A technique of building a value function at the stage of conceptual design of microprocessor systems

Abstract

The aim of this study is to formalize the selection of optimal technical solutions early in the design of microprocessor-based systems, which allows developers to analyze the recommended solutions, and has, in comparison with the traditional «intuitive» approach, at least two undeniable merits. First, the accepted assumptions and limitations are clearly formed. Secondly, it is defined precisely, in what sense the decision is optimal. When designing microprocessor systems (systems hereafter), several characteristics have to be taken into account at the same time. In general, when n properties are taken into account for each of the compared systems, then the solution of the task of choosing “the best” system depends on choosing a function-criterion. Such function is called a value function in the article. A simple quadratic function is suggested as the value function, it can be interpreted as the distance in Euclidean space of systems technical data. The system, which corresponds to the point nearest to the point characterizing the master system with “limiting” characteristics, is considered the best one. This function approximates the designer’s system of preferences signifi cantly better than a “classical” linear value function. In conclusion, note that the developed recommendations allow the designer of complex technical systems to analyze the proposed solutions in the early stages of design and, in case of disagreement with them, to indicate the reasons why he considers them inadequate. The designed machine optimization of technical solutions in conjunction with the traditional engineering approach should allow more reasonable choosing the structure of systems at the stage of systems conceptual design.

Keywords

• value function
• characteristics
• components
• master system
• parameters
• distance
• measure of quality