# Enter your data for EWMA Chart

Stat > Control Charts > Time-Weighted Charts > EWMA

Select the option that best describes your data.

## All measurements are in one column

Complete the following steps if the measurement data for an EWMA chart are in one column.

1. From the drop-down list, select All observations for a chart are in one column.
2. Enter one or more columns of measurement data. If you enter more than one column, Minitab creates a separate EWMA chart for each column.
3. In Subgroup sizes, do one of the following:
• If all subgroups are the same size, you can enter the subgroup size (for example, 5).
• Enter a column of values that identifies which subgroup each measurement is from.
4. 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.
In this worksheet, Length contains the length of parts, and Subgroup contains the subgroup indicators.
C1 C2
Length Subgroup
0.65 1
0.69 1
0.67 1
0.67 1
0.62 2
0.59 2
0.59 2
0.60 2

## Measurements for each subgroup are in a different row

Complete the following steps if each row in the worksheet contains the measurement data for one subgroup.

1. From the drop-down list, select Observations for a subgroup are in one row of columns.
2. Enter the columns that contain the measurement data.
3. 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 deafult 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.
In this worksheet, the data for subgroup 1 are in the first row across columns 1−4. The data for subgroup 2 are in the second row across columns 1−4.
C1 C2 C3 C4
Measurement 1 Measurement 2 Measurement 3 Measurement 4
0.65 0.69 0.67 0.67
0.62 0.59 0.59 0.60
1 J.M. Lucas and M.S. Saccucci (1990). "Exponentially Weighted Moving Average Control Schemes: Properties and Enhancements," Technometrics, 32, 1−12.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy