Study of empirical approaches to analyze the software metrics

Abstract

Software affects nearly every aspect of human lives. Software functional quality is a key to achieve industrial and business relevance, in particular to industrial development and growth. Software metrics are important indicators to improve the processes and products in all organizations. They define baselines of quality and productivity and enable comparisons against industry averages that help in identifying opportunities for improvement. In addition, Software metrics design and analysis are major activities in the software development life cycle. Software metrics play a vital role in software cost, quality, scheduling, reliability and maintenance. There are various methods to decide which metrics should be used for which purposes. Attributes of a metric may be independent; or attributes may depend on one another. Analytical Hierarchy Process (AHP) is used to assign weights to various parameters of a decision model when they are related to each other in a particular hierarchy. Analytical Network Process (ANP) and Fuzzy ANP are used to solve the decision problem, where attributes of decision parameters form dependency networks. The objective of this paper is to explore the possibilities of using empirical approaches like AHP, ANP and Fuzzy ANP to analyze the software metrics by measuring the weights of different attributes.