Zhao 1. Xu 1, Yao 2, Guo 2
(1. Zhejiang Entry-Exit Inspection and Quarantine Bureau, Hangzhou 310012; 2. Zhejiang Lide Product Technology Co., Ltd., Hangzhou 3 10005)
Based on the principle of "process quality control", the models based on different quality control are established by analyzing the test data of ecological textiles in the laboratory.
The early warning mode of the analysis object and a variety of different "quality control chart models" can realize the analysis and monitoring of the detection indicators, and also provide for enterprises and government departments.
Corresponding quality trend analysis information. The establishment of this method can achieve the purpose of rapid and effective early warning for relevant departments under the condition of using existing detection resources.
Keywords: eco-textiles; Detection; Early warning method; quality management
China library classification number: TSl0 1.9 document identification number: a.
Introduction to 0
Early warning analysis has been widely used in risk management and decision-making management in all walks of life. Establishing an appropriate early warning analysis system is helpful to effective prevention.
And reduce the losses caused by harmful substances, judge the possible situation of harmful substances in advance, and take corresponding management measures or adjust decisions in time.
Avoid risks. In the field of detection, through the analysis of harmful substances detection data in the laboratory, an early warning analysis method is established to realize the existing resource column.
In this case, we should strengthen inspection and control in a targeted manner, improve work efficiency, and provide corresponding quality monitoring information and trends for enterprises and government departments.
Analysis information' 1 |.
According to the working characteristics of the testing laboratory, the author established the testing and early warning analysis methods of ecological textiles from three angles.
A) The objective is to monitor the quality dynamics of import and export enterprises, and its purpose is to check the unqualified rate or inspection situation of import and export enterprises in a certain inspection period.
The analysis of output will send an early warning signal to enterprises with abnormal quality fluctuation.
B) The objective is to monitor some import and export textiles, and its purpose is to analyze the test data of some import and export textiles' ecological indicators.
For specific products with abnormal fluctuations in indicators, strengthen testing and realize effective quality monitoring.
C) For the purpose of monitoring a spot check project, its purpose is to analyze and compare the data of an ecological detection index in a certain detection period.
Send an early warning signal to the detection items with abnormal fluctuation of unqualified rate or detection rate, and remind relevant personnel to pay attention to the key spot check of this item, so as to realize the right inspection
Quality control of specific detection indexes.
No matter what kind of monitoring object, the generation of its early warning signal is based on the statistical analysis of the detection data, so a reasonable number is adopted.
Scientific analysis model is an important prerequisite for establishing early warning analysis method.
Through the analysis of a large number of data, it is shown that in general, the inspection failure rate is normal when the inspection units or inspection varieties or inspection items are limited.
Or the detection rate or over-standard rate always fluctuates in a certain range, and its value obeys statistical law and normal distribution in a large number of data. Therefore, before
By the way, we can use the principle of control chart to establish different mathematical models, make statistical analysis, and issue relevant early warning information by judging abnormal situations.
Control chart is a kind of graph, which gives the sample sequence information representing the current state of the process, and compares these information with the information established after considering the internal changes of the process.
Control limit [2].
1 Alert based on specific inspection unit (or specific inspection product) as the analysis object.
Usually, the products produced by an inspection unit can be roughly classified into one category. For the evaluation of quality characteristics of inspection lots, only combinations can be used.
There are two kinds of results: unqualified and unqualified. Therefore, the corresponding quality control chart can be established by statistical analysis of the test data of an inspection unit over the years.
That is, within a certain period of time or a certain sample size, within the range of +3a (U is the average value of unqualified number or unqualified rate, and o- is the standard deviation).
Fluctuating unqualified numbers are normal and are caused by accidental factors. If it exceeds this limit, it means that the quality fluctuation is abnormal and needs attention.
Judging the quantity of unqualified products or the determination of unqualified rate according to the detected product standards or indicators. The number of unqualified tests here
According to the calculation of natural inspection lot, no matter how many unqualified items appear in an inspection lot, they are calculated as 1 unqualified item, that is, an unqualified inspection lot.
Batch. The unqualified rate refers to the percentage of unqualified number in the number of elm applications during this period. The division of time period and sample size depends on the inspection.
Inspection quantity of the company. For units that are continuously and uniformly inspected, they can be divided by time period, and for units with large changes in inspection quantity, they can be divided by.
