Statistical process control
| Statistical process control |
|---|
| See also |
Statistical Process Control is a set of techniques and statistical methods used to assess the stability of the process. The purpose of SPC is to prevent non-conformity by detecting and early signalling of interference in the process.
SPC is a response to the ineffectiveness of traditional quality inspection. Instead of controlling final product quality inspectors or the employees themselves check produces parts. They do not wait until defective parts appear. If produced element is dangerously close to acceptable limits (tolerance), the immediately take the necessary actions. These actions may include: informing superiors, tuning machines, replacement of worn parts of the machine, etc.
SPC is geared to continuous improvement. It is primarily preventive. Analysis of the process not only provides information on occurring deviations, but also helps to understand the cause of process variation. Thanks to systematic monitoring, organization can minimize losses by removing identified defects and errors. At the same time managers, based on information about the problems, can design processes so as to prevent their next occurrence of the error using e.g.: Poka yoke, design quality or FMEA [1].
The statistical control
One method of quality inspection is a statistical control, often called sampling. Only samples of product are checked. This type of inspection is used for technical and economic reasons. It is not always possible to measure all the units of production batch. Some forms of control are associated with the destruction of the product.
Depending on the size and frequency of sampling, as well as how feedback is used in the manufacturing process, we can distinguish two types of statistical control:
- Statistical quality control - is passive. It aims to determine whether a given batch of products from which the sample was taken may be accepted.
- Statistical process control - has active character. The results are used not only to evaluate the products but also the whole process. It is about improving the process [2].
SPC Tools
The basic tools used in statistical process control are:
- 7 classic quality methods,
- Fish diagram (cause and effect diagram),
- Check sheet,
- Control charts,
- Histogram,
- Pareto chart,
- Scatter diagram,
- Stratification.
- 7 new quality methods.
- Affinity diagram,
- Interrelationship diagram,
- Matrix diagram,
- Matrix data analysis,
- Arrow diagram,
- Tree diagram,
- Process decision programme chart,
Purpose of using SPC
The purpose of using these tools is to identify the causes of problems. The causes are divided into:
- Random,
- Non-random.
Random causes (also called systemic, natural) are naturally associated with the process. They are common, and their effects are relatively small compared to the causes of non-random. These factors are difficult to identify and eliminate, as the only way out is to change the production system. Examples in this case may be: outdated equipment, inadequate lighting, poor quality material.
Non-random causes occur irregularly, and their effects are usually significant. Typically, they are easy to identify and remove. Examples of non-random causes might be: employee's error, broken or poorly programmed machine. The elimination of these reasons is the basic prerequisite for the achievement of control over the process [3].
More on this in an article on control chart.
References
- ↑ Bank J. Total Quality Management, Publishing Gebethner and Ska, Warsaw 1996
- ↑ Hamrol A., Mantur W., Quality Management. Theory and Practice ', PWN, Warsaw 2002
- ↑ Dahlgaard JJ Gopal K Kanji K. Kristensen, Fundamentals of quality management, PWN, Warsaw 2000
- Example of SPC using Excel or other spreadsheet
- MacGregor JF, Kourti T (1995) Statistical process control of multivariate processes, Control Engineering Practice, Volume 3, Issue 3
- Woodall WH, Mongomery DC (1999) Research issues and ideas in statistical process control, Journal of Quality Technology, 31.4
Author: Slawomir Wawak, Irena Śliwińska
