OPTIMIZING THE STRUCTURE OF SOFTWARE SYSTEMS SUPPORTING LOGISTICS AT THE DESIGN STAGE

Background: Computer software, widely used to support a broad range of logistics activities, is characterized both by increasing functionality and increasing complexity. For this reason, the process of software production, including the stages of specification of requirements, design, programming and testing, is time-consuming and expensive. The main goal of the software design phase is to determine the software architecture that identifies all software components and defines links and connections between them. The design phase also includes the development of the so-called internal logic of all extracted components, that is, detailed elaboration of algorithms for their operation and defining the structure of data used. It should be emphasized that the results of the software design process depend greatly on the knowledge and experience of the designer, because there are no universal behavioral patterns in this area. The main goal of the approach proposed at work is to reduce the role of the  subjective factor in the results of the software development process. The focus of this work is on this software development process within logistics processes.

Methods: The basic research method used in the work is mathematical modeling. The paper proposes a formal method of assigning the modular structure of the computer program by formulating and solving the corresponding double-criterion optimization problem. The module strength coefficient and module-coupling coefficients were established as modularization criteria of the program.

Results: The main result of the work is the method of determining the modular structure of the designed program by determining the solution of the two-criterion optimization problem. The numerical example developed to illustrate this entirel confirms the possibilities of its practical application. The modular structure of the program, based on the solution of the formulated polyoptimization task, is characterized by the maximum value of the modular power coefficient and the minimum value of the modular strength coefficient. According to the latest trends in software engineering, it is the optimal structure. The method can be useful in the process of designing software for systems supporting the implementation of logistics processes.

Conclusions: The author's method of determining the modular structure of the program, presented in the article, is an unprecedented attempt in the literature to use formal methods in the software design process, which could be implemented practically in the logistics processes. The lack of similar attempts probably results from the very low compliance of the software design stage with attempts to formalize it. In order to increase the possibility of practical application of the proposed method, it seems reasonable to conduct further work to implement the methods of developing software requirements specifications in a formalized form, e.g. with the use of mathematical notation.