Defining Software Quality Rank using Analytic Hierarchy Process and Object-Oriented Metrics

An alternative approach for measuring object-oriented (OO) software quality has been proposed by applying the Analytic Hierarchy Process (AHP) scheme. This approach is effective in defining the rank of software quality over a number of Java applications based on two sets of metric values representing main OO structural properties. The Metrics for Object-Oriented Software Engineering (MOOSE) and Metrics for Object-Oriented Design (MOOD) that have been describing individual characteristic of OO property now are integrated by means of AHPs pairwise comparison to deliver an estimate quantitative value of OO software design quality. A series of experiments were conducted by applying the approach to students works in Fundamental of Programming class. The results of applying AHP scheme were compared to experts judgments for verification. The result demonstrates the solidity of AHP mechanism for the intended purpose of measuring OO design quality.