Some standard content:
ICS 03.120.30
iiiKAa~cJouaKAa-
National Standard of the People's Republic of China
GB/T 17989.1—2020
Replaces (GB/T17989-2000
Control charts-Part 1:General Guide
Control charts-Part 1:General guidelines(IS07870-1;2014.MOD)
2020-03-06Released
State Administration for Market Regulation
National Standardization Administration
2020-10-01Implementation
iiiKAa~cJouaKAa-
Normative reference documents
Terms and definitions
Control charts
Statistical control of processes
Process acceptance
Natural drift process Pipeline
Determine the risk
Data collection design
Measurement control chart and counting control chart
6 Types of control charts
Control charts for determining process stability
Conventional control charts and some related control charts, 7.2
8 Acceptance control charts·
Acceptance control charts
Modified control charts (control charts with modified control limits. See GB/T17989.3) 8.3
9 Process adjustment·
References
GB/T 17989.1—2020
iiiKAa~cJouaKAa=
GB/T17989 "Control Chart" is planned to be divided into the following 9 parts: Part 1: General Guide:
Part 2: Conventional Control Chart;
-Part 3: Acceptance Control Chart;
-Part 1: Cumulative and Control Chart:
Part 5: Special Control Chart:
Part 6: Exponential Weighted Moving Average Control Chart;-Part 7: Multi-Child Control Chart;
-Part 8: Short Cycle and Small Batch Control Methods;-Part 9: White Correlation Process Control Chart
This part is the first part of GB/T17989. This part was drafted in accordance with the rules given in GB/T1.12009. 17989.1—2020
This part replaces GB/T17989—2000 "General Principles and Guidelines for Control Drawings". Compared with GB/T17989—2000, the main technical changes are as follows:
"Terms and Definitions" (see Chapter 3) has been added; "Symbols" (see Chapter 4) has been added;
"Concepts" (see Chapter 5) has been added;
"Economic Considerations" (see Chapter 13 of the 2000 edition) has been deleted. This part adopts the redrafting method to modify and adopt IS078701:2014 "Control Drawings Part 1: General Guidelines". The technical differences from IS078701:20 and their reasons are as follows: Regarding normative reference documents, this part has made technically different adjustments to adapt to my country's technical conditions. The adjustments are concentrated in Chapter 2 "Normative Reference Documents". The specific adjustments are as follows:, replace ISO35342 with GB/T3358.2, which is equivalent to the international standard (see Chapter 3);? Added reference to G3/T17989.2 (see 5.1);? Added reference to G3/T17989.3 (see 5.3); ● Added reference to GB/T17989.4 (see 7.2.3.2). This part has been edited as follows:
Adjust the order of references. Add references IS078705.IS078706, ISO225141, IS225147 This part is proposed and managed by the National Technical Committee for Standardization of Statistical Methods (SAC/TC21). The drafting units of this part are: Huzhou Mingfeng Enterprise Management Consulting Co., Ltd., Xiamen Minghongge Crafts Co., Ltd., China National Institute of Standardization, Tsinghua University, Huzhou Hongtuo Enterprise Information Service Co., Ltd., Piaoyang Market Comprehensive Inspection and Testing Center: The main drafters of this part are: Ding Lihui, Zhang Fan, Zhao Jing, Wai Jing, Tang Jiaotiao, Yu, Qian Xinhui, Jiang Wenhua. The previous versions of the standards replaced by this part are: -GB/T 17989-2000.
