The article addresses improving the performance of man-machine systems (MMS) by improving the quality of decision support system software (DSSS). The proposed solutions are based on existing models and methods, which have been analysed in detail in a scientific study, in particular the application of methodical static application debugging in certain solutions obtained. It is proposed to use the methods for solving the problem of MMS control according to the vector criterion as the basis for solving this problem. A mechanism for synthesizing a comprehensive plan of operations to improve the quality of the DSS software is proposed, and the possibility of analysing the step-by-step implementation of a plan to obtain an analytical assessment is shown. For this purpose, a comprehensive plan is presented in the form of a relevance tree, built in accordance with the provisions of graph theory and the rank approach. The time complexity of the algorithms implementing the relevance tree tasks was evaluated. Based on the formalized solutions obtained, it was found that the maximum number of edges for a graph that implements the relevance tree does not exceed , the number of views of edges does not exceed , that is, the total number of operations that implements the scanning of edges and adding them to the weighted minimum cover (WMC) does not exceeds . It is established that during the implementation of the operation of adding ribs in the WMC, at each step, after the fixation of the previous result, the number of ribs to be examined decreases, so the total complexity of the algorithms will not exceed . The solutions obtained meet the requirements of DSTU ISO/IEC 9126, ISO DSTU / IEC 14598, and take into account the requirements of the Software Quality Requirements and Evaluation series of standards as the value of the vertices of the event tree graph. In the process of solving the problem, the specificity of the functioning of the MMS was taken into account, the possibility of formalizing various aspects of knowledge (aesthetic, disizional, causal, diontic) and ensuring a given level of efficiency in finding solutions.
Architecture for a Multimodal and Domain-Independent Clinical Decision Support System Software Development Kit
