Fuzzy based Quantitative Evaluation of Architectures using Architectural Knowledge

Software architecture is a standard and essential part of system development. The knowledge produced and consumed during this process needs to be shared and reused among different stakeholders. Therefore, software architecture knowledge (AK) needs to be managed for improving the architectural capabilities. ADUAK (Architectural Development using Architectural Knowledge) model has been developed to support a framework for capturing and using architectural knowledge to improve the architecture design process. ADUAK guides the architect in selecting the most appropriate architecture, where the interactions between quality attributes, tactics should be analyzed and quantified and the results should be considered for the quality driven architectural design process. Ontology based AK tool is needed, because ontologies are used to combine the real world domain information and software architectural knowledge from various experts, to maintain up – to – date documentation that evolves over time. The primary focus of ADUAK is to acquire knowledge for the fuzzy based evaluation of the architectures of the system using ontologies. It is needed because the architecture represents an enormous risk in a development project. Making bad architectural designs may substantially slow down the system and even lead to failure. The quantitative impact of incorporating the quality in the software architectures is then calculated through a fuzzy based approach for the set of design alternatives. The paper aims to bring the contribution of ADUAK model for the fuzzy based quantitative evaluation of architectures with the help of architectural knowledge through a suitable case study.