iiiKAa~cJouaKAa
GB/T 17989.1—2020
Each production, service or management process has a certain degree of variation due to the influence of various factors. Therefore, the results observed in the process are not constant. Studying the variability of the process will help to understand its characteristics and provide a basis for taking appropriate measures. Control charts are a basic tool for statistical process control (SPC): they provide a simple graphical method that can be used to: determine whether the process is stable, that is, whether the process is operating in a system with individual random causes. The variation a)
in this case is called inherent variation, and the process is also called "under control"; estimate the degree of inherent variation of the process:
Compare the sample information representing the current state of the process with the control limits that reflect this variation to determine whether the process variation remains stable or changes:
Identify, investigate and possibly reduce/eliminate the influence of special causes of variation that may cause the process to reach an unacceptable level of d
;
Through the identification of trends, processes, cycles Various variable modes, such as: e) Determine whether the process behaves predictably and stably, so as to evaluate whether the process meets the specification: Determine whether the measured characteristics in the process meet the expected process capability required to meet the product or service requirements; g) h) When using statistical models for prediction, provide a basis for process adjustment; ) Help evaluate the performance of the measurement system, the main advantage of the control chart is that it is easy to draw and use. It provides online indicators of process behavior for production or service operators, engineers, managers and operators. However, in order to make the control chart a reliable and efficient process status indicator, in the design stage, attention should be paid to the process under study, select the appropriate control chart type, and determine the correct sampling plan. This part of GB/T17989 gives the factory a general concept of successfully designing a control chart. 1 Scope
iiiKAa~cJouaKAa=
Control chart
Part 1: General Guide
GB/T 17989.1—2020
This part of GB/T17989 gives the key elements and basic principles of control chart methods, and defines various types of control charts (including control charts related to conventional control charts, control charts that emphasize process acceptance or online process adjustment, and special control charts): This part introduces the basic principles and concepts of the factory, and gives examples to illustrate the relationship between various control chart methods to help select the most suitable control chart method in specific situations. This part does not give the statistical control methods used in control charts. This content is explained in detail in other parts of GB/T17989.
2 Normative referenced documents
The following documents are indispensable for the application of this document: For any referenced document with a date, only the version with the date applies to this document. For all referenced documents without IⅡ period, their latest versions (including all amendments) apply to this document, GB/T3358.2 Statistical Vocabulary and Symbols Part 2: Applied Statistics (GB/T3358.22009.IS3534-2:2006:IDT)
GB3/T【7989.2 Control Charts
Part 2: Conventional Control Charts ((B/T17989.2—2020,IS07870-2:2013,MOD)G3/T17989.3 Control Charts||tt| | Part 8: Acceptance Control Chart (3/T17989.32020, ISO7870-3: 2012.M0D) GB/T17989.4 Control Chart Part 4: Cumulative and Control Chart (GB/T17989.4—2020.ISO78704: 2011, MOD)
3 Terms and Definitions
The terms and definitions defined in GB/T3358.2 and the following terms and definitions apply to this document. For the purpose of wide use, some terms and definitions in GB/T3358.2 are repeated below: 3.1
Control chartcontrol charl
A chart that plots a sequence of sample statistics in a specific order to monitor, control and reduce process variation. Note 1: A specific order is a time sequence or sample acquisition sequence. Note 2: Control charts are most effective when used to monitor characteristics of final products or services. [GB/T 3358.22009, definition 2.3.1] 3.2
Control limitscontrollimits
A statistical value used to determine the expected stability of a characteristic. Note 1: A control chart usually has one or two control limits. Note 2: "Stability" does not refer only to the controlled process, it can also refer to the stability of the monthly standard value. 3.3 Conventional control chart Shewhart control chart Shewhart control chart is mainly used to determine whether the variation is caused by random causes or special causes from a graphical perspective. It uses the control chart (3.1) with conventional control limits (3.4), [G13/T 3358.2—2009. Definition 2.3.2 3.4 Conventional control limits Shewhart control chart is a statistical method, and the control limits (3.2) are determined based on the process changes caused only by random causes. 3.5 Acceptance control chart acceptance control chart chart
Control chart mainly used to determine whether the plotted points can be formed within the tolerance (3.1) [GB/T3358.22009, definition 2.3.3]
Process adjust control chart
Use the process prediction model to estimate the future trend without adjustment. Or determine the adjustment amount to make the system deviation within the acceptable range of control chart (3.1)
[G13/T 3358.2—2009-Definition 2.3.4]3.7
Measurement control chart
Variable conirol chart
The statistic used for plotting points is a conventional control chart of continuous scale (3.3). [GB/T3358.22009, definition 2.3.6]
Attribute control chart
Count control chart
Conventional control chart (3.3) where the statistic used for plotting points is countable or categorical variable LGB/T3358.2—2009, definition 2.3.7
cchart
