Abstract:
Software testing is a significant and complex phase of software development as it ensures the
complete functionality of the software. To evaluate the behaviour of the software the testing
process acquires generation of test cases as an input for the system under test. Generation of test
cases remains a challenging task usually as it is done based on human intelligence and it has
become much complicated and time-consuming process. Therefore, it can be enhanced by
automating the test case generation mechanism to identify and eliminate bugs. Unified Modelling
Language (UML) is a de facto standard that has been used in academia and industry currently.
Test case automation using UML diagrams is more effective and efficient as it is done early in the
software development life cycle. Therefore, to overcome the problem of time as well as budget
constraint, it is required to optimize the entire test suite. Many researchers have proposed
conventional based approaches for dealing with this problem and they have achieved optimized
test cases by selecting, minimizing, or reducing the test cases. However, in this research UML
sequence diagram and class diagram have been proposed for white box testing. To generate test
cases these two diagrams have been proved as compatible from existing literature. Currently,
existing approaches dealing with test case optimization have achieved 85% and 90% statement
coverage for the System Under Test 1 (SUT1) and System Under Test 2 (SUT2) using genetic
algorithm. However, in this research Unified Non-Sorting Genetic Algorithm (U-NSGA-III) has
been proposed for automated test case generation and optimization. The proposed approach has
achieved 95% statement coverage for SUT1 and 90.47% for SUT2. In this dissertation, two
benchmark case studies have been used and controlled experimentation have been performed for
optimization of test cases. The comparison of our approach has been done with GA-UNSGAIII
and PSO-UNSGAIII. From our results, it has been concluded that proposed U-NSGA-III has
performed better than other approaches.
