According to API Standard 936, when a sample fails to meet the required specifications, retesting is permitted to validate the result. To ensure the retest is scientifically valid and comparable to the original failed test, variables must be minimized. The standard explicitly states that retesting shall be performed using specimens drawn from the same batch, same pallet, and the same sample (the original quantity of material extracted) as the original test. Furthermore, the same testing procedure (methodology) must be utilized to ensure the results are correlated. Inspection methods (which typically refer to visual field monitoring or installation oversight) are distinct from the specific laboratory testing protocols and are not a variable that applies to the retesting of a specific material sample.

API Standard 936, Refractory Installation Quality Control—Inspection and Testing Monolithic Refractory Materials, Section: Testing and Quality Control (specifically subsections on Sampling and Retesting/Resampling).

Note: The standard dictates that to confirm compliance or failure, the re-evaluation must rule out material variation; thus, the original source material (Batch/Pallet/Sample) and the yardstick (Testing Procedure) must remain constant.

API 936 Body of Knowledge, Quality Control - Testing Variables, American Petroleum Institute.