CN108470229A - One kind being based on CpkThe reliability of bulk article enter detecting method - Google Patents

One kind being based on CpkThe reliability of bulk article enter detecting method Download PDF

Info

Publication number
CN108470229A
CN108470229A CN201710098170.1A CN201710098170A CN108470229A CN 108470229 A CN108470229 A CN 108470229A CN 201710098170 A CN201710098170 A CN 201710098170A CN 108470229 A CN108470229 A CN 108470229A
Authority
CN
China
Prior art keywords
reliability
supplier
product
value
key characteristic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201710098170.1A
Other languages
Chinese (zh)
Inventor
邵家骏
邢晓艳
吴晓华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sunlight Tech Beijing
Original Assignee
Sunlight Tech Beijing
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sunlight Tech Beijing filed Critical Sunlight Tech Beijing
Priority to CN201710098170.1A priority Critical patent/CN108470229A/en
Publication of CN108470229A publication Critical patent/CN108470229A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Factory Administration (AREA)

Abstract

The invention discloses a kind of typical products in mass production reliability incoming test methods.Determine the several key characteristic parameters for certain product that each supplier provides, it is desirable that while supplier delivers the product every time, provide the C of these key characteristic parameters respectivelypk.Enterprise samples when progress reliability enters to examine and tests these key characteristic parameters, calculate Cpk.By calculated CpkThe C provided with supplierpkIt compares and analyzes, it is determined whether this batch of product can be received.The present invention solves the problems, such as that the reliability of typical products in mass production common at present and difficult to deal with enters inspection, it is proposed that one kind being based on CpkCertificate authenticity method, both certain batch of product reliability level can be judged, it the presence or absence of reliability level difference can be distinguished between different batches again, the height of reliability level between different suppliers can also be compared, engineer application reality has been fully considered simultaneously, it is safe, lossless, inexpensive, simple, conscientiously operable.

