Application of optimization techniques to model based software testing for maximal test coverage

Abstract

The aim of this research is the application of nature-inspired hybrid meta-heuristic algorithms for test case generation in software systems. Hybrid meta-heuristic algorithms such as Bee Colony Firefly Algorithm (BCFA), Particle Swarm Bee Colony Algorithm (PSBCA) and Firefly Cuckoo Search Algorithm (FCSA) have been used for automatic generation and optimization of test cases from Unified Modeling Language (UML) diagrams. In order to find all possible test cases of a software system, the corresponding UML diagrams are first converted into their respective graphs for easy traversal.First, a state-chart diagram is converted into the corresponding State-Chart Diagram Graph (SCDG) then traversed to generate the test cases. The Depth First Search (DFS) traversal with backtracking has been employed for capable of generating suitable test cases in less time. The generated test cases cover all possible paths in the system. The generated test cases are further optimized by applying the hybrid Bee colony Firefly (BCFA) meta-heuristic algorithm. The result of hybrid BCFA is compared with the results obtained from individual algorithms i.e., BCA and FA. newline