1、BSI Standards Publication Guidance on how to conduct Round Robin Tests PD CLC/TR 50619:2014National foreword This Published Document is the UK implementation of CLC/TR 50619:2014. The UK participation in its preparation was entrusted to Technical Committee CPL/59, Performance of household electrical
2、 appliances. A list of organizations represented on this committee can be obtained on request to its secretary. This publication does not purport to include all the necessary provisions of a contract. Users are responsible for its correct application. The British Standards Institution 2014. Publishe
3、d by BSI Standards Limited 2014 ISBN 978 0 580 83408 0 ICS 03.120.20; 19.020 Compliance with a British Standard cannot confer immunity from legal obligations. This Published Document was published under the authority of the Standards Policy and Strategy Committee on 30 June 2014. Amendments/corrigen
4、da issued since publication Date Text affected PUBLISHED DOCUMENT PD CLC/TR 50619:2014 TECHNICAL REPORT CLC/TR 50619 RAPPORT TECHNIQUE TECHNISCHER BERICHT February 2014 CENELEC European Committee for Electrotechnical Standardization Comit Europen de Normalisation Electrotechnique Europisches Komitee
5、 fr Elektrotechnische Normung CEN-CENELEC Management Centre: Avenue Marnix 17, B - 1000 Brussels 2014 CENELEC - All rights of exploitation in any form and by any means reserved worldwide for CENELEC members. Ref. No. CLC/TR 50619:2014 E ICS 03.120.20; 19.020 English version Guidance on how to conduc
6、t Round Robin Tests Guide de conduite des essais Round Robin Leitfaden zur Durchfhrung von Ringversuchen This Technical Report was approved by CENELEC on 2014-01-24. CENELEC members are the national electrotechnical committees of Austria, Belgium, Bulgaria, Croatia, Cyprus, the Czech Republic, Denma
7、rk, Estonia, Finland, Former Yugoslav Republic of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and the United Kingdom. PD CLC/TR 5
8、0619:2014CLC/TR 50619:2014 - 2 - Contents Page Foreword 3 Introduction . 4 1 Scope 5 2 Normative references 5 3 Terms and definitions . 5 4 Process and responsibilities 6 4.1 Process . 6 4.1.1 Product to be tested 6 4.1.2 Parameters to be tested 6 4.1.3 Measurement procedure . 6 4.1.4 RRT procedure
9、. 7 4.2 Responsibilities . 7 4.2.1 Contracting body . 7 4.2.2 Coordinator 7 4.2.3 Subcontractor 7 4.2.4 Financing of RRT . 7 5 Testing laboratories 8 5.1 Potential laboratories 8 5.2 Announcement . 8 5.2.1 General 8 5.2.2 Questionnaire . 8 5.2.3 Assessment of selection of laboratories 8 5.3 Selectio
10、n of laboratories . 8 5.4 Final list of laboratories 8 6 Transportation of the product 9 6.1 Logistics . 9 6.2 Packaging . 9 7 Test 9 7.1 Execution of test 9 7.2 Laboratory visit 9 7.3 Transmission of result 9 8 Analysis, report and termination . 10 8.1 Analysis 10 8.2 Report . 10 8.2.1 Draft report
11、 . 10 8.2.2 Content of the internal report . 10 8.3 Termination and publication of final external report . 10 Bibliography 11 PD CLC/TR 50619:2014 - 3 - CLC/TR 50619:2014 Foreword This document (CLC/TR 50619:2014) has been prepared by CLC/TC TC59X “Performance of household and similar electrical app
12、liances“. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. CENELEC and/or CEN shall not be held responsible for identifying any or all such patent rights. PD CLC/TR 50619:2014CLC/TR 50619:2014 - 4 - Introduction It is the responsib
13、ility of each standardisation committee under TC 59X to establish the repeatability and reproducibility of the measurement standards developed. Results from inter-laboratory comparisons are important for a) identification of interlaboratory differences; b) establishment of the effectiveness and comp
14、arability of test or measurement methods; c) validation of uncertainties; d) evaluation of the performance of laboratories for specific tests or measurements and monitoring laboratories continuing performance; e) identification of problems in laboratories and initiation of actions for improvement wh
15、ich, for example, may be related to inadequate test or measurement procedures, effectiveness of staff training and supervision, or calibration of equipment; and f) education of participating laboratories based on the outcomes of such comparisons. The need for ongoing confidence in laboratory perform
16、ance is not only essential for laboratories and their contractors but also for other interested parties, such as regulators, laboratory accreditation bodies and other organisations that specify requirements for laboratories. EN ISO/IEC 17011 requires accreditation bodies to take account of laborator
17、ies participation and performance in proficiency testing. With this respect round robin testing was widely made in the past by TC59X for development of measurement procedures on purpose of EU regulatory measures on Labeling and Ecodesign. Round robin test results have been widely taken into account