Description

One kind being based on CpkThe reliability of bulk article enter detecting method
Technical field
The present invention relates to certificate authenticity method fields.It provides a set of method simple to operation, is particularly suitable for The inspection whether reliability level of typical products in mass production (such as high-volume component, high-volume machine product) meets the requirements.
Background technology
At present about the incoming test of product, substantially just for the inspection of performance indicator, and about reliability level There is no good ways and means for the inspection of height.Enter inspection especially for typical products in mass production, carrying out 100% inspection is Impossible or very high cost thing.Therefore, in enterprise engineering practical application, there is an urgent need to a kind of convenient and efficient, lossless, low The reliability incoming test method of cost, can quickly judge whether this batch of product meets reliability requirement.Due to CpkIt is higher, mistake Cheng Bianyi is smaller, and product reliability is higher, therefore can use CpkAs a kind of parameter for portraying reliability level height.The present invention carries It has supplied a kind of using CpkMethod carry out bulk article reliability enter detecting method.
Invention content
The invention aims to provide a kind of convenient and efficient, lossless, the inexpensive reliability of bulk article to enter factory's inspection Proved recipe method can judge the height of this batch of product reliability level；Simultaneously can also between different batches reliability Level has indifference to be compared；While the height of reliability level between different suppliers can also be compared.
In order to solve the above technical problems, method provided by the invention is as follows：
If there is n supplier to provide certain product (n >=1), determine that several key characteristic parameters of the product (are considered Engineer application is practical, generally can be 1~5), it is assumed that it is xi(i >=1), it is desirable that while supplier delivers the product every time, The C of these key characteristic parameters is provided respectivelypk.Enterprise is when carrying out reliability and entering to examine, the random selection product at least 25 As sample, the test result x of these key characteristic parameters is tested and recordedij(i≥1；J=1,2,3 ... k；K >=25), meter Calculate Cpk.By calculated CpkThe C provided with supplierpkIt compares and analyzes, it is determined whether this batch of product can be received.Work as confession Answer the product that quotient provides when not being the same batch, then different batches inspect to calculating C respectivelypk, different batches can be compared The difference of secondary reliability level.When different suppliers provides the product, C is calculated then different suppliers are inspected by random samples respectivelypk, The height of different supplier's reliability levels can be compared.
It is as follows：
Step 1：Determine several key characteristic parameters of certain product, it is assumed that be xi(i≥1)。
Step 2：Assuming that certain enterprise there are n different suppliers to supply the product (n >=1), different suppliers are randomly selected Certain product of (e.g., tri- different suppliers of A, B, C) different batches respectively at least 25.
Step 3：The test result for testing and recording these key characteristic parameters respectively, is denoted as xI, j, n(i≥1；J=1, 2,3 ... k；k≥25；n≥1).
Step 4：Test of normality is carried out to test data.If data Normal Distribution passes directly to the 5th step； Otherwise, the 5th step is gone to after carrying out normality conversion.
Step 5：According to following procedure Capability index CpkCalculation formula, calculate separately different supplier's different batches The C of every group of test datapkValue.
In above formula,
Related symbol is described as follows in formula：
CPL：Unilateral lower limit Measure of Process Capability
CPU：Unilateral upper limit Measure of Process Capability
Cpk：Measure of Process Capability
The mean value of all sample measurements
LSL：Lower specification limit
USL：Upper specification limit
σ：Process inherently fluctuates
μ：Mean value
Mean range
d2：Control figure coefficient, it is related with sample size
Step 6：When meeting following three kinds of conditions simultaneously, this batch of product can be received；Otherwise, then reject this batch of product or It proposes to carry out the improved requirement of product quality reliability to supplier.
i.Cpk1.33 should be not less than as far as possible；
Ii. by calculated CpkThe C that value is provided with supplierpkValue is compared, and the error of the two is within 10%；
Ii. the C between different batchespkValue difference cannot not be more than 10%.
Step 7：Respectively by the calculated C of different supplierspkValue is compared, CpkThe bigger supplier's reliability of value is more It is high.
The advantage of the invention is that：
I. the present invention be directed to common at present and intractable typicals products in mass production (such as high-volume component, high-volume complete machine Product etc.) reliability enter inspection problem, it is proposed that one kind be based on CpkCertificate authenticity method, this method both can to certain batch produce Product reliability level is judged, and the presence or absence of reliability level difference can be distinguished between different batches, can be with Compare the height of reliability level between different suppliers.
Ii is proposed by the present invention to be based on CpkReliability enter detecting method, can be filled by means of the test data of performance indicator Divide and considers engineer application reality, it is safe, lossless, inexpensive, simple, conscientiously operable.
Description of the drawings
Fig. 1 is the test of normality result of 1 key characteristic parameter x1 of supplier's A batches；
Fig. 2 is the test of normality result of 1 key characteristic parameter x2 of supplier's A batches；
Fig. 3 is the test of normality result of 2 key characteristic parameter x1 of supplier's A batches；
Fig. 4 is the test of normality result of 2 key characteristic parameter x2 of supplier's A batches；
Fig. 5 is the test of normality result of 3 key characteristic parameter x1 of supplier's B batches；
Fig. 6 is the test of normality result of 3 key characteristic parameter x2 of supplier's B batches；
Fig. 7 is supplier's A batches 1, the Cpk result of calculations of 2 key characteristic parameters x1, x2；
Fig. 8 is the Cpk result of calculations of 3 key characteristic parameter x1, x2 of supplier's B batches.
Symbol, code name are described as follows in figure：
N：Number of samples (see Fig. 1~Fig. 6)
AD：Inspection statistics magnitude (see Fig. 1~Fig. 6)
P values：Significance (see Fig. 1~Fig. 6)
LSL：Lower specification limit (see Fig. 7)
USL：Upper specification limit (see Fig. 7)
LSL*：Lower specification limit after transformation (see Fig. 8)
USL*：Upper specification limit after transformation (see Fig. 8)
Cp：Latent process Capability index (see Fig. 7, Fig. 8)
CPL：Unilateral lower limit Measure of Process Capability (see Fig. 7, Fig. 8)
CPU：Unilateral upper limit Measure of Process Capability (see Fig. 7, Fig. 8)
Cpk：Measure of Process Capability (see Fig. 7, Fig. 8)
Specific implementation mode
Below with reference to example is implemented, the present invention is further illustrated, which is used for illustrating and noting limit The present invention.
In the case study on implementation of the present invention, provide a kind of based on CpkComponent reliability enter detecting method.Certain enterprise Reliability, which is carried out, in certain component for the different batches supplied to Liang Jia suppliers A, B enters inspection, it is known that specification limit requires as 2 ± 0.2,3 ± 0.2, supplier A, B provide x1、x2CpkIt is 2.6,2.5.Examination carries out this batch of component of supplier A, B reliable Property enter inspection.
Specifically entering detecting method includes：
Step 1：Analysis determines there are two the determination key characteristic parameters of this kind of component, is denoted as x1、x2, specification limit requirement It is 2 ± 0.2,3 ± 0.2；
Step 2：The enterprise only supplies this kind of component there are two supplier A, B, randomly selects Liang Ge suppliers and this time carries Each 25 of this kind of component of the different batches of confession；
Step 3：It tests and records key characteristic parameter x1、x2Test result be shown in Table 1, simplification be denoted as xI, j(j= 1,2,3 ... k；K=25), x2, j(j=1,2,3 ... k；K=25)；
1 key characteristic parameter x of table1、x2Test result
Step 4：Test of normality is carried out to test data.If data Normal Distribution passes directly to the 5th step； Otherwise, the 5th step is gone to after carrying out normality conversion.
Six groups of data carry out the result is shown in Figure 1~Fig. 6 of test of normality respectively.Since 1 key characteristic of supplier's A batches is joined Number x1,1 key characteristic parameter x2 of supplier's A batches, 2 key characteristic parameter x1 of supplier's A batches, supplier's A batches 2 are crucial special The p value of property parameter x2 is all higher than 0.05, can directly be transferred to the 5th step and be calculated；3 key characteristic parameter x1 of supplier's B batches, The p value of 3 key characteristic parameter x2 of supplier's B batches is respectively less than 0.05, then needs to go to the 5th step after carrying out normality conversion.
Step 5：According to Measure of Process Capability CpkCalculation formula, 1 key characteristic parameter x of supplier's A batches1Test number According to CpkValue calculates as follows.
Cpk=min { CPL, CPU}=min { 2.560,2.684 }=2.56 (10)
Similarly, it can be calculated with same method：1 key characteristic parameter x2 of supplier's A batches, supplier's A batches 2 are crucial special Property parameter x1,2 key characteristic parameter x2 of supplier's A batches；3 key characteristic parameter x1 of supplier's B batches, supplier's B batches 3 are closed The C of key characterisitic parameter x2pkValue.Summarize result of calculation and is shown in Table 2.Energy F-histogram is shown in Fig. 7,8 respectively.
2 supplier's A, B batch of table, 1,2,3 key characteristic parameter x1、x2The C of test datapkResult of calculation
Step 6：Due to two different batches x of supplier A1、x2CpkIt is all higher than 1.33；And provided due to supplier A x1、x2CpkRespectively 2.6,2.5, calculated CpkThe C that value is provided with supplierpkValue is compared, and the two error is 10% Within；And the C between batch 1 and batch 2pkValue difference is not also within 10%；Therefore the two batches of supplier A can be received Product.
Due to supplier B batches x1、x2CpkRespectively less than 1.33；And due to the x of supplier B offers1、x2CpkRespectively It is 2.6,2.5, the two error is all higher than 10%, therefore, refusal receives this batch of product of supplier B, while proposing to carry out to supplier B The improved requirement of product quality reliability.
Step 7：Respectively by the calculated C of different supplierspkValue is compared, the C of supplier ApkIt is worth larger, reliability It is higher.Supplier B can be required to learn to supplier A, promote the reliability level for reaching supplier A.
The reliability that the seven above steps solve three different batches of product of Liang Ge suppliers through the invention enters inspection Problem not only judged each batch products reliability level, but between different batches reliability level height into It has gone differentiation, has also provided a comparison of the height of reliability level between different suppliers, while having fully considered engineer application reality, peace Entirely, lossless, inexpensive, simple, conscientiously operable.

