Complexity assessment based on UML-activity diagram

Abstract

Assessing complexity can significantly contribute to the attainment of the various quality attributes associated with a system. The avoidable complexity can be identified and reduced on the basis of the assessment. It holds the key to success of the system being developed. Various evaluation methods exist which have specific objectives and basis and all contribute to enhance product quality. In this paper a Complexity Assessment approach based on Activity Diagrams (CAAD) is proposed to evaluate the process view of the architecture of a system. The proposed approach estimates the complexity of the system/class/function from the UML representation of the process view of the architecture in the form of activity diagrams. This complexity measure may be used to assess and estimate the time and effort required to develop the system. This approach can estimate the coding complexity in terms of size without actually developing the code for the system/class/function. The paper is on calculating a complexity factor C from the given activity diagram and further develop a relationship between C and LOC metrics.