文档介绍：Control Charts
Module Objectives
Review the basics of Control Charts
Xbar & R and P-Charts
False Alarms and Sensitivity to Shifts
CUSUM and EWMA Charts
False Alarms and Sensitivity to Shifts
Examples and Simulations to Demonstrate Concepts
2
Review
Statistical Process Control (SPC) is a methodology for monitoring a process to identify special causes of variation and signal the need to take corrective action when such action is appropriate.
SPC relies heavily on Control Charts for that purpose.
In any process, regardless of how carefully designed or well maintained, a certain amount of natural or inherent variability will always exist.
This natural variability is the cumulative effect of many small, essentially unavoidable causes.
In the framework of SPC, such a process is operating with ‘common causes’ of variation and is said to be in statistical control.
3
Review
Other kinds of variability may be present or develop in time in the process, that are not inherent in the process but due to:
faulty equipment
operator error
defective materials
Such variability can either shift mean process performance or add variability around the mean. Sometimes it does both.
Since the source is not part of the system design, it is thought to be eptable. Resources can be assigned to detect it’s existence, and initiate corrective action.
Such variability is called “assignable cause” or “special cause” variation. A process that is operating in the presence of assignable cause variation is said to be out of control.
4
Review
A major objective of statistical process control is to quickly detect the occurrence of assignable causes of process shifts and increased variability so that investigation of the process and corrective action may be undertaken.
In the 1920’s, Walter Shewhart developed control charts as an on-line process control technique specifically for this purpose.
5
Review
The basic structure of a Shewhart Control Chart is as follows:
UCL = Upper Control Li