The sample size (inspection lot number) of, when compared with historical data, should adopt the same sample division method. This control chart belongs to counting numerical control.
Drawing, commonly used are unqualified numerical control diagram (rip diagram) and unqualified rate control diagram (evening diagram) [2].
1. 1 Establishment of models of unqualified NC chart (np chart) and unqualified rate control chart (household chart)
For these two types of early warning, K samples (more than 25) are usually taken under the same sub-sample size n, and a set is set by analyzing the fluctuation of unqualified numbers.
Center line (CL) and upper control line (UCI). ) and warning line (WCL) to determine whether the system is in a stable state, assuming the number of inspection lots per sample.
Similarly, for all excavations, P is the average of sample failure rate, and curse P is the average of sample failure rate. According to the national standard routine control chart, the following mathematics is established.
Model L 2|.
1. 1. 1 mathematical model of unqualified numerical control diagram (np diagram)
K
∑ (silicon) I
Average silicon of unqualified samples: np-F-( 1)
Central control line (Z: Cl-NP (2)
Upper control line UCL(3):UCL-NP+3. g- five p( 1-p) (3)
Warning line WCL(2 stare): WCL-NP+2 ~/Si (1-p) (4)
Mathematical model of 1. 1.z unqualified rate control chart (spectrogram)
K
，∑(rip)i
Average sample failure rate p: p-%-(5)
Σ curse I
i=l
Central control line cl: cl-p (6)
/2'●=———。 =2_
Upper control line UCL(3): UCL 1 stone +3√ Congshi I fish 2 (7)
F=—? =_———=_
Warning line WCL (2D): WCL = ≯+2 √ Congshi I Line (8)
1.2 Generation and cancellation of early warning
After the CL line, UCL line and WCL line are determined, the corresponding "sample-unqualified number" control chart (np chart) and "sample-unqualified number" are established.
Rate control chart (P chart), when the following situations occur, the system will generate an early warning signal.
UCI has an idea. (30) Off-line;
B) Two of the three consecutive points fall outside the WCL line (20);
C) Four of the five consecutive points are on the UCL line and WCI. Between lines (2a-3a);
D) It keeps rising at 6 o'clock.
When one of the above situations no longer occurs in the system, the early warning signal is released and the normal sampling state is restored.
Application example of 1.3
Taking "sample unqualified NC diagram (np diagram)" as an example, the application of this early warning method in the actual detection process is illustrated.
According to the statistics of historical data of an inspection unit in a certain period of time, the unqualified number of each sub-sample is shown in table 1 (all sub-samples are 50 inspections).
Table 1 A Statistics of the number of historical unqualified inspection units
Sample serial number Unqualified sample serial number Unqualified sample serial number Unqualified sample serial number
1 2 1 1 2 2 1 2
2 2 12 2 22 3
3 1 13 3 23 2
4 3 14 1 24 4
5 1 15 3 25 3
6 2 16 3 Total 59
7 3 17 l average 2.36
8 2 18 2
9 2 19 3 error coding rate is 4.72
10 3 20 4
According to the formula in 1. 1.2, calculate the central control line, upper control line and warning line respectively, and the results are as follows:
Central control line: CL = 2.36, average unqualified rate: P = 4.72.
Upper control line UCL: UCL = 6.86 Warning line WCL: WCL = 5.36.
Make an unqualified NC diagram for the sample 1 (line p) (figure1).
Sample serial number
Figure 1 A sample of unqualified NC drawing based on historical data
According to the established UCL line, CL line and wCL line, the number of unqualified products in this enterprise in the recent period is counted.
Methods: Fill in the control chart, and click the connection, we can see that the unqualified rate of the detection index of the enterprise has no abnormal fluctuation compared with the history.
Good dynamic and process control, no need to generate early warning. In the actual test work, the test can be carried out according to the normal sampling ratio.
The early warning analysis methods established by the above two control charts are mainly suitable for testing large-batch enterprises and varieties, as well as production varieties and processes.
The level and process should be relatively stable. However, for small batch, because the number of sub-samples can not reach the required number, it obeys the samples required by normal distribution.
Theoretically, this condition is difficult to be strictly applied. However, as a means to monitor the quality of an enterprise or a certain kind of products, at the beginning,
By appropriately reducing the statistics of sub-samples and the collection of total samples, and then through the accumulation and sum of detection data
The supplement and update of time make the control chart more reasonable and feasible.
