Subscriber only lesson.

Sign up to the Statistical Process Control course to view this lesson.

## About this lesson

The U chart relies on counting both defects and units and is appropriate if the process has erratic flow. This lesson explains how the data is recorded and interpreted on the chart. The lesson describes how to create this control chart in both Microsoft Excel and using Minitab. The lesson will include practice creating the chart.

## Exercise files

Download this lesson’s related exercise files.

U Chart.xlsx10.2 KB U Chart - Solution.docx

207.2 KB

## Quick reference

### U Chart

The U chart (plots Counts/Unit) is the attribute data control chart to be used when the focus is all of the counts of an attribute, but the subgroup size varies significantly. It can be easily created in either Microsoft Excel or Minitab.

### When to use

Use the U Chart when it counts the number of an attribute occurrence (defect) within all the units of a subgroup that varies in size with each instance of counting. It is frequently used to count the occurrences of an attribute in a process that has erratic flow.

### Instructions

The U Control Chart tracks the count of occurrences of an attribute (such as a defect) and normalizes those by the number of units in the subgroup. A unit in the subgroup could have no defects, one defect, or many defects. The value is the sum of all the defects divided by the number of units. U Charts are often used to count the number of occurrences in a process that has erratic or uneven flow.

U Charts can be created in Microsoft Excel or in Minitab. Within Minitab, control charts are created by using the “Stat” pull down menu, then selecting “Control Charts.” Within the Control Charts window, select “Attribute Charts” and then finally select “U.” In the Minitab U Chart panel, you will need to select the data column with your data and a second column that has the number of units in each subgroup.

If creating the U Chart in Excel:

- Establish a clear definition of what you are calling a unit. Units should be sized so that there are at least two units in a subgroup. Subgroups should be sized so there are at least five defects in a subgroup.
- Count the number of units in each subgroup. Depending upon how you define units, you can have a fractional number of units in a subgroup. For instance, if you classified a unit as a set of 10 PCB boards. And the subgroup include 95 PCB boards, you would say that your units for that subgroup was 9.5.
- Count the occurrences of the attribute or defect (not the number of defective units) within each subgroup.
- Calculate the Mean and the Upper Control Limit and Lower Control Limit. The UCL and LCL will change with each data point because the number of units (n) is changing with each data point.
- The u ratio for each subgroup is the total counts divided by the number of units in each subgroup.

- Plot the data points, the Mean and the control limits.
- Take appropriate actions to remove special causes or to center your data within the customer spec limits.

### Hints & tips

- Be very clear about your unit definition so that the unit count is accurate.
- Control limits will constantly change, but should remain near the same level.
- The LCL can never be less than zero. If the calculation is a negative number, just use zero for your value.
- When plotting the chart in Excel, use the “Line Graph” charting option with lines that overlay, not ones that stack.

Lesson notes are only available for subscribers.