Application of SQuaRE and Generalized Nets for extended validation of CE systems

Abstract

The need to develop robust, quality software architectures is more critical than ever today due to the significant complexity, size and interoperability requirements typical of modern systems. Rather than wait until the architecture, design and potentially implementation phases have been completed, this paper proposes that evaluation and testing methods should be applied during the architectural phase itself, and should be rooted in standard methodologies and processes, then complementary checked through the usage of Generalized Networks (GN) theory by a simulation platform. To guarantee the quality of the architecture the ISO/IEC CD 2504n of the SQuaRE series of standards (which describes a process for evaluating the quality of software products and also describing the requirements for the components of an architecture), are adopted as a reference methodology. (GN) is a tool for Discrete Event Simulation (DES), which is equally well suited for modelling simple and large, complex systems. For a complete assessment of quality, GNs seems to be a proper complement for the validations of the dynamics of the interoperable system, after tested by the quality procedure of SQuaRE. With these, a complete system can be validated through visualization, specification, simulation, analysis, development, and report-out of the test and evaluation procedures that is applied to the architecture and its components. By applying these quality assessment techniques earlier in the software development lifecycle, it is predicted that churn of both code and the architecture itself can be reduced. This would deliver improvements in the quality, reliability, consistency and indeed sustainability of the architecture and its implementations, compared when just either SQuaRE or GNs are used separately.