2 Early warning takes detection indicators as the analysis object.
This kind of early warning is the most important early warning analysis method in the laboratory, and the data is counted and analyzed according to different testing items.
There are several common early warning methods for different indicators.
2. 1 single indicator detection unqualified warning
2. 1. 1 Generation and application of early warning
This kind of early warning is applicable to the monitoring of an important sensitive indicator and the monitoring of newly added and prohibited dangerous indicators in a specific period. As for a certain period of time
Once the quality problem of an index reflected by foreign customers or the new banned substances in domestic and foreign technical regulations are found in the testing process,
If it is unqualified, it will generate an early warning signal to remind relevant personnel to pay attention and strengthen key testing.
2. 1.2 Determination of early warning threshold
For specific detection indicators, the determination of early warning threshold is mainly based on the maximum limit or minimum detection limit of harmful substances stipulated by national technical regulations.
2. 1.3 Early warning cancelled
For single indicators with unqualified early warning, the early warning can be cancelled under the following circumstances.
A) For an inspection unit or a certain variety, the monitoring indicators have not failed for 25 consecutive batches;
B) 65,438+000 data have been continuously detected by monitoring indicators, and none of them is unqualified.
2.2 Early warning of abnormal fluctuation of unqualified rate detected by specific indicators
In actual detection, the detection results of most indicators are less than the maximum limit value (MRL) and greater than the detection lower limit value. In this case, by
By analyzing the change trend of the test value, we can prompt a certain development trend of the index, so as to take measures in advance and strengthen the monitoring of the index. use
Mean-standard deviation control chart (X-S chart) can accurately identify whether the system is in a stable state [2].
2.2. 1 Establishment of mathematical model
The historical detection data are grouped reasonably. Each group of data is set to dig (usually 4 ~ 5 spells) and the data group is K (avoid ≥25), with samples respectively.
Taking this number as the abscissa and the mean or standard deviation of this group of test data as the ordinate, the "mean control chart" and "standard deviation control chart" are established respectively.
Calculate the average value of each sample and standard parameter I.
Zi- 13 ≥: zif (9)
Daozai
Si-√i and √-i)2
( 10)i= 1(xo
Where: Xij is the j-th value of the i-th sample; I is the average of group I; Played as the average of subgroup average; K is the number of samples, which is generally not counted.
The standard deviation of sliding to the ith sample is less than 25.
Calculate the average and standard deviation of k samples:
Eleven {≥: {≥:Xi (1 1+0)
Nickel -T- 1
J- chamber si ⅲ,
For the mean control chart:
Central control line cl: cl-j (13)
Upper control line UCL: UCL = X+A35 (14)
Among them, a. As a control factor, it can be directly obtained by looking up GB/T 409 1 standard.
At point a 4, A3- 1.682; when cursing a 5, A3= 1-427.
For standard deviation control chart
Central control line cl: cl-s (15)
Upper control line UCL: UCL-B4 south (14)
Similarly, B. is the controlling factor of S, which can also be obtained by directly looking up GB/T 409 1 standard.
When, 2-4, B4-2.266, when n=5, B4-2.089.
2.2.2 Generation and cancellation of early warning
Generally speaking, the "mean control chart" reflects the fluctuation concentration trend of the sample group, and the "standard deviation control chart" reflects the measurement of the sample.
Variation of deviation between test values. Therefore, from the "mean control chart", we can see the overall index level of the sample during this period, when the data points exceed
When crossing the UCL line, it shows that the content of harmful substances in the sample shows a great upward trend during this period; When in the standard deviation control chart
When the data points exceed the UCL line of the standard deviation, it shows that each sample in the sample changes greatly during this time period and has a high dispersion [3]. For these two feelings
Forms, all need to attract the attention of testers. When the data points in both control charts exceed the UCL line, it means that each sample in this time period has two fingers.
The standard value is on the high side and fluctuates greatly. Therefore, it is necessary to send out early warning signals, strengthen detection and reduce detection risks.
After strengthening sampling inspection for a period of time, when the average and standard deviation data points of the latest sample group are lower than UCL line for three consecutive points, the pre-inspection can be cancelled.
If it is unqualified, it will generate an early warning signal to remind relevant personnel to pay attention and strengthen key testing.
2. 1.2 Determination of early warning threshold
For specific detection indicators, the determination of early warning threshold is mainly based on the maximum limit or minimum detection limit of harmful substances stipulated by national technical regulations.
2. 1.3 Early warning cancelled
For single indicators with unqualified early warning, the early warning can be cancelled under the following circumstances.
A) For an inspection unit or a certain variety, the monitoring indicators have not failed for 25 consecutive batches;
B) 65,438+000 data have been continuously detected by monitoring indicators, and none of them is unqualified.
2.2 Early warning of abnormal fluctuation of unqualified rate detected by specific indicators
In actual detection, the detection results of most indicators are less than the maximum limit value (MRL) and greater than the detection lower limit value. In this case, by
By analyzing the change trend of the test value, we can prompt a certain development trend of the index, so as to take measures in advance and strengthen the monitoring of the index. use
Mean-standard deviation control chart (X-S chart) can accurately identify whether the system is in a stable state [2].
2.2. 1 Establishment of mathematical model
The historical detection data are grouped reasonably. Each group of data is set to dig (usually 4 ~ 5 spells) and the data group is K (avoid ≥25), with samples respectively.
Taking this number as the abscissa and the mean or standard deviation of this group of test data as the ordinate, the "mean control chart" and "standard deviation control chart" are established respectively.
Calculate the average value of each sample and standard parameter I.
Zi- 13 ≥: zif (9)
Daozai
Si-√i and √-i)2
( 10)i= 1(xo
Where: Xij is the j-th value of the i-th sample; I is the average of group I; Played as the average of subgroup average; K is the number of samples, which is generally not counted.
The standard deviation of sliding to the ith sample is less than 25.
Calculate the average and standard deviation of k samples:
Eleven {≥: {≥:Xi (1 1+0)
Nickel -T- 1
J- chamber si ⅲ,
For the mean control chart:
Central control line cl: cl-j (13)
Upper control line UCL: UCL = X+A35 (14)
Among them, a. As a control factor, it can be directly obtained by looking up GB/T 409 1 standard.
At point a 4, A3- 1.682; when cursing a 5, A3= 1-427.
For standard deviation control chart
Central control line cl: cl-s (15)
Upper control line UCL: UCL-B4 south (14)
Similarly, B. is the controlling factor of S, which can also be obtained by directly looking up GB/T 409 1 standard.
When, 2-4, B4-2.266, when n=5, B4-2.089.
2.2.2 Generation and cancellation of early warning
Generally speaking, the "mean control chart" reflects the fluctuation concentration trend of the sample group, and the "standard deviation control chart" reflects the measurement of the sample.
Variation of deviation between test values. Therefore, from the "mean control chart", we can see the overall index level of the sample during this period, when the data points exceed
When crossing the UCL line, it shows that the content of harmful substances in the sample shows a great upward trend during this period; When in the standard deviation control chart
When the data points exceed the UCL line of the standard deviation, it shows that each sample in the sample changes greatly during this time period and has a high dispersion [3]. For these two feelings
Forms, all need to attract the attention of testers. When the data points in both control charts exceed the UCL line, it means that each sample in this time period has two fingers.
The standard value is on the high side and fluctuates greatly. Therefore, it is necessary to send out early warning signals, strengthen detection and reduce detection risks.
After strengthening the sampling inspection for a period of time, when the average and standard deviation data points of the latest sample group are lower than the UCL line for three consecutive points, the pre-disqualification can be lifted, and an early warning signal can be generated to remind relevant personnel to pay attention and strengthen the key inspection.
2. 1.2 Determination of early warning threshold
For specific detection indicators, the determination of early warning threshold is mainly based on the maximum limit or minimum detection limit of harmful substances stipulated by national technical regulations.
2. 1.3 Early warning cancelled
For single indicators with unqualified early warning, the early warning can be cancelled under the following circumstances.
A) For an inspection unit or a certain variety, the monitoring indicators have not failed for 25 consecutive batches;
B) 65,438+000 data have been continuously detected by monitoring indicators, and none of them is unqualified.
2.2 Early warning of abnormal fluctuation of unqualified rate detected by specific indicators
In actual detection, the detection results of most indicators are less than the maximum limit value (MRL) and greater than the detection lower limit value. In this case, by
By analyzing the change trend of the test value, we can prompt a certain development trend of the index, so as to take measures in advance and strengthen the monitoring of the index. use
Mean-standard deviation control chart (X-S chart) can accurately identify whether the system is in a stable state [2].
