Measure extendibility/extensibility quality attribute using object oriented design metric

Abstract

Software design is one of the very important phases of the software engineering. The costs of software can be minimized if improvements or corrections made during this stage. Several of the current computer aided software engineering (CASE) tools like enterprise architect (EA) v12 do not have the capability to improve the design. This work aims to develop an algorithm that helps the software engineers evaluating the design quality utilizing one of the object-oriented (OO) design models namely quality metrics for object-oriented design (QMOOD) which represents as hierarchical model that describes the relationship between quality attributes such as reusability, extendibility and properties of the design of OO design. This algorithm describesed how the assessment of the extendibility/ extensibility using the software metrics has been done and the impact of the involved metrics in the extendibility value. Results obtained demonstrate the effect of OO design metrics such as inheritance, polymorphism, abstraction and coupling in quality characteristics like extensibility. The results show that lower values of abstraction and coupling, obtain higher value of extendibility which means the class diagram is ready to accept additional improvements. The proposed algorithm has been tested on two different systems (test cases) that vary in their class diagrams, functionalities, and complexities.