Industrial Automatic Control Systems and Controllers Annotation << Back Factor planning in sequential optimization A.A. Moiseev The article discusses the method of two-level experiment planning, based on the use of saturated plans, including, in particular, fictitious factors. The use of this method makes it possible to investigate the reproducibility of triple measurements according to the Cochrane criterion, the signifi cance of the introduced factors according to the Student’s criterion, the completeness of the system of introduced factors according to the Fisher criterion, and to carry out sequential optimization of the parameters corresponding to signifi cant factors during qualifi cation tests. Based on the results of planned experiments, a linear-regression model is formed, the use of which allows for statistical analysis, as well as optimization using linear programming methods. Local extrema of the linear regression model are interpreted as the centers of the next search areas. The consideration of the optimization of industrial rubber goods by the criterion of chemical stability, carried out in the article, is just one example of the application of the developed method of factorial optimization. A similar approach can, in particular, be used to solve the problems of choosing extreme conditions for qualifi cation tests for wear and to optimize the fuel composition by the criterion of thermal oxidative stability. Keywords: experiments planning; factor plan; regression model; fictive factor; Student criterion; Fisher criterion; Cochrane criterion; linear programming; simplex. DOI: 10.25791/asu.9.2020.1221 Pp. 53-63.