How to Create a Control Diagram: 13 Steps

Table of contents:

How to Create a Control Diagram: 13 Steps
How to Create a Control Diagram: 13 Steps
Anonim

Control charts are an effective tool for analyzing the performance of the data needed to evaluate a process. They have many uses. They can be used in industry to test, for example, if the machinery is making products within the pre-established quality specifications. They also have many simple applications: professors use them to evaluate test scores. To create a control chart, it is useful to have Excel - it will make your life easier.

Steps

Create a Control Chart Step 1
Create a Control Chart Step 1

Step 1. Check that your details meet the following criteria:

  • The data should usually be normally distributed around an average.

    In the example below, a company that produces bottles fills them around 500ml (average). In Anglo-Saxon measures it is 16 ounces. The company is evaluating the validity of their production process

  • The measurements must be independent of each other.

    In the example, the measurements are divided into subgroups. The data in the subgroups should be independent of the number of measurements; each data point will have a subgroup and a number of measurements

  • Example:
Create a Control Chart Step 2
Create a Control Chart Step 2

Step 2. Find the mean of each subgroup

  • To find the mean, add all the measurements in the subgroup and divide by the number of measurements in that subgroup.

    In the example, there are 20 subgroups and in each subgroup there are 4 measurements

  • Example:
Create a Control Chart Step 3
Create a Control Chart Step 3

Step 3. Find the mean of all means from the previous step (X)

  • This will give you the overall average of all data points.
  • The overall average will be the central axis of the graph (CenterLine = CL), which is 13.75 in our example.
Create a Control Chart Step 4
Create a Control Chart Step 4

Step 4. Calculate the standard deviation (S) of the data (see Tips)

Create a Control Chart Step 5
Create a Control Chart Step 5

Step 5. Calculate the upper and lower limit (UCL, LCL) using the following formula:

    • UCL = CL + 3 * S
    • LCL = CL - 3 * S
    • The formula represents 3 standard deviations above and 3 below the mean, respectively.
    Create a Control Chart Step 9
    Create a Control Chart Step 9

    Step 6. See the chart below with steps 7 to 10

    Example:

    Create a Control Chart Step 8
    Create a Control Chart Step 8

    Step 7. Draw a line at each detour

    • In the example above, there is a line drawn at one, two, and three standard deviations (sigma) from the mean.

      • Zone C is 1 sigma from the mean (green).
      • Zone B is 2 sigma from the mean (yellow).
      • Zone A is 3 sigma from the mean (red).
      BS Your Way Through a College Paper Step 9
      BS Your Way Through a College Paper Step 9

      Step 8. Draw the mean control chart (X barred), graphically representing the subgroup of means (x-axis) versus the subgroup of measurements (y-axis)

      The graph should look something like this:

      Example

      Create a Control Chart Step 8
      Create a Control Chart Step 8

      Step 9. Evaluate the graph to see if the process is out of control, ie beyond the allowable values

      The chart is out of control if any of the following occurs:

      • Any point falls beyond the red zone (above or below the 3 sigma line).
      • 8 consecutive points fall on the same side of the average line.
      • 2 of 3 consecutive points fall within zone A.
      • 4 of 5 consecutive points fall into zone A and / or zone B.
      • 15 consecutive points are within zone C.
      • 8 consecutive points are not in zone C.
      Create a Control Chart Step 10
      Create a Control Chart Step 10

      Step 10. Check if the system is within or out of all acceptability

      Advice

      Use Excel when creating graphs, because it contains functions that allow you to speed up calculations

      Warnings

      • Control diagrams (generally) are based on normally distributed data. In practice, however, they are reasonably outside the norm.
      • For some graphs, such as graph C, it may happen that the data is not normally distributed.
      • Moving average charts use different rules of interpretation to meet the demands of high non-normality of the data.
      • Barred average charts tend to be distributed normally even if the underlying data is not.

Recommended: