Defect classification
The error classification assigns errors to different classes according to different criteria and thus specifies an order of which errors must be treated with priority.
General
According to DIN EN ISO 9000, an error, sometimes also referred to as non-conformity , is the failure to meet a specified, presumable requirement. An error is therefore a characteristic value that does not meet the specified requirements. It also serves the company-wide, uniform handling of errors.
According to the cause of the error according to Poka Yoke
Poka Yoke means preventing avoidable mistakes and plays a key role in the Toyota production system. By using an “error list”, errors can be avoided from the outset or identified quickly and effectively and eliminated immediately.
Accidental, deliberate, and willful mistakes
Errors can initially be divided into three categories according to the intention of human error : Errors can happen accidentally or through carelessness or deliberately through sabotage or theft. It can also lead to willful mistakes by deliberately violating rules or regulations.
Other causes
Other causes for the occurrence of errors can be forgetfulness, misunderstandings, missing standards (incorrect or incomplete process or work instructions), lack of experience or incorrect execution (twisting, mixing up, mixing up parts or production steps).
After Six Sigma
Six Sigma is a method for increasing the quality and efficiency of business processes by permanently eliminating the causes of errors and by minimizing process spread. It is used to increase quality, avoid errors or save costs. The errors are divided into two classes:
A error
Missing, incorrect or undetected customer claims are called A errors.
B error
B errors are errors that incorrectly implement existing requirements, such as the incomplete implementation of specifications or classic software errors.
According to the requirements of the standards
The standards, such as ISO / TS 16949, require the monitoring, identification and evaluation of features that deviate from the standard or specification and a subsequent subdivision into two main classes. However, companies are free to define further classes and criteria.
Special / critical features
Special or critical characteristics, depending on the company and standard, can also be referred to as important characteristics or characteristics that require documentation, as well as critical or significant characteristics in English. Similar to the characteristics of the critical error, critical characteristics pose a risk due to serious deviations in terms of functionality and further processing (e.g. screw breaks during assembly), compliance with legal regulations, the safety of the product and thus for the lives of employees or customers (e.g. electric shock due to insufficient insulation, customer falls from the ladder due to the rung breaking). The deviation of the feature is not reliably noticed before delivery and can therefore have negative effects on the image and customer satisfaction. The FMEA evaluation is for the probability of occurrence> 3, the probability of detection> 3, and the meaning 8-10.
Normal characteristics
Normal features, on the other hand, do not deviate from the norm or specification and therefore do not represent an error.
For errors in form and content
For the transparent handling of errors found subsequently in publications, the statistical offices of the federal and state governments have established a classification and treatment based on this:
Particularly serious errors in content
Particularly serious errors in content can damage the image. However, it can be difficult to distinguish between minor and major content errors. This is therefore done on a technical basis by the relevant statistics department. The following error classes are divided:
Serious errors in content
Changing the content of the publication can result in economic consequences or data protection violations.
Minor errors in content
These do not change the statement of the publication or only change it slightly.
Formal mistakes
They are considered to be blemishes that have no bearing on the data and information published. For example, spelling and grammatical errors.
According to the severity of the consequences of the error
According to DIN 40080, three error classes can be distinguished according to the severity of the consequences and the effort required to rectify them:
Critical Defect
Critical errors have high consequences. There are safety-critical deviations due to a change in the product or a partial, or more rarely a complete, functional failure. They violate legal requirements, render the product unusable or pose a risk to the environment, health or life and require immediate corrective measures or product recalls, which can result in high costs. Examples of this are faults in safety devices, such as braking systems or food contamination, for example salmonella infestation.
Major Defect
Major errors are non-critical errors that cause moderate consequences. Although they allow the product to be used, the failure affects essential functions, which greatly reduces the usability of the product. In this case, the customer would return the product. Possible examples here are the failure of the gearbox or the over-salting of food.
