Decision-making methods based on binary comparison in the absence of a priori knowledge of the systematic errors are considered. Using simulation to find the probability characteristics in the conventional form for different distributions of additional error are described. Influence the choice of a threshold level of systematic errors difference in the result of the procedure are studied
Keywords-comparison, standards of measurement, not excluded systematic error, metrological analysisThe final cause of binary comparison is to establish a compliance of compared standards to the relevant requirements. This requires a corresponding rule that allows taking a decision on the basis of comparison results. Simulation can be used to select the optimal thresholds of the results of comparisons as well as to confirm the results of analytical calculations. It should be noted that the same uses a priori knowledge that uses during performing calculations on analytical basis. However, the absence of a priori information about properties of systematic errors comparison standards necessitates the use of conditional probability characteristics. This means that the simulation generated samples corresponding to a fixed value of the result of comparison, using them and set the conditional probability characteristics. The primary object in this case is the choice of software product that provides the best solution to the problem of a particular experiment. The number of set parameters, their character, quality characteristics, output format, recording, graphical display are possible selection criteria.
