## Statistical Process Control Concepts

Statistical process control (SPC) is simply a method for monitoring the statistical variations of a process, where the objective is to control or reduce variations between upper and lower process limits. Control charts (Fig. 1) are used to monitor the statistical variations, but this is just one part of the overall SPC method. For the techniques of SPC (specifically Shewhart control charts such as the schematic in Fig. 1) to be successfully employed as an off-line problem identification and problemsolving tool, it is essential to keep in mind that it is a three-step process, as follows:

1. Use statistical signals to find improvement opportunities through the identification of process faults.

2. Use experience, technical expertise, and fault diagnosis methods to find the root cause of the fault that has been identified.

3. Develop an action plan to correct the fault in a manner that will enable any gains that are realized to be held.

This three-step process can be explained by using the classical feedback control system perspective, as shown in Fig. 2. There are five distinct stages in the generic control loop (Fig. 2), which facilitate the three-step process in the following way:

1. Use of statistical signals (observation and evaluation)

2. Fault diagnosis (diagnosis)

3. Action plan (decision and implementation)

Fig. 1 Impact of having process initially in a state of statistical control versus improvement resulting from a breakthrough in performance
Fig. 2 Classical feedback control system view of SPC implementation

However, bringing a process into a state of statistical control does not necessarily mean that a fundamental improvement has been achieved. Clearly, a bad situation has been rectified by bringing the process into control, and quality and productivity are enhanced. However, bringing a process into control simply means that the process is back to where it should have been to begin with. At this point, it is then possible to begin to assess the present ability of the process to realize the potential it was initially intended to have. It may be failing to realize this potential because the implementation of the process is flawed or because the design of the process itself is flawed. In either case, the root cause(s) of the chronic common cause problem must be identified and removed at the system level. This constitutes a breakthrough in performance; that is, an improvement in the process has taken place. The results of the essential steps leading to such a breakthrough are shown in Fig. 1.

This article focuses primarily on the key statistical concepts and definitions that are essential for appreciation of the SPC methods. There are several important concepts that should be understood, including the requirements of rational sampling and the definition of measurements by attributes (or defects). Once these basic definitions are established, the next step in the three-step process is an appreciation of the root cause or fault diagnosis for variations in P/M processing. The factors related to fault diagnosis for P/M quality control and planning are discussed in the section "P/M Process Planning" in this article.

0 0