Computational System Architecture Development Using a Holistic Modeling Approach

Abstract

This paper presents an innovative system architecture development framework that allows the search of optimum architecture solutions within large design space by automating certain model construction, alternative generation, simulation, and assessment tasks. Such framework is facilitated by a holistic modeling approach that combines the capabilities of Object Process Methodology (OPM), Colored Petri Net (CPN) and feature model. The resultant holistic model not only can capture the structural, behavior, and dynamic aspects of a system, allowing strong analysis methods to be applied, but also can specify the architectural design space allowing generation of architecture alternatives that cover it. The proposed framework and suggested implementation is generic targeted at systems that can be specified by logic models using object-oriented paradigm. A partial implementation of the proposed approaches is presented with the design of reconfigurable manufacturing systems (RMSs) as an example, which is formulated as a multi-objective optimization problem with the Genetic Algorithm (GA, particularly, NSGA-2) as the search algorithm. The RMS is a multi-part flow line structure with identical machines in each production stage.