If this classification is not sufficient, a further distinction can be made between a main error A and B: The main error A results in the loss of the product or the uselessness and the product must be sorted out as scrap, whereas in the case of main error B the usability is only partially impaired and can be restored by rework.
Minor Defect
Minor defects are understood to be insignificant defects without significant consequences, which represent a reduction in quality with little or no impairment of the function and thus the usability of the product. For example visual defects ( blemishes ).
This class can also be further subdivided into minor defects A and B. With minor defects A, the usability is only slightly impaired. The secondary error B is understood to be an error that does not impair usability, but should not occur. It is also known as a trivial error.
According to the priority
The classification according to priority is particularly used in software development. For this reason, the examples are chosen from this area:
High priority errors (high)
High priority errors hinder the development team and make the system unusable. In this case, they should be remedied as soon as possible, even before the product is shipped, even if this is sometimes very time-consuming and results in the software testing being stopped.
Medium Priority Errors (Medium)
Here, too, the error must be corrected before the product is shipped. However, the development team is not hindered in its work, so that the problem can be resolved in the development cycle without interruption.
Low priority errors
These errors can be eliminated with simple actions. For this reason, they are only dealt with after the higher priority errors have been corrected.
After the error weight classification
Companies subdivide errors according to the cost of eliminating them, which can be expressed by the weighting of errors.
The following table shows the column values of the respective error characteristic, which then add up to the error weighting:
| Column Values | 1 | 4th | 9 | 16 | 25th |
|---|---|---|---|---|---|
| How can the error be identified? | very easy | light | conditionally difficult | heavy | very difficult |
| Where is the error found? | --- | still in the manufacturing department | still in operation | after assembly | at the customer |
| Will the part be rework or scrap? | --- | Rework | Rework with possible rejects | --- | Committee |
| How often does the error occur? | Rare | --- | frequently | --- | very often |
| How is the customer's reaction? | weak | --- | violently | --- | very violently |
| How high are the costs of the defective part in relation to the wages of a skilled worker? | low (≤ 1 h) | --- | medium
(1–5 h) |
--- | high (> 5 h) |
| How high is the wage expenditure to eliminate the error? | ≤ 10 min | 10 - 30 min | 30 - 60 min | 1 - 5 h | > 5 h |
| Is the delivery or production date at risk? | No | --- | possibly | --- | Yes |
The problem here, however, is that the evaluation scale is subject to the experience of the filler and is therefore not an objective mathematical evaluation, but a subjective evaluation based on practical experience.
According to the impact on the customer
Importance to the customer
According to the importance for the customer, a classification of “no importance at all” = 0, via “low” = 1 and “medium” = 2 to “very large” = 4 can be made. Since errors that have a strong impact on the customer are of particular importance in this classification, "4" is used in place of the number "3" to clarify the distance to category 2 errors.
Probability of perception
In addition, the likelihood of perception by the customer is examined. According to DIN EN ISO 9000: 2015-11, this is a new term for customer satisfaction that indicates the degree of fulfillment of customer expectations.
Here, too, numbers are assigned depending on the probability of taking notice:
- 100% = 4
- > 50% = 3
- <50% = 2
- <5% = 1
Customer relevance number (KRZ)
Similar to the FMEA analysis, the frequency of the error is multiplied by its significance and the probability of perception, which results in the customer relevance number, or KRZ for short. The larger this number, the more severe the impact on the customer. If there is an error with a high KRZ, it is assigned special importance and action is taken immediately, as it is very likely that it will be noticed and the customer will complain.
Classification in the product development process
According to the Pareto principle , 80% of the causes of errors arise in development and 20% in production. Product development is the phase in which features or defects are classified. An error collection card is often used to analyze the type and frequency of errors . This map can be used to determine the necessity to deal with the error. Based on the descending order of the frequencies of the individual errors, one can usually see that a few errors that occur with a high degree of probability make up a large proportion of the overall percentage of errors. By eliminating these causes of errors, the quality can be increased.
