ISO 11843-2 pdf download – Capability of detection 一Part 2: Methodology in the linear calibration case

ISO 11843-2 pdf download – Capability of detection 一Part 2: Methodology in the linear calibration case.
4 Experimental design
4.1 General
The procedure for determining values of an unknown actual state includes saming, preparation and the measurement itself, As every step of this procedure may produce distortion, it is essential to apply the same procedure tor characterizing, for use in the preparation and determinahori of the values of the unknown actual state, for all reference states and for the basic state used for calibration.
For the purpose of determining differences between the values characterizing one or more unknown actual states and the basic state. it is necessary to choose an expenmenlal design suited for comparison. The experimental units of such an experiment are obtained from the actual states to be measured and all reference states used for calibration. An ideal design would keep constant all factors known to influence the outcome and control of unknown lactors by providing a randomized order to prepare and perform the measurements.
In reabty It may be deficult to proceed In such a way, as the preparations and determination of the values of the states involved are performed consecutively over a period of time, However, in order to detect major biases changing with time. it is strongly recommended to perform one half of the calibration before and one half alter the measurement of the unknown states. However, this is only possible ii the size of the measurement series is known in advance and if there is sufficient time to fo$ow this approach. Ii it is not possible to control all influencing factors, conditional statements containing all unproven assumptions shall be presented.
Many measurement methods require a chemical or physical treatment of the sample prior to the measurement itself. Both ci these steps of the measurement procedure add variation to the measurement results. lilt is requwed to repeat measurements the repetition consists In a full repetition of the preparation and the measurement. However, in many situations the measurement procedure is not repeated fully, in particular not all of the preparational steps are repealed for each measurement; see note in 5.2.1.
NOTE The formulae for the crital values and the minimum detectable value In clause 5 are only valid under the assumption that the nuner of repealed measurements per preparation Is Identical for all measurements of reference states and actual stales.
As the variations and cost due to the preparation usually will be much higher than those due to the measurement, the optimal choice of J, K and L may be derived troni an optimization of constraints regarding variation and costs.
5 The critical values and . and the minimum detectable value ,r of a measurement
series
5.1 Basic assumptions
The I llOwin picockire Icr the computation of the critical values and the minimum detectable value iu luu the assumptions of ISO 1l095 The methods of ISO 11095 are used with one generalization; see 5.3.
Basic assumptions of ISO 11095 are that
— the calibration function is linear.
measurements of the response vanable of all preparations and reference states are assumed to be independent and normally distributed wilt, standard deviation referred to as ‘residual standard deviation,
— the residual slandard deviation is either a constant, i.e. ii does not depend on the values of the net state vanable [case 1]. or it forms a linear function of the values of the net state variable (case 2).
The decision regarding the applicability of this part of ISO 11843 and the oice of one of these two cases should be based on prior knowledge and a visual examination of the data.