ASTM E456-2006e1 780 Standard Terminology Relating to Quality and Statistics.pdf

1、Designation: E 456 06e1An American National StandardStandard TerminologyRelating to Quality and Statistics1This standard is issued under the fixed designation E 456; the number immediately following the designation indicates the year oforiginal adoption or, in the case of revision, the year of last

2、revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (e) indicates an editorial change since the last revision or reapproval.This standard has been approved for use by agencies of the Department of Defense.e1NOTEE456 was updated editorially in February 2008.1

3、. Scope1.1 This standard is the general terminology standard forterms defined in the standards of Committee E11 on Qualityand Statistics.1.2 A term in this standard which lists an attribution to anE11 technical standard indicates that the standard is normativefor that term. Any changes in the term d

4、efinition in thenormative standard will be editorially changed in this standard.Any terms added to an E11 standard will be editorially addedto this standard with an attribution to that standard.1.3 Term definitions that are similar to ISO 3534 will benoted in this standard, but ISO 3534 will not be

5、considerednormative for any E11 terms.2. Referenced Documents2.1 ASTM E11 Standards with terms in this standard:2E29 Practice for Using Significant Digits in Test Data toDetermine Conformance with SpecificationsE 177 Practice for Use of the Terms Precision and Bias inASTM Test MethodsE 178 Practice

6、for Dealing With Outlying ObservationsE 1169 Practice for Conducting Ruggedness TestsE 1325 Terminology Relating to Design of ExperimentsE 1488 Guide for Statistical Procedures to Use in Develop-ing and Applying Test MethodsE 1994 Practice for Use of Process Oriented AOQL andLTPD Sampling PlansE 223

7、4 Practice for Sampling a Stream of Product byAttributes Indexed by AQLE 2281 Practice for Process and Measurement CapabilityIndicesE 2282 Guide for Defining the Test Result of a Test MethodE 2334 Practice for Setting an Upper Confidence Bound Fora Fraction or Number of Non-Conforming items, or a Ra

8、teof Occurrence for Non-conformities, UsingAttribute Data,When There is a Zero Response in the SampleE 2489 Practice for Statistical Analysis of One-Sample andTwo-Sample Interlaboratory Proficiency Testing ProgramsE 2554 Practice for Estimating and Monitoring the Uncer-tainty of Test Results of a Te

9、st Method in a SingleLaboratory Using a Control Sample ProgramE 2555 Practice for Factors and Procedures forApplying theMIL-STD-105 Plans in Life and Reliability InspectionE 2586 Practice for Calculating and Using Basic StatisticsE 2587 Practice for Use of Control Charts in StatisticalProcess Contro

10、l2.2 ISO Standards:ISO 3534 Statistics-Vocabulary and SymbolsPart 1: Probability and General Statistical TermsPart 2: Applied Statistics3. Terminologyacceptance (control chart or acceptance control chartusage, ), na decision that the process is operating in asatisfactory manner with respect to the s

11、tatistical measuresbeing plotted: action limits: control limits.acceptance quality limit AQL , nquality limit that is theworst tolerable process average when a continuing series oflots is submitted for acceptance sampling. E 2234accepted reference value, na value that serves as anagreed-upon referen

12、ce for comparison, and which is derivedas: (1) a theoretical or established value, based on scientificprinciples, (2) an assigned or certified value, based onexperimental work of some national or international organi-zation, or (3) a consensus or certified value, based oncollaborative experimental w

13、ork under the auspices of ascientific or engineering group. E 177accuracy, nthe closeness of agreement between a test resultand an accepted reference value. E 177aliases, nin a fractional factorial design, two or more effectswhich are estimated by the same contrast and which,1This terminology is und

14、er the jurisdiction ofASTM Committee E11 on Qualityand Statistics and is the direct responsibility of Subcommittee E11.70 on Editorial/Terminology.Current edition approved Sept. 1, 2006. Published November 2006. Originallyapproved in 1972. Last previous edition approved in 2004 as E 456 04e1.2For re

15、ferenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM Customer Service at serviceastm.org. For Annual Book of ASTMStandardsvolume information, refer to the standards Document Summary page onthe ASTM website.1Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, We