Count control chart countcontrol chart Used to - quantify the number of specific types of nonconformities in a sample, evaluate and monitor the process level of the count control chart (3.8): Note 1: The count is the total number of events of a particular type, such as the number of missing people, the number of high sales, etc. In the field of quality, it is usually the number of nonconformities in a sample of a fixed sample size or a fixed amount of material, such as the number of defects in 100m of fabric, the number of errors in 100 invoices. Note 2: Renumbered G3/T3358.2—2009, definition 2.3.8, 3.10bzxz.net
uchart
Count per unit control chart An attribute control chart (3.8) that uses the average number of nonconformities of a particular type in a sample to assess and monitor the process level. Note 1: The count is the average number of events of a particular type. In the field of quality, it is usually the average number of nonconformities in a sample, such as the average number of defects per square meter of fabric, the average number of errors per invoice. Note 2: Rewrite GB/T3358.22009. Definition 2.3.93.11
np control chartnpchari
piece control chartnumberor categorized unitscontrol chartAn attribute control chart (3.8) that uses the number of units of a specified category in a sample of fixed group size to assess and monitor the process level. Note 1: In the field of quality control, it is often classified according to nonconforming products. In this case, it is called a nonconforming product number control chart. Note 2: Rewrite GB/T335S.22009-Definition 2.3.102
iiiKAa~cJouaKAa=
p control chartpchart
piece ratio control chart
proportion or percent calegorized comrol chari percentage classified unit product control chartpcrcntcatcgorizcdunitscontrolchartGB/T 17989.1—2020
counting control chart (3.8) that uses the proportion or percentage of units in a certain amount of sample belonging to a specified category to evaluate and monitor the process level
Note 1: In the field of quality control, it is often classified according to nonconforming products. In this case, it is called a nonconforming product rate control chart. Note 2: The chart is particularly suitable for situations where the group size changes. Note 3: It can be captured by proportion or percentage. Note 4: Rewrite GB/T 3358.2-2009. Definition 2.3.11. 3.13
standardized p chart
Standardized P control chart
An counting control chart that expresses the proportion of a specified category as a standardized normal variable (3.8). 3.14
X control chart
X har control chart
average control chart
average control chart
A variable control chart that uses the group mean to evaluate and monitor the process level (3.7). Note: Rewrite GB/T3358.22009. Definition 2.3.12 3.15
median control chart
median control chart
measurement control chart that uses the median of the subgroup to evaluate and monitor the process level (3.7), Note: Rewrite G13/T3358.2—2009+Definition 2.3.13, 3.16
moving average control chart movingayeragecontrolchart control chart that uses the arithmetic mean of every n consecutive observations to evaluate and monitor the process level (3.1). Note 1: This type of chart is particularly useful when there is only one observation per subgroup. For example, the process characteristic is temperature, pressure, time, etc. Note 2: The earliest value among the current observation and the most recent observation is replaced by the current observation. Note 3: One disadvantage of this control chart is that it does not weight the successive observations. Note 4: Rewrite GB/T3358.2200. Definition 2.3.11.3.17
日individual control chart
X control chart
X control chart
Variation control chart that uses a single observation from a sample to evaluate and monitor process performance (3.7) Note 1: The individual control chart is generally used in pairs with the moving range control chart (usually 1-2). Note 2: The individual control chart cannot use the average method to reduce random variation, nor can it use the central limit theorem. Note 3: The individual control chart is represented by the symbol.
Note 4: For single value charts, the symbol R represents the moving difference, that is, the absolute value of the difference between two successive values, such as -1:, -1, etc. Note 5: Rewrite GB/T3358.22009. Definition 2.3.15. 3.18
Cumulative sum control chartCUSUM chartCUSCMchar1
A control chart that plots the deviations of successive sample statistics from a reference value to detect drifts from the mean level of the statistics at the plotted points (3.1).
Note 1: The ordinate of each point is the algebraic sum of the ordinate of the previous point and the deviation of the current point from the reference value, standard value or selected value3
iiiKAa~cJouaKAa=
GB/T17989.1—2020
Note 2: When the reference value is equal to the total average value, the change in the mean level can be best judged. Note 3: This control chart can be used for control, diagnosis and prediction. Note 4: When the cumulative control chart is excavated, a template (for example, a V-shaped plate) can be superimposed on the existing chart to make a graphical interpretation: When the cumulative sum curve reaches the boundary of the V-shaped template or intersects with it, an alarm signal is issued [G13/T 3358.22009, definition 2.3.5] 3.19
EWMA control chartEWMAchart
ExponentiallyweightedmovingaveragecontrolchartA variable control chart (3.7) that uses an exponential sliding weighted average to evaluate and monitor the process level. [GB/T 3358.22009, definition 2.3.16
Z chartZ chart
A variable control chart (3.7) that uses the standardized end-state variables of subgroups to evaluate the process. 3.21
group control chart for averages
Samples from multiple sources form subgroups, and the maximum and minimum values of the sample means of each source in the subgroup are used to evaluate and monitor the measurement control chart of the process level (3.7)
Group control chart for rangesSamples from multiple sources form subgroups, and the maximum value of the sample range of each source in the subgroup is used to evaluate and monitor the measurement control chart of process variation (3.7).