In order to keep the error elimination costs low, the production processes must be set up in such a way that the occurrence of critical errors is minimized. Only features of the product are considered that are of great importance, i.e. the z. B. impair customer satisfaction, endanger the safety of life or violate the legal regulations and thus justify a high expenditure.
The features can be classified using various evaluation criteria. It should be noted that the error elimination costs increase according to the "rule of ten". This means that the costs increase tenfold with each production stage, so they are many times higher during production than in the development stage.
The APQP methodology is used for quality planning in product and process development. Errors should be avoided through a uniform, product-related documentation structure. The relevant characteristics are planned, monitored and managed centrally.
Defect capture and analysis
In order to increase the quality by preventing the recurrence of an error, it is important to record errors and analyze the cause of the error. This is possible with the help of the static process control and the seven quality tools.
Seven quality tools
The seven quality tools were put together by Kaoru Ishikawa in 1943. They consist of several methods and serve problem-solving processes. This allows errors to be recorded and the causes of errors to be analyzed. The seven quality tools build on one another. They are also divided into two groups, failure detection and failure analysis techniques.
Statistical process control
With static process control , standardized processes are monitored and controlled. The quality features are measured regularly and the measurement results are transferred to quality control cards. This allows you to intervene in the production process when the values are above or below the limit values and prevent undesired developments. The quality control card has the advantage that possible trends can be recognized at an early stage and action can be taken immediately.
Defect detection techniques
Used to record the errors by graphically displaying the errors according to type and frequency:
Failure analysis techniques
These are used to determine the causes of the error:
- Pareto chart
- Correlation diagram ( scatter diagram ): graphic representation of characteristic values in a coordinate system for checking cause-effect relationships between the variables under consideration
- brainstorming
- Ishikawa diagram ( cause and effect diagram )
literature
- Martin Drobits: Bachelor thesis I. Methods and tools in quality management. Systematic approach to error classification, root cause analysis, error correction and prevention. Diplomica Verlag GmbH, Hamburg 2010
- Ralf Dillerup, Roman Stoi: corporate management. Management & Leadership. Strategies-tools-practice. 5th edition. Verlag Franz Vahlen GmbH, Munich 2016
- Martin Klein: Introduction to the DIN standards. BG Teubner, Stuttgart. Leipzig. Wiesbaden + Beuth Verlag, Berlin. Vienna. Zurich 2001
Individual evidence
- ↑ a b c Heinz Langmack: Error management. Behr's Verlag, Hamburg 2001
- ↑ a b c d e f Anni Koubek: Practice book ISO 9001: 2015. Understand and implement the new requirements. Carl Hanser Verlag, Munich 2015
- ↑ a b c Martin Drobits: Bachelor thesis II. Methods and tools in quality management. Systematic approach to error classification, root cause analysis, error correction and prevention. Diplomica Verlag GmbH, Hamburg 2010
- ↑ a b c d Gerhard Wolf. Marek Leszak: Defect classification for software. Guideline. BITKOM, Berlin-Mitte 2001
- ↑ Dealing with errors Website of the statistical offices of the federal and state governments. Common statistics portal. Retrieved June 20, 2018.
- ↑ a b c d Grit Reimann: Successful quality management according to DIN EN ISO 9001: 2015. Solutions for practical implementation. 3rd completely revised edition. Beuth Verlag GmbH, Berlin. Vienna. Zurich 2016
- ↑ a b c d e Walter Masing: Introduction to the quality theory Beuth Verlag GmbH, Berlin. Vienna. Zurich 2016
- ↑ Ron Eringa: Dealing with Bugs in an Agile Environment. Prowareness GmbH, Düsseldorf
- ↑ Classification of the bug priorities website of AugMenVis GmbH. Retrieved June 21, 2018.
- ^ Hans Dieter Seghezzi. Fritz Fahrni. Thomas Friedli: Integrated quality management. The St. Gallen approach. 4th revised edition. Carl Hanser Verlag, Munich 2013