Claims (1)

1. one kind being based on CpkThe reliability of bulk article enter detecting method, which is characterized in that include the following steps：
Step 1：Determine several key characteristic parameters of certain product, it is assumed that be xi(i≥1)；
Step 2：Assuming that certain enterprise there are n different suppliers to supply the product (n >=1), different suppliers are randomly selected (e.g., A, tri- different suppliers of B, C) different batches certain product respectively at least 25；
Step 3：The test result for testing and recording these key characteristic parameters respectively, is denoted as xI, j, n(i≥1；J=1,2, 3 ... k；k≥25；n≥1)；
Step 4：Test of normality is carried out to test data：If data Normal Distribution passes directly to the 5th step；Otherwise, The 5th step is gone to after carrying out normality conversion；
Step 5：According to following procedure Capability index CpkCalculation formula, calculate separately every group of different supplier's different batches The C of test datapkValue；
In above formula,
Related symbol is described as follows in formula：
CPL：Unilateral lower limit Measure of Process Capability
CPU：Unilateral upper limit Measure of Process Capability
Cpk：Measure of Process Capability
The mean value of all sample measurements
LSL：Lower specification limit
USL：Upper specification limit
σ：Process inherently fluctuates
μ：Mean value
Mean range
d2：Control figure coefficient, it is related with sample size
Step 6：When meeting following three kinds of conditions simultaneously, this batch of product can be received；Otherwise, then this batch of product is rejected or to confession Quotient is answered to propose to carry out the improved requirement of product quality reliability：
i.Cpk1.33 should be not less than as far as possible；
Ii. by calculated CpkThe C that value is provided with supplierpkValue is compared, and the error of the two is within 10%；
Ii. the C between different batchespkValue difference cannot not be more than 10%；
Step 7：Respectively by the calculated C of different supplierspkValue is compared, CpkThe bigger supplier's reliability of value is higher.
CN201710098170.1A 2017-02-23 2017-02-23 One kind being based on CpkThe reliability of bulk article enter detecting method Pending CN108470229A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710098170.1A CN108470229A (en) 2017-02-23 2017-02-23 One kind being based on CpkThe reliability of bulk article enter detecting method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710098170.1A CN108470229A (en) 2017-02-23 2017-02-23 One kind being based on CpkThe reliability of bulk article enter detecting method

