An effective model-based integration testing technique for component-based software

Abstract

Recently, many issues and challenges have been identified in integration testing of current Component-Based Software Systems (CBSS), that lead to ineffective testing and, ultimately, to CBSS of poor quality. In this research, a model-based integration testing technique for CBSS has been proposed by combining model learning and testing techniques for the formal analysis and testing of a system of integrated black-box software components. The effectiveness of the proposed technique is evaluated by applying a case study, namely, the elevator system. The performance analysis is carried out by performing a comparative analysis of proposed technique and existing techniques by using three evaluation parameters. The results of the comparison showed that the proposed technique is effective and outperforms the existing well-known technique in terms of learned components that performance is 100% better, output queries that performance is 25.01% better, as well as error detection capability that is 75%.