ISO 4259 Petroleum products - Determination and application of precision data in relation to methods of test
4 Stages in the planning of an inter-laboratory test programme for the determination of the precision of a test method
4.1 General
The stages in planning an inter-laboratory test programme are as follows:
a) preparing a draft method of test;
b) planning a pilot programme with at least two laboratories;
c) planning the inter-laboratory programme;
d) executing the inter-laboratory programme.
The four stages are described in turn in 4.2 to 4.5.
4.2 Preparing a draft method of test
This shall contain all the necessary details for carrying out the test and reporting the results. Any condition that could alter the results shall be specified.
A clause on precision is included in the draft method of the test at this stage only as a heading. It is recommended that the lower limit of the scope of the test method is not less than the region of the lowest value tested in the inter-laboratory programme, and is at least 2R greater than the lowest achievable result (see 8.2), where R is the reproducibility estimate. Similarly, it is recommended that the upper limit of the scope of a test method is not greater than the region of the highest value tested in the inter-laboratory programme, and is at least 2R less than the highest achievable result.
4.3 Planning a pilot programme with at least two laboratories
A pilot programme is necessary for the following reasons:
a) to verify the details in the operation of the test;
b) to find out how well operators can follow the instructions of the method;
c) to check the precautions regarding samples;
d) to estimate approximately the precision of the test.
At least two samples are required, covering the range of results to which the test method is intended to apply; however, at least twelve laboratory/sample combinations shall be included. Each sample is tested twice by each laboratory under repeatability conditions. If any omissions or inaccuracies in the draft test method are revealed, they shall now be corrected. The results shall be analysed for bias and precision; if either is considered to be too large, then alterations to the test method shall be considered.
4.4 Planning the inter-laboratory programme
There shall be at least five participating laboratories, but it is preferable that there are more in order to reduce the number of samples required.
The number of samples shall be sufficient to cover the range of the property measured at approximately equidistant intervals and to give reliability to the precision estimates. If precision is found to vary with the level of results in the pilot programme, then at least five samples shall be used in the inter-laboratory programme. In any case, it is necessary to obtain at least 30 degrees of freedom in both repeatability and reproducibility. For repeatability, this means obtaining a total of at least 30 pairs of results in the programme.
For reproducibility, Table A.1 gives the minimum number of samples required in terms of L, P and Q, where L is the number of participating laboratories, and P and Q are the ratios of variance component estimates obtained from the pilot programme. Specifically, P is the ratio of the interaction component to the repeats component and Q is the ratio of the laboratories component to the repeats component. Annex B gives the derivation of the equation used. If Q is much larger than P, then 30 degrees of freedom cannot be achieved; the blank entries in Table A.1 correspond to, or an approach to, this situation (i.e. when more than 20 samples are required). For these cases, there is likely to be a significant bias between laboratories.
4.5 Executing the inter-laboratory programme
One person shall be responsible for the entire programme, from the distribution of the texts of the test method and samples to the final appraisal of the results. He shall be familiar with the test method, but shall not personally take part in the tests.
The text of the test method shall be distributed to all the laboratories in time to allow any queries to be raised before the tests begin. If any laboratory wants to practice the method in advance, this shall be carried out with samples other than those used in the programme.
The samples shall be accumulated, subdivided and distributed by the organizer, who shall also keep a reserve of each sample for emergencies. It is most important that the individual laboratory portions be homogeneous. They shall be blind coded before distribution and the following information shall be sent with them:
a) agreed (draft) method of test;
b) handling and storage requirements for the samples;
c) order in which the samples are to be tested (a different random order for each laboratory);
d) statement that two results shall be obtained consecutively on each sample by the same operator with the same apparatus. For statistical reasons, it is imperative that the two results are obtained independently of each other, that is, that the second result is not biased by knowledge of the first. If this is regarded as impossible to achieve with the operator concerned, then the pairs of results shall be obtained in a blind fashion, but ensuring that they are carried out in a short period of time;
e) period of time during which repeated results are to be obtained and the period of time during which all the samples are to be tested;
f) blank form for reporting the results. For each sample, there shall be space for the date of testing, the two results, and any unusual occurrences. The unit of accuracy for reporting the results shall be specified;
g) statement that the test shall be carried out under normal conditions, using operators with good experience but not exceptional knowledge and that the duration of the test shall be the same as normal.
The pilot-programme operators may take part in the inter-laboratory programme. If their extra experience in testing a few more samples produces a noticeable effect, it serves as a warning that the test method is not satisfactory. They shall be identified in the report of the results so that any effect can be noted.
