Statistical process control
The statistical process control (including statistical process control and statistical process control, English statistical process control, SPC called ) is usually understood as a procedure for the optimization of production and service processes due to statistical methods.
A more comprehensive definition is: " Statistical Process Control is, first and foremost, a way of thinking Which happens to have some tools attached. " May be translated as: " Statistical process control is, first and foremost, a way of thinking that happens to have some tools with. "
History
SPC was developed by Walter A. Shewhart. The scientific foundations were 1931 fully derived and described by him in the book Economic Control of Quality of Manufactured Product. This work was triggered by the intention of management of the Hawthorne Plant of the Western Electric Company in Chicago to produce convergent and thus reliable products. An attempt to accomplish this by means of common sense, failed. As a result, Shewhart was asked by the Bell Telephone Laboratories in New York for support.
Shewhart was based on the assumption that the quality of the final product depends essentially on the combination of the scattering parameters of the items. The cause of this scattering he found two fundamentally different mechanisms:
The second important finding of Shewhart was that now in an attempt to minimize this scattering, two mistakes can be made:
- Assign a deviation of a particular cause, even though it was caused by a general Cause: Error 1.
- Assign a deviation of a general cause, even though it was caused by a particular cause: Error 2.
While it can be either one or the other errors are completely avoided, but never both at the same time. There had to be some way to minimize the cost of error prevention. Extensive statistical research and theory led eventually to the development of Shewhart control charts (German quality control charts ) as the optimal tool to implement the findings into daily practice.
Your first industrial-scale application found SPC in World War II, where it was used in the production of armaments.
Later William Edwards Deming realized that can be applied with the same positive results, these insights and tools on all kinds of processes (business processes, management processes, etc.). This teaching was mainly in Japan on fertile ground, where it was further developed, among others, within the Toyota Production System.
Today, statistical process control is seen as part of a quality management system and supports as a service process the core process of production or service. All statistical methods that are used to monitor and optimize the core process are summarized under the term statistical process control. These methods go beyond the various control charts techniques and also include, for example, the methods of statistical experimental design, the FMEA or the collection of methods Six Sigma with a. Sizes of the SPC flow in customer-supplier relationships as a process capability indices.
Procedure
After the process has to be examined clearly defined, must be defined by a process expert, which measured variables are important. These must be recognized as scheduled during production. The analysis is then performed using control charts (eg, -, - or card).
There are now also software packages that are trying to establish the statistical process control as a key component of computer-aided quality assurance CAQ. The measurements made for this purpose are partly carried out automatically by a machine data acquisition (MDA ) and processed statistically accordingly.
Benefit
SPC is used to comply with a predefined level of quality possible cost, it is not suitable to increase the quality of products. One that goes beyond the required quality level location could produce extra cost, which would be assigned to only an insignificant additional benefit. Typically, the measure is needed to error- free parts at a value of only 99.73 %, which can be achieved with relatively little control effort and thus low cost. An increase in the quality to 100% defect-free parts would increase greatly the control effort, and indeed much more than the difference of 0.27 %, as the total cost increase exponentially with the desired quality level ( including through more tests, better testing equipment and production machinery, appropriate production processes, etc.). It is here spoken of uneconomic " over-provisioning ". Thus, SPC is the economic minimum principle (English minimal principle).
For the targeted increase in the desired quality level of other quality management methods are necessary, such as FMEA.
Software
SPC is usually applied with software support. There are three kinds of software are used. First, generic calculation software such as spreadsheets or statistical packages, secondly Full range CAQ solutions such as from the following manufacturers ( in alphabetical order): AHP, ASIDATAMYTE, Babtec, Böhme & Weihs, CAQ AG, Gewatec, IBS, iqs Software GmbH, Pickert & Partner, Q -DAS, QUIPSY, SCIIL AG or Syncos. Beginners usually use spreadsheets. However, special SPC and CAQ products can reduce the workload and central evaluations enable: PLC software is usually offered with vendor-specific regulations and can read data directly from measurement tools and machinery, while a CAQ system besides SPC further quality management modules such as FMEA, supplier evaluation or process control plans includes several levels and products.