A new method for controlling the quality of technological products

Abstract

The problem of conformity of products to the requirements of technical control or between products of different batches puts the methods of experimental determination of product properties high on the agenda. The measurement of the parameters of these properties is stochastic. Some product properties can only be determined in specially designed experiments. Thus, in the production of powders, their dispersion can be normalized, the specific charge of condenser powders can increase, the location of magnetic granules of nanoparticles on information carriers can be controlled, the content of physiologically active substances in drugs can be controlled, and the spectrum of mobilities of atmospheric particles can be analyzed. Possible approaches to solving the problem of checking a batch of products for compliance with the requirements of technical specifications (TS) are considered using the example of powder dispersion. For the first time, a method has been proposed for comparing product properties, which are continuous random variables, by determining the joint area under their differential curves. The decision on the conformity of the measured product properties to the requirements of technical specifications is based on a comparative assessment with the directive established joint area under the distribution density curves. The technique of determining the probability of correspondence of normal distribution laws to each other is described. A final analytical solution is obtained for two exponential distribution laws.