Uncertainty of measurement and reliability of the decision making on compliance

Abstract

It is shown that the probability of making the right decision on the compliance of the object with the norms depends on the measurement uncertainty associated with the result of the measurement control of the object parameters and the length of the tolerance interval. The decisive rule on the suitability (unfitness) of the object of control should take into account both the possible uncertainty of the result and the requirements for the cost and complexity of the control procedure, as well as to the specified reliability. To improve the reliability of this procedure, a control method based on a sequential adaptive decision-making procedure is proposed, which takes into account the relationship of the parameters of the distribution law of the possible values of the controlled quantity, random effects in the measurement and the length of the tolerance interval. The decision-making procedure assumes consistently, depending on the result at the current stage, to introduce additional tolerance intervals, the length of which is determined by the parameters of the distribution function of random variables accompanying the measurements of the controlled parameters. The result of measurement of the controlled parameter is compared with these intervals, and decisions on continuation or completion of control with introduction of additional limits are made. This provides a reduction in the probability of erroneous decisions at each additional stage of control, since the time of the initial hit of the controlled value in a consistently calculated controlled interval is determined step by step. The use of a sequential adaptive procedure provides a given reliability of control and does not allow to carry out the control procedure in full for all possible values of the controlled value.