An automated support for evaluating alternative design decisions

Abstract

In recent years, there have been increasing interests in using Component-Based System Development (CBSD) approach to develop large and complex applications. It's believed that satisfying quality attributes is an essential factor for the success of such systems. This paper introduces a novel approach to support the right decisions on selecting design alternatives decision at the early stage of component-based system development. In this approach, we implement Particle Swarm Optimization (PSO) algorithm, as the selected Intelligent Swarm (IS) optimization technique, to explore the design space for optimal design that satisfies the quality requirements. A case study from the Embedded System (ES) domain is used to demonstrate the applicability of the approach. The objectives of this paper are, to aid architects to decide on the level of redundancy within each subsystem based on reliability, and to make trades-off between reliability and performance based on the technical and cost preferences. The results of some preliminary experiments indicate the potentiality of the approach.