Model-based test case generation and optimization using UML diagram: a research proposal

Abstract

Software testing is an essential activity in software development life cycle (SDLC). It has been used as a process to discover bugs in software product with a specific end goal to determine and enhance software quality and reliability. Because the size of software system is growing faster, more time and effort are needed to do testing. Manual testing technique is not relevant nowadays because of the complexity and the size of system that require more time and manpower for testing. To address this problem, Model-Based Testing (MBT) has been introduced as one of the techniques for test case generation because of the early detection of fault at design phase. However, although MBT gives many advantages on testing, it also has some problem on test case generation such as a large number of test cases will be generated using the generator tools. To overcome this problem, optimization approach has been proposed in this study. In this paper, we present a research proposal to develop a new strategy for generating test case from Activity Diagram with Genetic Algorithm (GA) used to optimize the generated test cases. We summarize the initial findings of the experiments carried out and draw conclusion on future works.