Extreme control chart high-lowcontrolchartA measurement control chart that evaluates and monitors the measurement control chart of the process level based on the maximum and minimum values of the subgroup observations (3.7). 3.24
Trend control chartA control chart that uses the deviation of the group mean from the expected value of the process level to evaluate and monitor the process level (3.1). Note 1: Trends can be determined by empirical methods or convergence methods. Note 2: Plot the observations in time sequence. After eliminating random variation and periodic effects, the trend control chart shows the trend of changes in the process level. Note 3: Modified G3/T3358.2—2009. Definition 2.3.173.25
R chart Rchart
Range control chart rangecontrol chart A measurement control chart that uses subgroup ranges to evaluate and monitor process variation (3.7) Note 1: The subgroup range value is represented by a scale symbol, that is, the maximum and minimum measured values in the subgroup. Note 2: The mean of the subgroup slope difference is represented by the symbol R. Note 3: Rewrite GB/T3358.22009+Definition 2.3.183.26
s control chartschart
Standard deviation control chartstandlarddeviationcontrolchartA measurement control chart that uses the standard deviation of the group to evaluate and monitor process variation (3.7). Note 1: The standard deviation of the group is represented by the symbol,. Note 2: The mean of the subgroup standard deviation is represented by the symbol 5. Note 3: Rewrite GB/T3358.22009.Definition 2.3.193.27
iiiKAa~cJouaKAa=
Moving range control chartmovingrangeconirolchartA measurement control chart that uses the difference of n consecutive observations to evaluate and monitor process variation (3.7). Note 1: The previous observation replaces the most recent n1 observations. Note 2: Rewrite GB/T3358.22009. Definition 2.3.20 3.28
control chart for coefficient of variation Coefficient of variation control chart
Quantitative control chart for evaluating and monitoring variation based on the coefficient of variation of subgroups (3.7) 3.29
Multivariate control chart
multivariate control shart
GB/T 17989.1—2020
Control chart obtained by synthesizing a sample statistic using two or more related variables in a subgroup (3.1). Note 1: Multivariate control chart is also called multivariate control chart. Note 2: Rewritten from GB/T 3358.2—2009. Definition 2.3.21.3.30
Multiple characteristic control chart An attribute control chart based on multiple characteristics to evaluate and monitor process levels (3.8). Note: Rewritten from GB/T 3358.2—2009. Definition 2.3.22, 3.31
demerit control chart
defect control chart
quality score chart
quality score chart
Multidimensional characteristic control chart (3.30) that assigns different weights to each defect event (or quality score) according to the perceived significance. [G13/T 3358.22009, definition 2.3.233.32
process adjustment
processadjustment
action to reduce the deviation of output characteristics from the standard estimate through feedforward control and/or feedback control, note: real-time monitoring is carried out to determine whether the process and process adjustment system are in a state of statistical control. [(G13/T 3358.2—2009. definition 2.3.213.33
control variable
Control variable
A variable in the process used as an alarm signal to change the output of the process. Note 1: The signal can be triggered by a measurable process change. Note 2: Rewrite GB/T3358.2-2009, definition 2.3, 27. 3.34
Autocorrelation
Internal correlation between a series of observations obtained in time sequence. LGB/T3358.2-2009. Definition 2.3.283.35| |tt||Special causespecialcaust
(process variation) Causes of process variation other than the inherent variation of the process. Note 1: "Special cause" is sometimes also called "identifiable cause", but there is a difference between the two. Only when non-special causes are clearly identified can they be considered as identifiable causes.
Note 2: Special causes are caused by some special circumstances that do not occur very often. Therefore, in a process affected by special causes, the magnitude of the variation changes over time and is unpredictable [(G13/T 3358,2—2009. Definition 2.2.1]5
iiiKAa~cJouaKAa=
GB/T 17989.1—2020
randomcause
random cause
commoncausc
general cause
chance cause catuse
(Process variation) The cause of the inherent variation in the process over time. Note 1: If the process is only affected by random causes, the variation is statistically predictable. Note 2: Reducing the number of random causes can lead to process improvement. However, the identification, reduction and elimination of random causes must be subject to a cost-benefit analysis from the perspective of technical feasibility and economy.