16、st Conshohocken, PA 19428-2959, United States.therefore, cannot be estimated separately. E 1325assignable cause, nfactor that contributes to variation in aprocess or product output that is feasible to detect andidentify (see special cause). E 2587attributes data, nobserved values or test results tha

17、t indi-cate the presence or absence of specific characteristics orcounts of occurrences of events in time or space. E 2587attributes, method of, nmeasurement of quality by themethod of attributes consists of noting the presence (orabsence) of some characteristic or attribute in each of theunits in t

18、he group under consideration, and counting howmany units do (or do not) possess the quality attribute, orhow many such events occur in the unit, group, or area.E 2334average outgoing quality (AOQ)the average percent defec-tive of outgoing product including all accepted lots orbatches, after any defe

19、ctive units found in them are replacedby acceptable units, plus all lots or batches which are notaccepted after such lots or batches have been effectively100 % inspected and all defective units replaced by accept-able units. E 1994average outgoing quality limit (AOQL)the maximum oftheAOQs for all po

20、ssible incoming percentages defective forthe process, for a given acceptance sampling plan. E 1994average quality protectiona type of protection in whichthere is prescribed some chosen value of average percentdefective in the product after inspection (average outgoingquality limit (AOQL), that shall

21、 not be exceeded in the longrun no matter what may be the level of percent defective inthe product submitted to the inspector. E 1994average run length (ARL)the average number of times thata process will have been sampled and evaluated before ashift in process level is signaled. E 2587average standa

22、rd deviation, s, narithmetic average ofsample standard deviations. E 2281balanced incomplete block design (BIB), nan incompleteblock design in which each block contains the same numberk of different versions from the t versions of a singleprincipal factor arranged so that every pair of versionsoccur

23、s together in the same number, l, of blocks from the bblocks. E 1325batch, na definite quantity of some product or materialproduced under conditions that are considered uniform.NOTE 1A batch is usually smaller than a lot.batch, in inspection, na collection of units of productproduced under condition

24、s that are considered uniform andfrom which a sample is drawn and inspected, and may differfrom a collection of units designated as a batch for otherpurposes, for example, production, shipment, etc. E 2234batch size, nthe number of units of product in a batch.E 2234bias, nthe difference between the

25、expectation of the testresults and an accepted reference value. E 177c chartcontrol chart that monitors the count of occurrencesof an event in a defined increment of time or space.E 2587center lineline on a control chart depicting the average levelof the statistic being monitored. E 2587chance cause

26、 source of inherent random variation in aprocess which is predictable within statistical limits (seecommon cause). E 2587characteristic, na property of items in a sample or popula-tion which, when measured, counted or otherwise observed,helps to distinguish between the items. E 2282check samplesee c

27、ontrol sample. E 2554classification of defects, nthe enumeration of possibledefects of the unit of product arranged according to theirseriousness, that is, critical, major, or minor defect.E 2234cluster sampling, nwhen the primary sampling unit com-prises a bundle of elementary units or a group of s

28、ubunits,the term cluster sampling may be applied.DISCUSSIONExamples of cluster sampling are: selection of cityblocks as primary sampling units; selection of a household as a clusterof people (of which only one may be interviewed); selection of bundlesof rods or pipe from a shipment; and selection, f

29、rom a shipment, ofcartons that contain boxes or packages within them.coefficient or variation, CVfor a nonnegative characteristic,the ratio of the standard deviation to the mean for apopulation or sample. . E 2586collaborative study, ninterlaboratory study in which eachlaboratory uses the defined me

30、thod of analysis to analyzeidentical portions of homogeneous materials to assess theperformance characteristics obtained for that method ofanalysis. E 2489collaborative trial, nsee collaborative study. E 2489common causesee chance cause. E 2587completely randomized design, na design in which thetrea

31、tments are assigned at random to the full set of experi-mental units. E 1325completely randomized factorial design, na factorial ex-periment (including all replications) run in a completelyrandomized design. E 1325component of variance, na part of a total variance identifiedwith a specified source o

32、f posite design, na design developed specifically forfitting second order response surfaces to study curvature,constructed by adding further selected treatments to thoseobtained from a 2nfactorial (or its fraction). E 1325confidence bound, nsee confidence limit. E 2334confidence coefficient, nthe va

