On improving indicators for assessing the decision support systems’ software quality

Abstract

The article deals with the applied software quality assessing issues. The work relevance is determined by the fact that currently, in the normative documents there is a list of the main characteristics assessing the application programs quality, which, as shown, is not fully fit for purpose, providing no quality control and compliance testing programs to customer requirements, formulated in the development task. One of the reasons for the current situation is the systems’ quality evaluating exclusively quantitative indicators impossibility that combines both technical means and a human operator. An attempt to use world practice, for example, relatively successful quality models from the ISO/IEC 25000: 2014 standards have not yet been implemented: the model itself is allowed for use by regulatory documents (GOST R ISO/IEC 25010-2015), but the quality indicators described in it are not accepted. Private improvements to existing methods do not solve the problem systematically. In the article, based on the software existing approaches to assessing the quality analysis, a scientific and practical problem’s formulation is formulated and one of the approaches to its solution is proposed, based on the existing methods’ refinement for assessing quality based on the model described in GOST R ISO/IEC 25010, taking into account real users needs, interpreted through reducing the errors’ likelihood of the first and second kind that arise when using the software. The formulated problem solution will provide a general increase in the automated control effectiveness through the use of the quantitative and qualitative assessments of the software being developed