LGB/T3358.2—2009. Support 2.2.5
4Symbols
The following symbols apply to this document.
nSubgroup size
The proportion of unit products of a certain characteristic
Red range
RThe average of the group ranges Value
Subgroup standard deviation
Average of subgroup standard deviation
2Single value
Subgroup mean
5Concepts
5.1 Control chart
Control chart is a graphical representation of process data, which can be used to visually evaluate the variation of the process. Within a given interval, the red man is specified to obtain quality characteristic data, thereby determining the characteristic value or characteristic of the product: These data are usually used to obtain appropriate statistics and plot these statistics on the control chart: A typical control chart contains a center line and control limits on both sides of the center line: Among them, the center line reflects the central level of the expected variation of the statistic. If the process is controlled.The statistic will randomly fall within the area determined by the two control limits. The two control limits are used to determine whether the process is in a controlled state. The control limits define an interval, and the width of the interval is determined to some extent by the inherent variation of the process. If the statistics of the points plotted in the control chart are within this area, indicating that the process is in a state of statistical control, then the process can continue to run with the current settings. However, if the statistics of the points plotted in the control chart are outside this area, it means that the process may be "out of control." When the control chart shows an "out of control" signal, it indicates that there may be special causes that cause process variation, and the process needs to be corrected by the most necessary measures:
The measures that can be taken include:
a) Conduct an investigation to determine the source of the special cause, with the aim of eliminating, correcting or reducing the impact of the cause: b) Perform c) Continue the process based on the risk assessment: d) Stop the process or take corrective measures e) If the special cause shows positive characteristics (such as process improvement), keep the special cause and keep it as long as possible. Sometimes there is a second set of control limits on the control chart called "warning limits". If the warning limits are plotted on the chart but not beyond the control limits, it indicates that there is a suspected cause affecting the process. It should be noted that no "measures" need to be taken on the process at this time. You can shorten the sampling interval with the next two red and/or increase the next sample to help determine whether the process has changed: When the control chart includes warning limits, the control limits are sometimes also called "action limits". GB/T 17989.1—2020 There are other forms of rules for judging the state of the process, such as points within the control limits showing an abnormal arrangement pattern. These rules are usually called "decision criteria", see GB/T17989.2. When the control chart is used for process acceptance, the acceptance limit can be used as a criterion for judging whether the process is acceptable: see 5.3, 5.2 Statistical control of the process
Control charts are often used to judge the stability of the process. If it is only affected by random (or general or accidental) causes, that is, if there are no special, unexpected or special (or specified) causes affecting the system, the process is considered to be in a "statistically controlled state". These special causes may affect the level of process operation. Or the degree of variability of the process level, or both. The variation caused by random or accidental causes occurs randomly and often conforms to certain statistical laws. In essence, when the process is under statistical control, the operation of the process can be stably predicted, and when special (or specific) causes affect the system, the process depends on the results of these causes: If the process is found not to be in a state of statistical control, it is called "out of control", and ten preventive measures need to be taken at this time to restore the process to a state of statistical control. For some economic or natural phenomena, there may be no known preventive measures, and control charts are used only to identify out-of-control conditions.
