or zone B.
- 15 consecutive points are within Zone C.
- 8 consecutive points not in zone C.

Sample Control Charts

Sample Control Chart

Sample Control Chart for Subgroups

Sample Control Chart for Multiple Groups

Community Q&A

Search

Add New Question

Question
What is the minimum number of samples that I would need to create a valid control chart?
Community Answer
You need for there to be a sample size of at least 30 for the distribution to be considered normal.
Thanks! We're glad this was helpful.
Thank you for your feedback.
If How.com.vn has helped you, please consider a small contribution to support us in helping more readers like you. We’re committed to providing the world with free how-to resources, and even $1 helps us in our mission.Support How.com.vn
Yes No
Not Helpful 5 Helpful 9

Ask a Question

200 characters left

Include your email address to get a message when this question is answered.

Submit

Video

Tips

Excel commands that will make your data analyze easier are:
Thanks
Helpful6 Not Helpful17
Use Excel when creating your charts, because it has built in functions (shown below) that will speed up your calculations.
Thanks
Helpful2 Not Helpful11

Submit a Tip

All tip submissions are carefully reviewed before being published

Submit

Thanks for submitting a tip for review!

Warnings

Control charts are (generally) based on data being normally distributed. In practice, however, they are reasonably robust to non-normal data.
Thanks
Helpful1 Not Helpful0
X-bar charts, such as the one explained above, chart means rather than individual values. As explained by the Central Limit Theorem, Means tend to be normally distributed even if the underlying data are not.
Thanks
Helpful0 Not Helpful1
Moving Range charts (often used with the Individuals Chart (also known as the X or I chart) use different interpretation rules to cater for the very strong non-normality of Moving Range data.
Thanks
Helpful0 Not Helpful4
For some charts, e.g. the c chart, data probably will be non-normally distributed.
Thanks
Helpful0 Not Helpful0