The method of improving the efficiency of assessment of support of object-oriented software
A. V. Lykhvar · Telecommunication and information technologies · 2021
The work examines issues of improving the quality and level of support of object-oriented software. Analysis of existing software quality models demonstrated the relevance of maintaining an appropriate level of software support. At the same time, the existing approaches are mostly distinguished by terminological ambiguity, variety of formulations and methods of measurement and interpretation of results. The problem is also the complexity of the support process itself, which arises from a large number of internal and external factors, which makes the adjustment of the support model a long-term and complex process. A modification of the delta maintainability model (DMM) is proposed by expanding the measurable properties of the source code, determining their relationship with the maintainability sub-characteristics defined by the ISO/IEC 25010:2016 standard. A matrix of relations of sub-characteristics of ISO 25010 quality with the characteristics of the source code properties is constructed. The work uses WMC, NOC, NOM, MLOC, PAR, VG metrics, which take into account the features of the object-oriented programming paradigm. A mathematical model has been developed for calculating compliance indicators. To check the adequacy of the model, an experimental study was conducted on the example of 6 open source software products with the implementation of the object-oriented part in the Python programming language, which have different functionality. The metrics of the DMMS+ model together with DMMS have been proven to reflect the relationship between source code changes that affect maintainability. The correlation of DMMS and DMMS+ indicators according to the analysis of 1000 changes made in the repository is from 0.77 to 0.86. Predictive assumptions about object development include maintaining object-oriented software by modifying the delta maintainability model (DMM) by extending the measurable properties of the source code.