Joshua is a graduate student at the USF. He has interests in business technology, analytics, finance, and lean six sigma.
Why We Use P-Charts
A P-chart is one of the most commonly used types of statistical process control (SPC) charts. The P-chart tracks the percentages of attributes over a period of time. Whenever a process is going out of range, the person administering the P-chart can make adjustments to the process when trends appear to be occurring.
The number of samples in a P-chart can range anywhere from 2 to 30, but the sample size can be unlimited. The data used in this Minitab tutorial is not from random sampling. Each sample had a size of 25. A sample was marked either satisfied or dissatisfied. An important piece of data for the analysis would be the proportion of dissatisfaction present in the samples. Download a copy of the Minitab file to follow along.
Considerations for P-Chart Data
Before getting into an example take a look at some guidelines that should be followed to ensure that P-Chart results are valid. Below are guidelines for collecting data.
Data Must Be Ordered by Time
Control charts detect change overtime, so it is important that your data is in the correct order of time. The oldest data must be collected first and will appear at the top of the column if using columns or to right of the rows if using data that is set in rows.
Collect Data From Specific Time Intervals
Before the data is collected a specific time interval to collect that data needs to be established. It’s very important that an interval be as short as possible, so changes can be made to compensate for variations in the chart. Intervals could be every 10 minutes, every hour, every production shift, or every day. After the interval is decided on, the data collector must be consistent with that interval while collecting data.
Data Should Be Collected in Subgroups
Data should be collected in subgroups under the same conditions. Subgroups are samples of similar items in the same process that are under evaluation. If the subgroup data are not collected under the same conditions, distinguishing between special cause and common cause may be impossible.
The Subgroups Size Should Be Large Enough
To ensure that control limits are accurate even though they are estimated, the subgroup size for your P-Chart needs to be large enough. There is a formula to determine whether subgroups are large enough. The required subgroup size (n) depends on the mean proportion of defects (P-hat). The formula to determine whether subgroups are big enough is n=.5 / P-hat. The result of n will determine the least amount of a sample size you can have rounded to the nearest whole number. You won’t be able to calculate this until you know the proportion of defects in a process.
The Data Must Have Enough Subgroups
Your data must have enough subgroups to have accurate control limits. It is still possible to use a P-Chart on a preliminary basis if there are not enough subgroups. As more subgroups are collected the control limits can be recalculated.
Add Data to Minitab
Data can be pasted into Minitab easily from Excel and other applications. Select of the data that needs to be copied and press Ctrl +C. Next, open Minitab and paste the data in the upper left-hand corner of the table.
Set up the P-Chart
Now that we have data P-chart adjustments can be made. Start the P-chart generation process by clicking on Stat→Control Chart→Attribute Chart→P.
Add the "No. Dissatisfied" to the variables text box by double-clicking on "No. Dissatisfied" while the cursor is the in variables text box. Now add the sample size of 25 to the subgroup sizes.
Click on the "P Chart Options" button and click on the "Storage" tab to bring you to the storage options. Now click in the box labeled proportions. This will save the proportion of dissatisfied data on the worksheet within Minitab. Next, while still in the "Storage" tab, click on all the checkboxes in the below section except for the stage checkbox. This will allow these values to display at each data point. Next, click on the "Tests" tab and select "Perform all tests for special causes." Click "Ok" to exit P Chart Options.
Click on "Labels" to modify the labels on the chart. In the title section I want to add "P-chart Dissatisfaction" then click OK. Now click OK in the P-chart window to generate the chart.
At first, it appears that there is a problem with the P-chart because there are many points on the lower tolerance limit (LTL). This is actually a good thing. All of the points on the LTL represent samples where there was no dissatisfaction. The greater concern would be the high proportion of dissatisfaction trends in the chart. When these trends are caught the appropriate corrections to the process can be made.
Saving Your Project And Exporting The Graph
To save your file go to File →Save As, then select the location to save the project. There are plenty of ways to export the P-chart from the project. One way is to right-click on the graph and copy the image. Next, the image can be pasted where it needs to go. Another method is to right-click on the image and export the image to a Microsoft Word or Excel document.
- Boyer, K. & Verma, R. (2010). Operations & supply chain management for the 21st century. Mason, OH: South-Western.
- Minitab. (n.d.). Overview for P Chart. Retrieved January 5, 2019, from http://support.minitab.com/en-us/minitab-express/1/help-and-how-to/control-charts/how-to/attribute-control-charts/p-chart/before-you-start/overview/
- Minitab. (n.d.). Overview for importing data. Retrieved January 5, 2019, from https://support.minitab.com/en-us/minitab/18/help-and-how-to/data-input-and-output/open-files-and-import-data/overview/
- Minitab. (n.d.). Data considerations for P Chart. Retrieved January 5, 2019, from http://support.minitab.com/en-us/minitab-express/1/help-and-how-to/control-charts/how-to/attribute-control-charts/p-chart/before-you-start/data-considerations/
- How to Create an Xbar-R Chart in Minitab 18
This article describes why we choose to use X-Bar R charts and how to construct the graphs in Minitab, with an example.
- How to Create a Pareto Chart in Minitab 18
This is a tutorial on how to create a Pareto chart in Minitab 18. It includes sample data to follow along with while a chart is constructed and explained.
- How to Complete a Regression Analysis in Minitab 18
This tutorial on how to create a regression in Minitab includes sample data to follow along with. Explore this straightforward explanation and short analysis of the results.
- How to Calculate Process Capability in Minitab 18
This article shows you how to compute process capability using Minitab 18. You can read along or follow tasks with a provided Minitab file data. A short analysis of the data is conducted.
To learn more about using MiniTab I recommend the following affordable book Minitab Demystified by Andrew Sleeper.
This content is accurate and true to the best of the author’s knowledge and is not meant to substitute for formal and individualized advice from a qualified professional.
© 2018 Joshua Crowder
Joshua Crowder (author) from Tampa, FL on February 11, 2018:
Great! If it helped at least one person I'm satisfied.
Louise Powles from Norfolk, England on February 11, 2018:
That was very helpful to read. Thankyou.