The PhD thesis is devoted to solving current scientific and technical problem of improving quality of development and maintenance of "Smart Home" systems (SHS) by applying methods and tools of domain modeling to create variable software components of such systems. In the dissertation introduction the relevance of the research topic is substantiated, connection of the work with scientific topics is shown. The research purpose is formulated, the object, subject and research methods are defined, which are based on application of applied system analysis principles, models and methods of modern software engineering, in particular on use of domain modeling methods, object-oriented methods of software analysis and synthesis, quantitative metrics quality of software, as well as the use of basic principles of control theory, mathematical apparatus of set theory, expert methods of decision theory and use of unified modeling language UML and modeling language SysML for analysis of hardware and software solutions.
The scientific novelty and practical significance of obtained results are described. Information on practical application and personal contribution of the applicant, approbation of research results and their coverage in publications are given. Information on the structure and scope of the dissertation is given. The first section provides an analytical review of approaches to the maintenance of variable components of software systems for various purposes. The current state of software development in such new high-tech subject areas as the creation of Smart-Home systems, which in turn belong to a wider class of modern computerized systems: Internet of Things systems. A review of developments in the field of SHS allowed us to conclude that there is significant progress in creating effective hardware solutions, but at the same time the issues of improving the efficiency of development and maintenance of software components by ensuring their variability and reusability in promising SHS, and this determines the relevance of this dissertation research. The second section of dissertation presents the methodological basis for the development and maintenance of variable components of the software of SHS. The possibility of using domain modeling as a conceptual basis for analyzing the properties of variability of software components of SHS is considered and an analytical review of methods for developing variable components of software for SHS is performed. The third section describes the process of developing models, procedures and tools for building and maintaining variable software components of SHS. The operational model (OM) of requirements processing in the context of development of variable software components according to Scrum methodology is offered. OM allows to formalize the process of requirements variability management in variable SHS, which opens the possibility of collecting information, based on the information to analyze commonness and classification of requirements, which in turn reduces the number of requirements at the stage of sprint requirements catalog, due to which it becomes possible to improve the quality of the requirements analysis process. The domain model of variable software components at the stage of support of the SHS is proposed, and the procedure of evaluation of the quality indicator of the maintenance of the software components of the SHS is developed. Architectural models of appropriate tools have been developed to provide automated support for the proposed procedures for building variable software components. The fourth section of the dissertation is devoted to software implementation and experimental research of developed models and tools. The calculations of the proposed indicator are based on the calculation of the amount of time spent, which is necessary to maintain certain functionalities of the system at the stage of its maintenance. As a result of the analysis of calculations of the criterion of support, the application of the proposed modification of the architecture of the SHS leads to an increase in the maintenance of the system. The architecture of the CASE toolkit is proposed - a tool for automating a number of software support processes for SHS. This provides an opportunity to automate the processes of preliminary analysis and quality assessment of alternative options for the development of new components of the software SH. In the dissertation work the actual scientific and applied problem of quality improvement of processes of development and support of variable components of the software of SHS on the basis of use of methods of domain modeling is solved. As a result of the experiments, data were obtained that allow us to draw conclusions about the efficiency and appropriateness of the proposed approach to determine the effectiveness of alternative architectural solutions.