2.2. 1 Establishment of mathematical model
The historical detection data are grouped reasonably. Each group of data is set to dig (usually 4 ~ 5 spells) and the data group is K (avoid ≥25), with samples respectively.
Taking this number as the abscissa and the mean or standard deviation of this group of test data as the ordinate, the "mean control chart" and "standard deviation control chart" are established respectively.
Calculate the average value of each sample and standard parameter I.
Zi- 13 ≥: zif (9)
Daozai
Si-√i and √-i)2
( 10)i= 1(xo
Where: Xij is the j-th value of the i-th sample; I is the average of group I; Played as the average of subgroup average; K is the number of samples, which is generally not counted.
The standard deviation of sliding to the ith sample is less than 25.
Calculate the average and standard deviation of k samples:
Eleven {≥: {≥:Xi (1 1+0)
Nickel -T- 1
J- chamber si ⅲ,
For the mean control chart:
Central control line cl: cl-j (13)
Upper control line UCL: UCL = X+A35 (14)
Among them, a. As a control factor, it can be directly obtained by looking up GB/T 409 1 standard.
At point a 4, A3- 1.682; when cursing a 5, A3= 1-427.
For standard deviation control chart
Central control line cl: cl-s (15)
Upper control line UCL: UCL-B4 south (14)
Similarly, B. is the controlling factor of S, which can also be obtained by directly looking up GB/T 409 1 standard.
When, 2-4, B4-2.266, when n=5, B4-2.089.
2.2.2 Generation and cancellation of early warning
Generally speaking, the "mean control chart" reflects the fluctuation concentration trend of the sample group, and the "standard deviation control chart" reflects the measurement of the sample.
Variation of deviation between test values. Therefore, from the "mean control chart", we can see the overall index level of the sample during this period, when the data points exceed
When crossing the UCL line, it shows that the content of harmful substances in the sample shows a great upward trend during this period; When in the standard deviation control chart
When the data points exceed the UCL line of the standard deviation, it shows that each sample in the sample changes greatly during this time period and has a high dispersion [3]. For these two feelings
Forms, all need to attract the attention of testers. When the data points in both control charts exceed the UCL line, it means that each sample in this time period has two fingers.
The standard value is on the high side and fluctuates greatly. Therefore, it is necessary to send out early warning signals, strengthen detection and reduce detection risks.
After strengthening sampling inspection for a period of time, when the average and standard deviation data points of the latest sample group are lower than UCL line for three consecutive points, the pre-inspection can be cancelled.
Police, return to normal detection rate.
This control chart based on a large number of historical detection data is mainly suitable for monitoring some detection failure rates, and the detection results can be used.
Physical examination shows the detection indicators in the form of data, such as the current formaldehyde content, heavy metal content, pH value of water extract and other items.
In daily analysis and control, because the calculation of range is much simpler than the calculation of standard deviation, range R can be used instead of "average"
The standard deviation s in the quasi-difference control chart is obtained, and the "mean-range control chart (x_ r chart)" is obtained.
3. Early warning based on hazard risk coefficient
The appearance of early warning signal means that the unqualified rate of harmful substances in the detection of this period has obviously increased or the detection indicators have obviously abnormal.
Fluctuation, that is, the possibility of harmful substances exceeding the standard in the product increases and the risk is greater. But in fact, because a specific indicator failed to pass the test.
Rate is related to the number of spot checks, that is, it is closely related to the self-inspection rate, so in a sense, for some projects with low self-inspection rate, although testing
The unqualified rate found in this paper is very low, but its potential risk is still high. Because, from a statistical point of view, the smaller the number of samples, the smaller the representativeness of the population.
The worse, the lower the credibility of the results. Draw lessons from the experience of food safety monitoring and early warning system, and also set the parameter of "hazard risk coefficient"
Number to comprehensively analyze the generation of early warning information [1].
Because the risk caused by harmful substances is related to the unqualified rate of its detection indicators, the self-inspection rate and the sensitivity to harmful substances, it can be done.
The following definition: R'-AP+6/F+S (15)
Where: R7 is the risk coefficient of danger; P is the failure rate of hazard detection; F is the sampling ratio of hazard sources, that is, the self-inspection rate; S is in danger
Hazard risk factors; Mouth and b are the corresponding weight coefficients.
In practical application, it is suggested that a value is 100 and b value is 0. 1.