18、in the establishment of regulations, in defining tolerance levels for verification of declared values and/or limits. This Technical Report is intended to provide a consistent basis for performing round robin testing. It gives guidance to all interested parties to determine the competence among each
19、other. It provides common ground for reliable statistical data (repeatability and reproducibility levels, etc.) as needed for regulation purposes (like for Labeling and Ecodesign). PD CLC/TR 50619:2014 - 5 - CLC/TR 50619:2014 1 Scope This Technical Report provides guidance for carrying out round rob
20、in tests (RRT) and hence for the determination of levels of repeatability (intra-laboratory variability) and reproducibility (inter-laboratory variability). This Technical Report can also be used to verify the measurement methods, to improve the measurement method, and to qualify laboratories. It is
21、 not applicable for the determination of production variation for a particular product. General advice on proficiency testing of laboratories is given in EN 17043. This Technical Report can be used in addition to this document. NOTE The repeatability and reproducibility levels are important factors
22、for the establishment of uncertainty margins of the measurement methods and for the definition of tolerances levels in verification schemes. 2 Normative references The following documents, in whole or in part, are normatively referenced in this document and are indispensable for its application. For
23、 dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies EN ISO/IEC 17043:2010, Conformity assessment General requirements for proficiency testing (ISO/IEC 17043) IEC/TR 61923, Household electrical app
24、liances Method of measuring performance Assessment of repeatability and reproducibility 3 Terms and definitions For the purposes of this document, the terms and definitions given in EN ISO/IEC 17043:2010 and the following apply. 3.1 contracting body organisation or individual for which a round robin
25、 testing is provided through a contractual arrangement SOURCE: EN ISO/IEC 17043:2010, definition 3.3, modified 3.2 repeatability precision under repeatability conditions Note1 to entry: Repeatability includes the variability of the appliance under test. SOURCE: ISO 3534-1, definition 3.15 3.3 reprod
26、ucibility precision under reproducibility conditions Note 1 to entry: Reproducibility includes repeatability SOURCE: ISO 3534-1, definition 3.20 3.4 round robin testing process in which one or more items is tested according to a specific protocol by a number of different laboratories PD CLC/TR 50619
27、:2014CLC/TR 50619:2014 - 6 - Note 1 to entry: It is the intention to derive levels of repeatability and reproducibility and hence making classifications of laboratories. Note 2 to entry: An alternative term is “ring testing”. The term “proficiency testing” is more general as covers other aspects tha
28、n derivation of repeatability and reproducibility values SOURCE: EN ISO/IEC 17043:2010, definition 3.7. Note 3 to entry: It may be used for the evaluation of laboratory performance against pre-established criteria. 4 Process and responsibilities 4.1 Process 4.1.1 Product to be tested The product cat
29、egory to be tested should be clearly specified and one or more representative products should be selected. If only one product is selected, it should be representative in the sense that it reflects the typical behaviour and performance of the defined product category. It is recommended that two or m
30、ore products be tested in order to get an indication on the measurement uncertainty across the range of performance that is likely to be encountered in the market. (For example, one product with high and one with low energy/water consumption or performance). The products to be used for the RRT can b
31、e selected either by pre-testing, pre-selection or special production. Additional sample(s) should be put aside as replacements in case any RRT samples are damaged. NOTE It may be appropriate to circulate key items of test equipment or other test objects together with the RRT samples. 4.1.2 Paramete
32、rs to be tested The parameters to be tested should be clearly defined. Usually the full parameter set of a measurement procedure should be covered (e.g. parameters as set out in EN 60456 washing machine like washing performance, consumption values, time, rinsing efficiency. Noise may be seen as a di
33、fferent measurement procedure). If repeatability and reproducibility of the measurement standard are to be assessed, all additional materials used for the testing should be as defined in the measurement procedure, and should not be specially selected. (Examples of such materials in the case of washi
34、ng machine tests would include test swatches, soils and detergent). If it is considered likely that the reproducibility of results could be affected by batch-to-batch variability within given tolerances of the test material this can form part of the RRT. In this case, different batches of material w
