【国外标准】 Standard Practice for Utilization of Test Data to Determine Conformance with Specifications

4.1 This practice provides a means whereby the parties can resolve disputes over specification conformance for those product properties which can be tested and expressed numerically.4.1.1 This practice can be used to ensure that such properties are correctly stated on labels or in other descriptions of the product.4.1.2 This practice can be implemented in those cases where a supplier uses an in-house or a commercial testing laboratory to sample and test a product prior to releasing the product to a shipper (intermediate receiver) and the ultimate receiver also uses an in-house or commercial testing laboratory to sample and test the product upon arrival at the destination. The ATV would still be determined according to 8.3.4.2 This practice can be applied in the determination of tolerances from specification limits based on a mutually agreed probability between parties for making the conformance to specification decision if the true value of a property is sufficiently close to the specification limit. Such tolerances are bounded by an acceptance limit (AL). If the ATV value determined by applying this practice falls on the AL or on the acceptable side of the AL, the product property can be considered to have met the specification; otherwise it shall be considered to have failed to meet the specification.4.3 Application of this practice requires the AL be determined prior to actual commencement of testing. Therefore, the degree of criticality of the specification, as determined by the Probability of Acceptance (P value) that is required to calculate the AL, shall have been mutually agreed upon between both parties prior to execution of actual product testing.4.3.1 This agreement should include a decision as to whether the ATV is to be determined by the absolute or rounding-off method of Practice E29, as therein defined.4.3.1.1 If the rounding-off method is to be used, the number of significant digits to be retained must also be agreed upon.4.3.1.2 These decisions must also be made in the case where only one party is involved, as in the case of a label.4.3.1.3 In the absence of such an agreement, this practice recommends the ATV be rounded in accordance with the rounding-off method in Practice E29 to the number of significant digits that are specified in the governing specification.4.4 This practice is designed to be suitable for reference in contracts governing the transfer of petroleum products and lubricants from a supplier to a receiver.4.5 As a prerequisite for acceptance for lab test results to be used in this practice, the following conditions shall be satisfied:4.5.1 Site precision (R′) as defined in Practice D6299 for the appropriate test method(s) from each lab, as substantiated by control charts meeting the requirement of D6299 from in-house quality control programs, for property typical of the product in dispute, should have a TPI > 1.2 for methods with Precision Ratio <4 and TPI > 2.4 for methods with Precision Ratio ≥4 (see Practice D6792 for TPI explanation).4.5.2 Each lab shall be able to demonstrate, by way of results from interlaboratory exchange programs, a lack of a systemic bias relative to exchange averages for the appropriate test method(s) as per methodology outlined in Guide D7372.4.5.3 In the event that the site precision of laboratories from two parties are statistically different as confirmed by the F-test (see Annex A4), then, for the purpose of establishing the ATV, each laboratory's test result shall be inversely weighted in accordance with laboratory's demonstrated variance.4.6 It is recommended that this practice be conducted under the guidance of a qualified statistician.1.1 This practice covers guidelines and statistical methodologies with which two parties (see Note 1) can compare and combine independently obtained test results to obtain an Assigned Test Value (ATV) for the purpose of resolving a dispute over product property conformance with specification.NOTE 1: Application of this practice is usually, but not limited to, between supplier and receiver of a product.1.2 This practice defines a technique for establishing an Acceptance Limit (AL) and Assigned Test Value (ATV) to resolve the dispute over a property conformance with specification by comparing the ATV to the AL.1.3 This practice applies only to those test methods which specifically state that the repeatability and reproducibility values conform to the definitions herein.1.4 The statistical principles and methodology outlined in this practice can also be used to obtain an ATV for specification conformance decision when multiple results are obtained for the same batch of product within a single laboratory. For this application, site precision (R') as defined in Practice D6299 shall be used in lieu of test method published reproducibility (R).1.5 This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.