1、 ISO 2012 Water quality The variability of test results and the uncertainty of measurement of microbiological enumeration methods Qualit de leau - Variabilit des rsultats dessais et incertitude de mesure des mthodes dnumration microbienne INTERNATIONAL STANDARD ISO 29201 First edition 2012-01-15 Ref
2、erence number ISO 29201:2012(E) ISO 29201:2012(E) ii ISO 2012 All rights reserved COPYRIGHT PROTECTED DOCUMENT ISO 2012 All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocop
3、ying and microfilm, without permission in writing from either ISO at the address below or ISOs member body in the country of the requester. ISO copyright office Case postale 56 CH-1211 Geneva 20 Tel. + 41 22 749 01 11 Fax + 41 22 749 09 47 E-mail copyrightiso.org Web www.iso.org Published in Switzer
4、land ISO 29201:2012(E) ISO 2012 All rights reserved iii Contents Page Foreword . v Introduction .vi 1 Scope 1 2 Key concepts 1 2.1 Uncertainty of measurement 1 2.2 Estimation of the uncertainty of measurement . 1 2.3 Intralaboratory reproducibility . 2 2.4 Combined standard uncertainty 2 2.5 Relativ
5、e standard uncertainty 2 2.6 Relative variance . 3 2.7 Expanded uncertainty and expanded relative uncertainty 3 3 Microbiological methods . 4 3.1 Common basis . 4 3.2 Quantitative instruments . 4 3.3 Uncertainty structure . 4 3.4 Expression of combined uncertainty . 4 4 Choices of approach 5 4.1 Gen
6、eral . 5 4.2 Choices of evaluation approach . 6 4.3 Choices of expression and use of measurement uncertainty 7 5 The component approach to the evaluation of operational uncertainty 7 5.1 General . 7 5.2 Identification of the components of uncertainty . 7 5.3 Evaluation 7 6 The global approach to the
7、 determination of the operational uncertainty 8 6.1 General . 8 6.2 Evaluation 9 7 Combined uncertainty of the test result .10 7.1 Basic principle .10 7.2 Operational variability 10 7.3 Intrinsic variability .10 7.4 Combined uncertainty 10 7.5 Borderline cases 10 Annex A (informative) Symbols and de
8、finitions . 11 Annex B (normative) General principles for combining components of uncertainty 13 Annex C (normative) Intrinsic variability Relative distribution uncertainty of colony counts .18 Annex D (normative) Intrinsic variability of most probable number estimates .20 Annex E (normative) Intrin
9、sic variability (standard uncertainty) of confirmed counts .23 Annex F (normative) Global approach for determining the operational and combined uncertainties .26 Annex G (normative) Component approach to evaluation of the combined relative uncertainty under intralaboratory reproducibility conditions
10、 .31 Annex H (normative) Experimental evaluation of subsampling variance .35 Annex I (normative) Relative repeatability and intralaboratory reproducibility of volume measurements .38 Annex J (normative) Relative uncertainty of a sum of test portions 40 Annex K (normative) Relative uncertainty of dil
11、ution factor F 44 ISO 29201:2012(E) iv ISO 2012 All rights reserved Annex L (normative) Repeatability and intralaboratory reproducibility of counting .46 Annex M (normative) Incubation effects Uncertainty due to position and time .50 Annex N (informative) Expression and use of measurement uncertaint
12、y 55 Bibliography .61 ISO 29201:2012(E) Foreword ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International Standards is normally carried out through ISO technical committees. Each member bo
13、dy interested in a subject for which a technical committee has been established has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrot
14、echnical Commission (IEC) on all matters of electrotechnical standardization. International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2. The main task of technical committees is to prepare International Standards. Draft International Standards adopted b
15、y the technical committees are circulated to the member bodies for voting. Publication as an International Standard requires approval by at least 75 % of the member bodies casting a vote. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent ri
16、ghts. ISO shall not be held responsible for identifying any or all such patent rights. ISO 29201 was prepared by Technical Committee ISO/TC 147, Water quality, Subcommittee SC 4, Microbiological methods. ISO 2012 All rights reserved v ISO 29201:2012(E) Introduction Testing laboratories are required
17、to apply procedures for estimating uncertainty of measurement (see ISO/IEC 17025 5 ). Without such an indication, measurement results cannot be compared, either among themselves or with reference values (see ISO/IEC Guide 98-3:2008 7 ). General guidelines for the evaluation and expression of uncerta
18、inty in measurement have been elaborated by experts in physical and chemical metrology, and published by ISO and IEC in ISO/IEC Guide 98-3:2008. 7However, ISO/IEC Guide 98-3:2008 7does not address measurements in which the observed values are counts. The emphasis in ISO/IEC Guide 98-3:2008 7is on th
