Control Charts for Variable and Attributes

Variable Control Charts

Consider that you are evaluating the output from a process.  Conceptually, you could evaluate the products in two basic ways.  In the first way you would simply classify the products as "conforming" or "non conforming."  This produces attribute (discrete) data.  In the second way you could measure a key characteristic using a continuous scale.  This produces variable (continuous) data.

Variables control charts are used to evaluate variation in a process where the measurement is a variable--i.e. the variable can be measured on a continuous scale (e.g. height, weight, length, concentration). There are two main types of variables control charts.  One (e.g. x-bar chart, Delta chart) evaluates variation between samples. Non-random patterns (signals) in the data on these charts would indicate a possible change in central tendency from one sampling period to the next.  One way of thinking about the use of a variables control chart is that you are testing the hypothesis that a particular sample mean came from the population of sample means represented by the control limits of the process.  If the particular sample mean is within the control limits, your concusion is that it does come from that population.  If the particular sample mean is outside the control limits, you conclusion is that it may have come from some other distribution (i.e. a distribution with a mean that is higher or lower than this population mean.  [NOTE:  There are other signals that may indicate an out-of-control signal that will be discussed in the Lesson Six Presentation.]

The other type of variables control chart (e.g. R-chart, S-chart, Moving Range chart) evaluates variation within samples.  Non-random patterns (signals) in the data on these charts would indicate a possible change in the variation within the samples.

Non-random patterns in the data plotted on the control charts provide evidence of the process being in-control (only common cause variation present; predictable) or out-of-control (common cause andassignable cause variation present; unpredictable).  Adjusting a process which is in-control will result in increased variation.  Failing to adjust a process which is out-of-control results in a loss of predictability.  Control charts help a machine operator or manager to decide when it is appropriate to make an adjustment and when it is better to leave the process alone.

                                                 Attribute Control Charts

These charts are applied to data that follow a discrete distribution.

Types of attributes control chart:

p chart

This chart shows the fraction of nonconforming or defective product produced by a

manufacturing process.

It is also called the control chart for fraction nonconforming.

np chart

This chart shows the number of nonconforming. Almost the same as the p chart.

c chart

This shows the number of defects or nonconformities produced by a manufacturing process.

u charts

This chart shows the nonconformities per unit produced by a manufacturing process.