Statistical Process Control (SPC)
Traditionally enterprises have depended on their production departments to make products and on their quality control departments to inspect and screen out items that do not meet specifications. Often this approach results in reiterative inspections in an effort to detect instead of prevent problems. Obviously this approach is wasteful because it allows time and materials to be invested in products or services that are not always usable. After the fact inspection is both uneconomical and unreliable.
Statistical process control, on the other hand, is a preventative system. Because it provides immediate feedback, it can minimize or eliminate waste. There are essentially four elements involved in SPC:
- The process: The combination of people, equipment, materials, methods, and environment that work together to produce output.
- Information about Performance: Process output provides qualitative and quantitative information about process performance. In a broad sense, process output includes not only the products that are produced, but also any intermediate 'outputs' that describe the operating state of the process, such as temperatures or cycle times. Collected and interpreted correctly, this data can provide the information you need to determine whether the product or process or both require corrective action.
- Action on the process: Action on the process is future oriented because it prevents the production of out-of-specification products. Corrective actions can include changes in operations (e.g. operator training), raw materials, and even in the process itself. Process changes might include equipment repair and maintenance or the addition of temperature and humidity controls.
- Action on the output: Action on the output is past-oriented, because it involves detecting out of specification output already produced. Unfortunately, if current output does not consistently meet customer requirements, it may be necessary to sort all products and to scrap or rework any nonconforming items.
Process control focuses on gathering process information and analyzing it so that actions can be taken to correct the process itself.
Process Variation
To use process control, it is important to understand the concept of variation. Some sources of variation in the process cause short term or piece to piece differences, such as backlash and clearances within a machine and its fixturing. Other sources of variation cause changes in the output over the long term. Consequently, the time period and conditions under which measurements are made have a direct affect on the amount of total variation.
They are two types of variations: common cause and special cause variations. Common cause variations occur when processes are in statistical control. They are inherent to the system and are therefore difficult to reduce or eradicate. The variability that exists within the control limits of a typical control chart is usually due to common causes. Special cause (often called assignable cause) variations can be attributed to factors or sets of factors that are external to the system. Examples of special cause variations include operator errors, poor machine maintenance, and missed process steps. Special cause variations can be detected by simple statistical techniques one of which is the control chart.
See Also
Control Charts