Implementation of traceability in interlaboratory comparisons of the measurement results of human blood haematological parameters processed jointly through the least squares method
摘要
Imagine a worldwide experiment consistently conducted across multiple international laboratories, where blood parameters from one individual were tested on devices from either the same or different manufacturers. Theoretically, the result should be the same on different devices. However, in practice, interlaboratory discrepancies remain persistent. The question remains whether such experiments have yielded their maximum potential information. In reality, the implementation of these experiments was limited by the absence of a clear mathematical model for the processing and optimal use of such comparative measurements, as well as the lack of an organizational structure for their performance. The proposed measurement model enables the joint processing of any measurement combination of any human blood samples, certified and non-certified reference materials, manufacturers’ control samples, etc., during international and interlaboratory comparisons of measurement results. It is proposed that measurements be conducted in accordance with the Metrological Measurement Network (MMN) scheme, which has significant advantages over the circular and radial schemes. In the absence of certified reference materials (CRMs) or reference methods, it is proposed to average the reference numerical axis based on the measurement results made by a large number of working devices from all manufacturers. This will provide a local reference, relative to which the measurement biases of each device will be determined. This concept represents the core contribution of this research, for processing interlaboratory comparisons measurement results, specifically within the domain of human blood haematological parameters.