MONITORING OF SOCIOTECHNICAL SYSTEM CLUSTERS BASED ON A FUZZY COGNITIVE APPROACH AND DYNAMIC CLUSTERING METHODS

UDC 004.89

Authors

Arbuzov Alexey D.
The Branch of National Research University "Moscow Power Engineering Institute" in Smolensk, Smolensk, Russia
Borisov Vadim V.
The branch of National Research University "MPEI" in Smolensk, Russia
Dli Maxim I.
The branch of National Research University "MPEI" in Smolensk, Russia

Abstract

The specificity of sociotechnical systems (STS) is the heterogeneity of objects, as well as the diversity of interrelations and interdependencies between them. The authors suggest monitoring clusters of STS basing on a combination of fuzzy cognitive approach and dynamic clustering methods. The use of fuzzy cognitive approach allows us to represent relationships between STS objects as fuzzy relationships of mutual influence. Analysis of these fuzzy relationships is used to select indicators for identifying and monitoring the dynamics of STS cluster structure, as well as to analyze the stability of identified STS clusters and the whole STS. Dynamic clustering methods are applied in the monitoring of the STS clusters dynamics. It consists in analyzing changes in the cluster structure of STS, including the drift of cluster centers, the disappearance of old clusters and appearance of new ones, their integration and separation.An example of monitoring clusters dynamics in the power engineering system of the Smolensk region is presented

Keywords

Sociotechnical system, Monitoring, Fuzzy cognitive model, Dynamic clustering