DEVELOPMENT OF AN EXPERT-ORIENTED SYSTEM OF SUPPORT FOR DESIGNING LARGE-SCALE INFORMATION SYSTEM
Roman V. Erzhenin
The article deals with the actual problem of improving the design quality of large-scale information systems of public administration (KISGU). According to the author, the choice of acceptable parameters for the functioning of KISGU at the stages of its macro-design is constrained not so much by the lack of effective modeling methods as by the lack of convenient tools that allow one to acquire, accumulate and use heterogeneous knowledge for decision-making when building adequate models of a com-plex information system (IS). As a methodological basis for intellectual support for modeling and assessing the effectiveness of IS, it is proposed to use a complex DEA-model of assessment, built on the principles of structural, functional and territorial stratification of information on the functioning of a complex KISGU. A review of existing approaches to assessing complex IS and an analysis of various tools implemented on the principles of DEA-models made it possible to conclude that it is impossible to use them for assessing systems of the KISGU class. To support the examination procedure, the author proposes to develop a new web application using the Yii2 Framework. For this, the article describes the software architecture of an expert-oriented decision support system (DSS) and presents the results of its testing. As information support, we used open machine-readable data containing information on the characteristics of the functioning of the state budget management information system (ISUGB). The obtained results of the ISUGB assessment confirm the author's hypothesis about the possibility of enhancing the intelligence of an expert at the design stages through the use of special tools. The proposed approach can be used in the design of the system to select the parameters of the functioning of the ISUGB, in the audit of the effectiveness of the activities of institutions and as a practical guide for the training and retraining of IT specialists of specialized specialties.
budget; performance audit; macro design; modeling; DEA; decision support systems