19、e “law of propagation of uncertainty” principle, whereby combined estimates of the uncertainty of the final result are built up from separate components evaluated by whatever means are practical. This principle is referred to as the “component approach” in this International Standard. It is also kno
20、wn as the “bottom-up” or “step-by-step” approach. It has been suggested that the factors that influence the uncertainty of microbiological enumerations are not well enough understood for the application of the component approach (see ISO/TS 19036:2006 6 ). It is possible that this approach underesti
21、mates the uncertainty because some significant uncertainty contributions are missed. Reference 19 shows, however, that the concepts of ISO/IEC Guide 98-3:2008 7are adaptable and applicable to count data as well. Another principle, a “black-box” approach known as the “top-down” or “global” approach,
22、is based on statistical analysis of series of repeated observations of the final result (see ISO/TS 19036:2006 6 ). In the global approach it is not necessary to quantify or even know exactly what the causes of uncertainty in the black box are. According to the global philosophy, once evaluated for
23、a given method applied in a particular laboratory, the uncertainty estimate may be reliably applied to subsequent results obtained by the method in the same laboratory, provided that this is justified by the relevant quality control data (EURACHEM/CITAC CG 4 10 ). Every analytical result produced by
24、 a given method thus should have the same predictable uncertainty. This statement is understandable against its background of chemical analysis. In chemical analyses the uncertainty of the analytical procedure and the uncertainty of the final result of analysis are usually the same. The global princ
25、iple dismisses the possibility that there might be something unique about the uncertainty of a particular analysis. The uncontrollable “variation without a cause” that always accompanies counts alters the situation for microbiological enumerations. The full uncertainty of a test result can be estima
26、ted only after the final result has been secured. This applies to both the global and the component approaches. The unpredictable variation that accompanies counts increases rapidly when counts get low. The original global design is therefore not suitable for low counts, and therefore also not appli
27、cable to most probable number (MPN) methods and other low-count applications, such as confirmed counts. It is often necessary, and always useful, to distinguish between two precision parameters: the uncertainty of the technical measuring procedure (operational variability), which is more or less pre
28、dictable, and the unpredictable variation that is due to the distribution of particles. A modification of the global principle that takes into account these two sources of uncertainty is free from the low-count restriction. This is the global model detailed in this International Standard. In theory,
29、 the two quantitative approaches to uncertainty should give the same result. A choice of two approaches is presented in this International Standard. Offering two approaches is appropriate not only because some parties might prefer one approach to the other. Depending on circumstances one approach ma
30、y be more efficient or more practical than the other. Neither of the main strategies is, however, able to produce unequivocal estimates of uncertainty. Something always has to be taken for granted without the possibility of checking its validity in a given situation. The estimate of uncertainty is b
31、ased on prior empirical results (experimental standard uncertainties) and/or reasonable general assumptions. vi ISO 2012 All rights reserved Water quality The variability of test results and the uncertainty of measurement of microbiological enumeration methods 1 Scope This International Standard giv
32、es guidelines for the evaluation of uncertainty in quantitative microbiological analyses based on enumeration of microbial particles by culture. It covers all variants of colony count methods and most probable number estimates. Two approaches, the component (also known as bottom-up or step-by-step)
33、and a modified global (top-down) approach are included. The aim is to specify how values of intralaboratory operational variability and combined uncertainty for final test results can be obtained. The procedures are not applicable to methods other than enumeration methods. NOTE 1 Most annexes are no
34、rmative. However, only the annexes relevant to each case are to be applied. If the choice is the global approach, then all normative annexes that belong to the component approach can be skipped and vice versa. NOTE 2 Pre-analytical sampling variance at the source is outside the scope of this Interna
