To add a control chart, drag and place the control chart from the Assets pane onto the canvas. Minitab Connect adds a placeholder asset card that contains the default title, a representation of the asset, and a drop down that contains the data views associated with the dashboard. Select the data view you want to use for the control chart. Then select Open to open the dialog for the control chart.
You must select a data view before you can create an asset. Use the Data panel to select a data view.
The method to enter your data depends on the type of control chart. Select the option below that matches your data entry setup.
For an Individuals Chart, I-MR Chart, and C Chart, all the measurements for a variable must be in one column. In Variable, enter one column of measurement data.
C1 |
---|
Weight |
785.92 |
830.36 |
798.43 |
812.56 |
799.63 |
822.72 |
786.64 |
803.66 |
811.29 |
826.74 |
If the measurement data are in one column, select this option, then enter a column of measurement data. In Subgroup sizes, if all subgroups are the same size, enter the subgroup size. Otherwise, enter a column of values that identifies which subgroup each measurement is from.
C1-T | C2 |
---|---|
Subgroup | Measurements |
One | 10 |
One | 13 |
One | 10 |
Two | 8 |
Two | 8 |
Two | 9 |
If each row in the worksheet contains the measurement data for one subgroup, select this option, then enter the columns that contain the measurement data.
C2 | C3 | C4 |
---|---|---|
Measurement_1 | Measurement_2 | Measurement_3 |
10 | 13 | 10 |
8 | 8 | 9 |
For a Laney P' Chart, U Chart, and P Chart, you must enter a variable column and specify the subgroup size. In Variable, enter the column that contains the process data.
C1 | C2 |
---|---|
Irregularities | Yards Sampled |
10 | 120 |
8 | 110 |
7 | 110 |
7 | 115 |
In Weight of EWMA, enter the weight to use in the exponentially weighted moving average. The value must be between 0 and 1. If you change the default weight (0.2) and the number of standard deviations for the control limits, you can construct a chart with specific properties. Choose the weight based on how small of a shift you want to detect and how fast you want to detect that shift while maintaining an acceptable false alarm rate. You can choose combinations of these two parameters by using an ARL (average run length) table. See Lucas et al.1 for an extensive table.
Specify a time scale for the x-axis.
Label the x-axis with the number of the subgroup.
Label the x-axis with values from a column. In Stamp column, enter a column that contains date/time, numeric, or text values for the scale.
The following options are on the Estimate tab of the Options sub-dialog.
If some observations have erratic data because of special causes that you have corrected, you can omit these observations from the calculations to avoid incorrect parameter estimates.
List the observations. Use a colon to indicate a range of observations. Leave a space between each observation or range of observations.
For example, to specify observations 2, 9, 44, 45, 46, and 47, enter 2 9 44:47.
Select Use unbiasing constant to eliminate bias that can occur when you estimate certain process parameters. Often, the choice to use unbiasing constants depends on company policy or industry standards.
Enter the number of observations that are used to calculate the moving range. By default, a length of 2 is used because consecutive values have the greatest chance of being alike. The length must be 100.
If the data are cyclical, consider using a moving range of a different length. For example, if you collect quarterly data, consider using a moving range of length 4 to ensure that one observation from each season is included in the calculation.
On the Estimate tab of the I-MR Options subdialog, select Use Nelson estimate to correct for unusually large moving range values in the calculation of the control limits. The procedure is similar to the procedure proposed by Nelson. Connect eliminates any moving range values that are more than 3 standard deviations larger than the average moving range, then recalculates the average moving range and the control limits.
By default, Connect calculates the control limits using the actual subgroup sizes. When the subgroup sizes differ, the control limits are uneven, but you can force the control limits to be straight. Under When subgroup sizes are unequal, calculate control limits, select Assuming all subgroups have size, and enter a subgroup size.
When you specify a subgroup size, you change the calculations for the control limits, which can change the results of the tests for special causes. Use this option only if the differences between the subgroup sizes are small. Don't use this option when the difference between subgroup sizes is more than 25%. For example, if the largest subgroup has 10 observations and the smallest subgroup has 8 observations, then the difference is 20% ((10 - 8) / 10 = 0.2 = 20%).
For example, the data for the following charts is the same, but the control limits for the second chart were calculated based on a fixed subgroup size.
On the Options subdialog, Connect provides one to eight tests for special causes, depending on the control chart. By default, Connect uses only Test 1. Select additional tests based on company or industry standards. Use the tests to determine which observations to investigate, and to identify the specific patterns and trends in your data.
In the drop-down list, specify whether to perform some, all, or no tests for special causes. You can make each test more or less sensitive by changing the value of K.
In the example below, the values in the Stage column change from 1 to 2 to indicate the beginning of stage 2. The values change from 2 to 3 to indicate the beginning of stage 3.
C1 | C2 |
---|---|
Measurements | Stage |
... | ... |
9 | 1 |
10 | 1 |
8 | 2 |
11 | 2 |
... | ... |
7 | 2 |
8 | 2 |
5 | 3 |
7 | 3 |
... | ... |
To see only the most recent observations instead of all observations, enter the number of observations to include in the chart. Connect still uses all the data to calculate the control limits and the center line.
By default, Connect displays the last 25 observations. This is different than Minitab Statistical Software, which displays all observations by default. This could result in different results if you don't change the settings making charts with the same data between the two applications.
Connect displays all observations.
This chart displays only the most recent number of observations that you enter.
If you turn on an alert for a reference line, you still need to create a new alert to receive a notification. See the following section for more information on how to set up a new alert.
After you add a reference line, you can use the Graph Options tab to change the value, color, and name of the line, change the alert status, or delete the line.
For emails and texts, use the dropdown list to select a user to add them to the alert. If you want to add an email or phone number for a person that is not in the dropdown list, type it into the box. After Connect verifies that it is valid, select Press Enter to Add Email or Press Enter to Add Phone Number to add it to the alert.
United States phone numbers must be 10 digits with any combination of parentheses, periods, dashes, and spaces as delimiters. Parentheses can only be used on the area code. International phone numbers must start with a + symbol and can contain up to 15 digits with no spaces or delimiters.
To enable, disable, edit, or delete an alert after you create it, select the ellipsis in the header bar and select Manage Alerts.