35、ould need to be included in the test design of the RRT. 4.1.3 Measurement procedure The measurement procedure(s) to be used for testing should be clearly defined. Preferably RRTs should be based on published EN standards. For RRTs supporting standardisation work it may be necessary to use working pa
36、pers like prEN or even DC documents as the measurement procedure. Alternatively, national, international and/or industrial standards may be used and where necessary, combinations of more than one measurement method can be used. It is essential therefore that clear instructions are given on which ver
37、sion of a measurement standard the RRT is based. If deviations between the standard used for the RRT and a later published standard exist, care shall be taken in interpretation of the results of the RRT in relation to the final published standard. PD CLC/TR 50619:2014 - 7 - CLC/TR 50619:2014 4.1.4 R
38、RT procedure The RRT procedure is based on sending one or more products to different laboratories to be tested according to the defined measurement procedure(s). Therefore, the same sample of the product(s) shall be forwarded from laboratory to laboratory (serial procedure). Care should be taken to
39、ensure no damage or change (e.g. aging) occurs to the product during this process. Damage can be limited by ensuring the sample is properly packaged and that it is transported from laboratory to laboratory by a carrier with a proven track record of transporting delicate items. The aging effect may b
40、e reduced using the parallel procedure described below but in any case, the extent of any aging should wherever possible be determined by re-testing the same product again in the first laboratory after it has gone through all the testing in the other laboratories. As a second best alternative, more
41、than one sample of the product selected is sent in parallel to the laboratories (parallel procedure). In this case, before the RRT begins, it should be demonstrated that all samples show similar behaviour under test. Again, the samples should wherever possible, be re-tested by the same laboratory at
42、 the end of the RRT to check for any change in behaviour under test. For calculating reliable figures of the repeatability and reproducibility, a minimum of five laboratories should participate in an RRT. If more than 10 laboratories take part in an RRT, a parallel procedure combined with a serial p
43、rocedure may be advisable for time reasons (e.g. samples showing similar behaviour are each sent to five laboratories). 4.2 Responsibilities 4.2.1 Contracting body The body(ies) which is (are) running and/or financing the RRT should operate in a transparent manner towards all involved parties. The c
44、ontracting body should appoint an individual or a team to coordinate the RRT. 4.2.2 Coordinator The coordinator should be on call throughout the testing phase. If only one person is coordinating the RRT he/she should be neutral and independent. For example, they should not be employed by a product m
45、anufacturer or a test house involved in the RRT. NOTE Deviations from this rule may be necessary in individual cases. Any deviations from the recommendations of this document should be documented by the coordinator and included in the final internal report. 4.2.3 Subcontractor Subcontractors (collab
46、orators) in RRT should inform the coordinator regularly about the progress. The coordinator should be informed about unforeseeable proceedings without any delay. 4.2.4 Financing of RRT The coordinator should establish and agree with the contracting body the financing of the RRT. All costs (e.g. coor
47、dinator, subcontractor, product samples to be tested, logistics, test materials, testing) for the RRT are to be defined and agreed before recruiting participant laboratories. PD CLC/TR 50619:2014CLC/TR 50619:2014 - 8 - 5 Testing laboratories 5.1 Potential laboratories A list of potential laboratorie
48、s should be prepared by the coordinator, in consultation with the contractor and relevant standardisation bodies. 5.2 Announcement 5.2.1 General The round robin test should be announced publically together with the qualification criteria as defined in document “Qualification of laboratories for veri
49、fication procedures” (e.g. tender procedure, TC59X- working group procedure) so that all laboratories that might be interested have the chance to apply to take part in the RRT. NOTE Potential laboratories may be contacted directly. 5.2.2 Questionnaire The announcement should be accompanied by a questionnaire where the general qualification of the laboratory (equipment, experience, certification, qualification) should be assessed. The laboratories are asked to answer within a fixed time frame. If there are minimum qualification criteria given to take part in th
