TABLE OF CONTENTS Statistical process control is the application of statistical techniques to control a process. In 1924, Walter. A. Shewhart of Bell Telephone Laboratories developed a statistical control chart to control important production variables in the production process. This chart is considered to be the beginning of statistical process control (SPC). It is one of the first quality assurance methods introduced in modern industry.
Statistical process control is based on a comparison of "what is a normal process," which is based on the data from a period of normal operation, with "what is happening now," which is based on a sample of data from current operation. The data collected from the normal operation condition is used to build control charts and control limits. The control charts and limits are built based on relevant statistical theory. The control limits are so designed that if the current operation is not much different from the normal operation, the statistic calculated from current data is within control limits. If the current operation is significantly different from the normal operation, then the statistic calculated from the data is outside the control limits. We treat this as an out-of-control situation. In statistical process control theory, the out-of-control situation is usually caused by assignable causes, or special causes, such as a sudden change in incoming material, degradation or malfunction of the machine, change of operators, etc. Usually, the production is stopped and investigation is conducted to find and eliminate assignable causes.
For univariate cases, that is, when...