35、tional Standard, but needs to be addressed in sampling designs and monitoring programmes. NOTE 3 The doubt or uncertainty of decisions based on the use of analytical results whose uncertainty has been estimated is outside the scope of this International Standard. NOTE 4 The extra-analytical variatio
36、ns observed in proficiency tests and intercalibration schemes are also not detailed in this International Standard, but it is necessary to take them into consideration in analytical control. The use of intercalibration data in uncertainty estimation offers the possibility for the bias between labora
37、tories to be included (Nordtest Report TR 537 12 ). 2 Key concepts 2.1 Uncertainty of measurement Uncertainty of measurement according to ISO/IEC Guide 98-3:2008 7is defined as a “parameter, associated with the result of measurement, that characterizes the dispersion of the values that could reasona
38、bly be attributed to the measurand”. It is a measure of imprecision. The parameter is expressed as a standard uncertainty or relative standard uncertainty. 2.2 Estimation of the uncertainty of measurement According to ISO/IEC Guide 98-3:2008, 7the parameter can be evaluated by statistical analysis o
39、f series of observations. This is termed type A estimation of uncertainty. Any other type of procedure is called type B estimation of uncertainty. The most common type B estimates in microbiological analysis are those based on assumed statistical distributions in the component approach. Types A and
40、B may refer to the uncertainty of individual components of uncertainty as well as to the combined uncertainty of the final result. Type A evaluations of standard uncertainty are not necessarily more reliable than type B evaluations. In many practical measurement situations where the number of observ
41、ations is limited, the components obtained from type B evaluations can be better known than the components obtained from type A evaluations (ISO/IEC Guide 98-3:2008 7 ). INTERNATIONAL STANDARD ISO 29201:2012(E) ISO 2012 All rights reserved 1 ISO 29201:2012(E) 2.3 Intralaboratory reproducibility A so
42、mewhat abstract expression of uncertainty, intralaboratory reproducibility, is frequently considered the most appropriate parameter of the uncertainty of measurement, see ISO/TS 19036:2006. 6I t i s a l s o known as intermediate reproducibility or intermediate precision, e.g. time + equipment + oper
43、ator-different intermediate precision standard uncertainty as defined by ISO 5725-3. 2The idea is to evaluate how much the analytical result might have varied if the analysis had been made by another person in the same laboratory using different equipment and batches of material and different analyt
44、ical and incubation conditions than those actually employed. The value of intermediate precision estimated never belongs to any actual analytical result, but is assumed to give a general estimate of reasonable uncertainty for the application of a method in one particular laboratory. Intralaboratory
45、reproducibility is estimated either by combining separate components of uncertainty determined under intralaboratory reproducibility conditions (component approach) or by special experiments in which the analytical conditions are varied by design (global approach). 2.4 Combined standard uncertainty
46、2.4.1 General The final test results of microbiological analyses are calculated from intermediate observed values. The main intermediate observation is the count. Most of the other observed values are connected with volume measurements. Combined standard uncertainty, as defined in ISO/IEC Guide 98-3
47、:2008, 7is the “standard uncertainty of the result of a measurement when that result is obtained from the values of a number of other quantities, equal to the positive square root of a sum of terms, the terms being variances or covariances of these other quantities weighted according to how the meas
48、urement result varies with changes in these quantities”. NOTE 1 Observation of covariances is only necessary if significant correlations occur between components of uncertainty. Otherwise a simple root sum of variances is sufficient (see 2.4.2 and 2.5). NOTE 2 In cases of microbiological enumeration
49、, it can be assumed that all components of uncertainty are independent, i.e. statistically uncorrelated. In such instances, the combined standard uncertainty is the positive square root of the sum of component variances, i.e. the root sum of squares (Annex B). (ISO/IEC Guide 98-3:2008. 7 ) 2.4.2 Signific ant property of combined uncertainties According to EURACHEM/CITAC CG 4 10 , “Unless there is a large number of them, components (standard uncertainties) that are less than one-third of the largest need not