Publications (1)

Publication Number Publication Date
CN108470229A true CN108470229A (en) 2018-08-31

Family

ID=63266849

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710098170.1A Pending CN108470229A (en) 2017-02-23 2017-02-23 One kind being based on CpkThe reliability of bulk article enter detecting method

Country Status (1)

Country Link
CN (1) CN108470229A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112883332A (en) * 2019-11-29 2021-06-01 沈阳石蜡化工有限公司 Cpk-based method for testing control stability of oil refining chemical device
CN114114044A (en) * 2021-11-15 2022-03-01 天津市捷威动力工业有限公司 Method for evaluating welding overcurrent reliability of tab-busbar module

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101937212A (en) * 2009-07-03 2011-01-05 中芯国际集成电路制造(上海)有限公司 Process detection method and device
CN102982234A (en) * 2012-10-26 2013-03-20 西安电子科技大学 Computing method of multivariable process capability index
CN104268392A (en) * 2014-09-23 2015-01-07 北京航空航天大学 Manufacturing process product reliability decline risk evaluating method based on quality deviation
CN104484747A (en) * 2014-12-01 2015-04-01 西安电子科技大学 Method for determining qualified rate of products by utilizing truncation samples

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101937212A (en) * 2009-07-03 2011-01-05 中芯国际集成电路制造(上海)有限公司 Process detection method and device
CN102982234A (en) * 2012-10-26 2013-03-20 西安电子科技大学 Computing method of multivariable process capability index
CN104268392A (en) * 2014-09-23 2015-01-07 北京航空航天大学 Manufacturing process product reliability decline risk evaluating method based on quality deviation
CN104484747A (en) * 2014-12-01 2015-04-01 西安电子科技大学 Method for determining qualified rate of products by utilizing truncation samples

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112883332A (en) * 2019-11-29 2021-06-01 沈阳石蜡化工有限公司 Cpk-based method for testing control stability of oil refining chemical device
CN114114044A (en) * 2021-11-15 2022-03-01 天津市捷威动力工业有限公司 Method for evaluating welding overcurrent reliability of tab-busbar module

Similar Documents

Publication Publication Date Title
CN107478455B (en) Timing truncation reliability test method suitable for Weibull distribution type product
CN112382582A (en) Wafer test classification method and system
CN105702595B (en) The yield judgment method of wafer and the changeable quantity measuring method of wafer conformity testing
CN104062305B (en) A kind of analysis method of integrated circuit defect
CN106021928A (en) Comprehensive stress acceleration test method
CN108400098A (en) The method for verifying wafer test correlation
CN101261285A (en) Method and system for automatically managing probe mark shifts
CN108470229A (en) One kind being based on CpkThe reliability of bulk article enter detecting method
CN109579963A (en) Autonomous type weighing anti-cheating method and system
CN102945222A (en) Poor information measurement data gross error discrimination method based on Grey System Theory
De Mast et al. Measurement system analysis for binary inspection: Continuous versus dichotomous measurands
CN110486350B (en) Electro-hydraulic servo valve fault diagnosis method and device, storage medium and electronic equipment
CN110687492A (en) Method and system for monitoring detection quality of automatic verification system of low-voltage current transformer on line
CN101592692B (en) Evaluation method of measuring machines
CN108063101A (en) A kind of methods of sampling of wafer defect test
CN114841573B (en) Grouping inspection scheme determining system and method for exponential lifetime type products
KR101482758B1 (en) Fault detection method
Ramaswamy et al. Time truncated chain sampling plans for generalized exponential distribution
CN111967799B (en) Method for identifying materials and process problems of circulator by sticking integrated parameter table
WO2005093819A1 (en) Processing unit, display method and display program
CN114578209A (en) Automatic FCT turntable tester control system
CN107340487A (en) A kind of method checked test system and be in actual processing ability under stable state
CN103559566B (en) A kind of work dispatching method of Defect Scanning output control process work bench
Celano A bootstrap method for the measurement error estimation in Gauge R & R R\&R Studies
Cox et al. Discrete-event simulation of process control in low volume high value industries

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20180831

WD01 Invention patent application deemed withdrawn after publication