ANALYSIS AND DATA PROCESSING SYSTEMS

The issues related to the use of expert assessment methods in order to increase the optimality and objectivity of developed management decisions are considered. The expediency of involving experts is formulated when a collegial management decision is necessary. The basic stages of the research related to the application of the expert assessment method are given. The relevance of the chosen topic is justified by the need to develop effective methods and algorithms aimed at automating the processes of analysis and processing of data obtained during the expert survey, as well as summarizing its results. The aim of the study is to develop and test an algorithm for finding coordinated groups (coalitions) of experts, as well as a method that makes it necessary to work out the final decision on the results of the expert survey. The proposed algorithm for finding coalitions is based on the application of the mathematical apparatus of the theory of sets and algebra of relations; it uses a strict criterion in their formation and replenishment of coalitions, but does not exclude the belonging of experts to several coalitions, which provides a high informative value of the results of expert assessment. The method includes the calculation of weighting coefficients assigned to the formed coalitions taking into account the degrees of experts' membership in them.  The resulting decision is formed on the basis of the data obtained by the coalition, which has the maximum dimensionality and the required degree of consistency. The developed method and algorithm were tested on the results of an expert survey for determining the degrees of significance of the processes implemented in communication networks. As a result, coalitions of experts who agreed with each other were formed, as well as the experts, whose opinions are more deviated from the rest, were identified. The coherence of the coalitions is evaluated using the concordance coefficient, and the final decision is formed on the basis of the method of average arithmetic ranks assigned to the studied processes of functioning of communication networks by the experts belonging to the largest coalition.

1. Orlov A.I. Teoriya prinyatiya reshenii [Theory of Decision Making]. Moscow, Mart Publ., 2004. 656 p.

2. Mikoni S.V. Teoriya prinyatiya upravlencheskikh reshenii [Theory of managerial decision-making]. St. Petersburg, Lan' Publ., 2015. 448 p.

3. Demin G.A. Metody prinyatiya upravlencheskikh reshenii [Methods of managerial decision-making]. Perm, Perm State University Publ., 2019. 88 p.

4. Isaeva M.K. [Decision-making methods based on expert evaluations of machine experiments]. Strategicheskoe planirovanie i razvitie predpriyatii [Strategic planning and enterprise development]. Proceedings of the Nineteenth All-Russian Symposium. Moscow, 2018, pp. 288–290. (In Russian).

5. Orlov A.I. Ekspertnye otsenki [Expert evaluations]. Moscow, 2002. 31 p.

6. Gorskii V.G., Gritsenko A.A., Orlov A.I. Metod soglasovaniya klasterizovannykh ranzhirovok [A reconciliation method for clustered rankings]. Avtomatika i telemekhanika = Automation and Remote Control, 2000, no. 3, pp. 159–167. (In Russian).

7. Madeeva V.A. [Automation of calculation of generalized opinion of experts and determination of the degree of agreement of experts' opinions]. Informatsionno-telekommunikatsionnye sistemy i tekhnologii [Information-telecommunications systems and technologies]. Proceedings of the All-Russian scientific-practical conference. Kemerovo, 2020, pp. 244–245. (In Russian).

8. Kuzmin I.E., Safronov A.A., Baranova E.M., Borzenkova S.Yu. Razrabotka sistemy vychisleniya stepeni soglasovannosti mnenii ekspertov v sfere informatsionnoi bezopasnosti metodom nakhozhdeniya koeffitsienta konkordatsii [Development of a computing system degree of consistency of expert opinions in the field of informa tion security by finding the concordance coefficient]. Izvestiya Tul'skogo gosudarstvennogo universiteta. Tekhnicheskie nauki = News of the Tula state university. Technical sciences, 2020, no. 5, pp. 11–18.

9. Sovertkov P.I. Formirovanie umenii po opredeleniyu mery soglasovannosti otsenok ekspertov na konkurse [Formation of skills in determining the measure of consistency of expert evaluations in the competition]. Severny region: nauka, obrazovanie, cultura, 2008, no. 1 (17), pp. 55–64. (In Russian).

10. Sudakov V.A., Osipov V.P., Kurennykh A.E. [Increasing the consistency of expert judgments in the matrix of pairwise comparisons]. Upravlenie razvitiem krupnomasshtabnykh sistem MLSD'2018 [Proceedings of 2018 11th International Conference "Management of Large-Scale System Development", MLSD 2018]. Moscow, 2018, pp. 346–352. (In Russian).

11. Saaty T. Prinyatie reshenii: metod analiza ierarkhii [Decision-making: analytic hierarchy process]. Moscow, Radio i svyaz' Publ., 1993. 278 p. (In Russian).

12. Bukachev D.S. Sposob otsenki soglasovannosti ekspertnykh dannykh v ekonomicheskikh informatsionnykh sistemakh [Method for assessing the consistency of expert data in economic information systems]. Nauka i biznes: puti razvitiya = Science and Business: Development Ways, 2018, no. 11 (89), pp. 47–50.

13. Totsenko V.G., Tsyganok V.V. Method of paired comparisons using feedback with expert. Journal of Automation and Information Sciences, 1999, vol. 31, no. 7–9, pp. 86–96. DOI: 10.1615/JAutomatInfScien.v31.i7-9.480.

14. Jin F., Cao M., Liu J., Martínez L., Chen H. Consistency and trust relationship-driven social network group decision-making method with probabilistic linguistic information. Applied Soft Computing, 2021, vol. 103, p. 107170. DOI: 10.1016/j.asoc.2021.107170.

15. Podvesovskii A., Mikhaleva O., Averchenkov V., Reutov A., Potapov L. A model of control of expert estimates consistency in distributed group expertise. Communications in Computer and Information Science, 2017, vol. 754 (5), pp. 361–374. DOI: 10.1007/978-3-319-65551-2_26.

16. Kuzmina N.M., Ridley A.N. Incomplete pairwise comparisons method for estimating the impact criteria for hub airports network optimization. Smart Innovation, Systems and Technologies, 2021, vol. 217, pp. 259–270. DOI: 10.1007/978-981-33-4826-4_18.

17. Kobzar' A.I. Prikladnaya matematicheskaya statistika [Applied mathematical statistics]. Moscow, Fizmatlit Publ., 2006. 816 p.

18. Simushkin S.V. Mnogomernyi statisticheskii analiz. Ch. 2 [Multivariate statistical analysis. Pt. 2]. Kazan, Kazan State University Publ., 2009. 114 p.

19. Popov A.V. Sravnenie metodov otsenki protsessov ergotekhnicheskikh sistem organov vnutrennikh del [Comparison of methods for assessing the processes of ergotechnical systems for internal affairs bodies]. Avtomatizatsiya protsessov upravleniya = Automation of Control Processes, 2022, no. 3 (69), pp. 67–76. DOI: 10.35752/1991-2927_2022_3_69_67.