S value is mainly based on the degree of concern and sensitivity to specific hazards. Usually, S= is set for the current high-concern and sensitive hazards.
2; For low attention and low sensitivity, set S = 0.5 and set S-L to moderate.
Take S= 1 as an example to explain the setting process of risk coefficient.
A) According to historical data, divide the samples within a certain period of time (such as one month) and calculate the sampling rate e and the unqualified rate p-sum of each sample respectively.
Average sampling rate r and average failure rate p.
B) Calculate the risk coefficient R 7i and the average risk coefficient R 7 of each sample. And standard deviation.
C) with R7. +/-port is the upper and lower control line, setting:
When R7≤R7. When a bite is in a low-risk state, at this time, the proportion of spot checks can be reduced;
When R7≥R7. +tag, it is in a high-risk state. At this time, the proportion of spot checks should be increased;
When R7. stare
D) Under the average sampling rate level, when each batch of tests is qualified, if the aP is 0, R7-6/r+s is the lowest risk.
The statistical results of banned aromatic amine dyes (azo) in a laboratory from 1 to 10 (taking one month as a sample) are shown in Table 2.
From the calculation in Table 2, we can draw the following conclusions: According to the statistics of AZO test results of a laboratory from June 5438+0 to June 5438+00, when R7 is less than or equal to 2.29, it is in a low-risk state, and at this time it is possible to reduce the risk-I can be 10,000 in the face of summer-I can be 10,000 in the face of rain.
Low sampling rate;
When R7≥5.35, it is at high risk. At this time, it should be
Increase the proportion of spot checks;
When 2.29
At this point, maintain the existing spot check ratio.
E) at the average sampling rate of 15.59%, when every
When all batches pass the test, R7-6/R+S if aP is 0.
A 1.64, when the risk is the lowest.
4 conclusion
Using the principle of "control chart", the ecological textiles are inspected.
Some items in the indicators are not established by different analysis methods.
1 17. 18
2 20.06
3 16. 14
4 17.3 1
5 14.93
6 15. 14
7 14.73
8 14.5 1
9 13.35
10 14.3 1
Average 15.59
mean standard deviation
Lower limit: 3.82— 1.53—2.29.
2.47 4.05
1.57 3.07
1.63 3.25
2. 19 3.77
2.83 4.50
2.37 4.03
2.57 4.25
1.62 3.3 1
0.47 2.22
4.00 5.70
2. 17 3.82
1.53
Upper limit: 3.82+ 1.53—5.35.
The same early warning mode. In practical application, different analysis control chart modes should be selected according to different monitoring objects. If it is necessary to analyze a certain type,
When monitoring products or a certain type of enterprise, the methods of unqualified numerical control chart (rip chart) and unqualified rate control chart (household chart) should be given priority; p.r.n.
When monitoring an ecological detection index, "mean-standard deviation control chart (X-S chart)" or "mean-range control chart (X. _R R.
Figure) Establish an early warning analysis model; The early warning model established by "hazard risk coefficient" method can better reflect the detection of a hazard.
This index shows the quality risk under different sampling ratios, so it is more suitable for inspection and quarantine laboratories.
References:
[1] Li Cong, Huang Yimin. Food safety monitoring and early warning system [M]. Beijing Chemical Industry Press 2006: 76.
[23 GB/T 409 1-200 1, general control chart [S].
[3] You Jianxin, Zhang Jiantong and Du Xuemei. Quality management [M]. Beijing Science Press 2003.
Establishment of early warning method for ecological textiles testing
Zhao Shan-Hong 9 1,-Chun, Yao Jiping 2, Guo Fang-Long 92
(1. Zhejiang Entry-Exit Inspection and Quarantine Bureau. China Hangzhou Quarantine Bureau, 310012;
2. Zhejiang Ryder Product Technology Co., Ltd., Hangzhou, China 3 10005)
Abstract: This paper studies the theory of process quality control and mathematical statistics methods.
Test data of ecological textiles in application laboratory. Different early warning methods and quality control models
The analysis object is established to realize analysis supervision (or test indicators) and provide-
Provide corresponding quality trend analysis information for enterprises and government departments. as—
The establishment of this method is helpful for the relevant departments to achieve rapid and effective goals as soon as possible.
Warnings under existing test resources.
Keywords: eco-textiles; Testing; Early warning method; Quality control 0 1
(Editor: Xu Huier)