5.3 Process Acceptance
In addition to monitoring process stability, control charts can also be used to determine whether a process is acceptable. When a process is in a state of statistical control, its decision risk is controlled, and control charts can determine whether the process output meets the requirements of the product or service. The most effective is for tolerance ranges. When the process variation is small, even if the process level temporarily drifts to a certain "out-of-control state", the process can still meet product and service requirements. At this time, the control chart is used to monitor whether the process is in an acceptable state, regardless of the dynamic changes in the process level. In this case, the acceptance control chart in GB/17989.3 should be used. 5.4 Management of Natural Drift Processes
When a disturbance that cannot be eliminated causes the process level to drift, such as the concentration of a specific chemical in a batch, there may be a compensating variable that can be used to adjust the process level. In this case, a control chart can be specially designed to explain when and how to adjust to compensate for the influence of the disturbance. This type of control tends to significantly reduce the variability of the process without causing the process to be over-adjusted. 5.5 Judgment Risk
When a set of judgment criteria and a finite sample of data points are used to judge the control status of a process, two types of errors may occur: Type I error (Type II error) is the judgment that the process is not in statistical control and appropriate measures should be taken, when in fact the process is only affected by random causes. Therefore, the process will be incorrectly judged as "out of control". The risk of this error is called "risk". When there are special causes affecting the process, but the collected data cannot judge that the process is "out of control", a Type II error (Type II error) will occur. At this time, the process will be incorrectly judged as "in statistical control". The risk of this error is called "beta risk". These two risks can be controlled by appropriately selecting control limits, judgment criteria, and subgroup sizes. 5.6 Data Collection Design
5.6.1 Overview
The most important factors in data collection are the selection of characteristics and the identification of the location or stage of control. The data collection method helps to distinguish between random causes and special causes, which is very important for the effective operation of control charts. Based on the understanding of the process and the nature of the data collected, consider the method of defining samples or subgroups, the appropriate subgroup size, and the frequency of data collection! 5.6.2 Feature Selection
First, the feature to be controlled by the process should be selected. The following factors should be considered. First, the feature is a key feature that can reflect the state of the process. Second, the characteristic is related to the quality characteristics of the product: Table 1 gives an example of selecting characteristics based on the results of FMEA and process analysis. The more important the product characteristic is, the earlier it should be controlled in the process. In Table 1, the rolling pressure and forming torque applied to the hinge can be used as candidates for control chart characteristics: Table 1 Characteristic selection
Importance ranking (according to FMEA table)
9-10, Chain characteristics affecting safety
5~~8, Important characteristics
2~4, General characteristics
5.6.3 Measurement process evaluation
Product characteristic examples
Insulator originality
Motion resistance
Product component characteristic examples
Screw pitch diameter||tt ||Surface texture
Example of process parameters
Rolling force
Stress torque applied to the chain
Operating parameters
Before implementing any type of process control, it is important to ensure the validity of the measurement process: the variation caused by the measurement is small enough to detect the variation of the process characteristics, so the measurement variation (see 1S022511-7) must be estimated. In this case, the measurement method (including method, instrument, etc.) must be selected based on the specification or process variation. 5.6.4 Subgroup selection
Subgroups are samples of unit products collected from the process in a certain way. Statistics such as the number of non-conforming products, mean value, range, etc. can be calculated based on the characteristic values of these samples. And plot these statistics on the control chart. The selection of reasonable samples or subgroups should ensure that all subgroups are homogeneous. Within reasonable subgroups, it is assumed that the variation is caused by random causes. These causes are caused by the inherent variation in the process over time. Reasonable selection of subgroups can help to discover special causes of inter-group variation. Short-term variation is measured by a series of reasonable and homogeneous within-group variations, and the location of control limits can be determined by the measurement, while long-term variation is usually evaluated based on inter-group variation. It is usually constructed in a time sequence because special causes may appear over time. Other factors can also be used to construct groups. For example, changes in operators, changes in equipment, or changes in suppliers, subgroups can be defined according to different operators, equipment or suppliers. If you want To make sense of rational subgroups, care should be taken to include all the effects of normal random causes. For example, a test dummy is used to test a material placed in it, and the series of repeated test values obtained may not include the effects of the material placement process or the sampling process: if these effects are inherent in the normal test environment, these repeated test data will unrealistically underestimate the inherent variation of the measurement process. Therefore, almost all actual measurements obtained from this process will show "out of control". 5.6.5 Subgroup size
Choose the subgroup size to balance the ability to detect small process shifts and the risk of failing to find special causes. Although larger subgroup sizes are more expensive, they provide a more accurate basis for process assessment. Therefore, monitoring efficiency is higher. If the sample size is too large: during the sample collection period, special causes are more likely to appear, increasing the variation of the sample itself; therefore, the control limits may be over-expanded. It may be difficult to find special causes.
When dealing with count data, since count data contains less information than measurement data, the subgroup size required to detect changes in the defective rate of the process is usually much larger than the subgroup size using measurement data. In some cases.Subgrouping is not feasible or meaningful, and only information can be collected on a single unit of product. Therefore, subgroups are inherently small. 1: If destructive testing is required, or sampling costs are high, or repeated measurements of a process (continuous or batch) differ only due to instrument or analytical errors, this is also a situation where subgrouping cannot be performed. 5.6.6 Sampling Frequency
The sampling frequency depends on the magnitude of the process excursion and the cost of running the process in a state of statistical out-of-control. The smaller the excursion, the more likely it is to be detected.
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.