33、lue, C, of the probabilityassociated with a confidence interval or statistical coverageinterval. It is often expressed as a percentage. ISO 3534-1E 2334confidence level, nsee confidence coefficient. E 2334confidence limit, neach of the limits, T1and T2, of the twosided confidence interval, or the li

34、mit T of the one sidedconfidence interval. E 2334confounded factorial design, na factorial experiment inwhich only a fraction of the treatment combinations are runin each block and where the selection of the treatmentcombinations assigned to each block is arranged so that oneor more prescribed effec

35、ts is(are) confounded with the blockeffect(s), while the other effects remain free from confound-ing. E 1325E45606e12confounding, ncombining indistinguishably the main effectof a factor or a differential effect between factors (interac-tions) with the effect of other factor(s), block factor(s) orint

36、eractions(s). E 1325consumers riskthe probability that a lot whose percentagedefective is equal to the LTPD will be accepted by the plan.E 1994contrast, na linear function of the observations for whichthe sum of the coefficients is zero. E 1325contrast analysis, na technique for estimating the param

37、-eters of a model and making hypothesis tests on preselectedlinear combinations of the treatments (contrasts). E 1325control(evaluation), nan evaluation to check, test, orverify; (authority): the act of guiding, directing, or manag-ing; (stability): a state of process in which the variability isattr

38、ibutable to a constant system of chance causes.control chart , n-chart on which are plotted a statisticalmeasure of a subgroup versus time of sampling along withlimits based on the statistical distribution of that measure soas to indicate how much common, or chance, cause variationis inherent in the

39、 process or product. E 2587control chart factor, na tabulated constant, depending onsample size, used to convert specified statistics or parametersinto a central line value or control limit appropriate to thecontrol chart. E 2587control chart method, nthe method of using control chartsto determine w

40、hether or not processes are in a stable state.control limits, nlimits on a control chart that are used ascriteria for signaling the need for action or judging whethera set of data does or does not indicate a state of statisticalcontrol based on a prescribed degree of risk. E 2587control sample, nsam

41、ple taken from a stable, homogeneousmaterial for the purposes of monitoring the performance ofa test method in a laboratory. E 2554critical defect, na defect that judgment and experienceindicate would result in hazardous or unsafe conditions forindividuals using, maintaining, or depending upon the p

42、rod-uct, or a defect that judgment and experience indicate islikely to prevent performance of the function of a major enditem. E 2234critical defective, na unit of product which contains one ormore critical defects and may also contain major and/orminor defects. E 2234defect, nany nonconformance of

43、the unit of product withspecified requirements. E 2234defective, na unit of product which contains one or moredefects. E 2234defects per hundred units, nany given quantity of units ofproduct is one hundred times the number of defects con-tained therein (one or more defects being possible in any unit

44、of product) divided by the total number of units of product,that is:Defects per hundred units 5Number of defects 3100Number of units inspectedE 2234dependent variable, nSee response variable. E 1325design of experiments, nthe arrangement in which anexperimental program is to be conducted, and the se

45、lectionof the levels (versions) of one or more factors or factorcombinations to be included in the experiment. Synonymsinclude experiment design and experimental design.E 1325deviation, nthe difference between a measurement or quasi-measurement and its stated value or intended level.DISCUSSIONDeviat

46、ion should be stated as a difference in terms ofthe appropriate data units. Sometimes these units will be originalmeasurement units; sometimes they will be quasi-measurements; thatis, a scaled rating of subjective judgments; sometimes they will bedesignated values representing all continuous or disc

47、rete measurementsfalling in defined cells or classes.double sampling plan, na multiple sampling plan in whichup to two samplings can be taken and evaluated to accept orreject a lot. E 2555empirical percentile, nestimate of a population percentileusing the sample data. This is a sample value such tha

48、t apercentage p of the sample is less than that value. E 2586error of result, nthe test result minus the accepted referencevalue (of the characteristic).NOTE 2It is not possible to correct for random error.evolutionary operation (EVOP), na sequential form ofexperimentation conducted in production fa

49、cilities duringregular production. E 1325experimental design, nsee design of experiments. E 1325experiment space, nthe materials, equipment, environmen-tal conditions and so forth that are available for conductingan experiment. E 1325experimental unit, na portion of the experiment space towhich a treatment is applied or assigned in the experiment.E 1325factorial experiment (general), nin general, an experimentin which all possible treatments formed from two or morefactors, each being studied at two