Combinatorial test case generation from sequence diagram using optimization algorithms

Combinatorial Testing plays an essential role in generating optimized test cases to detect defects that occurred by interactions among input parameters of the systems. To generate combinatorial test cases, information about parameters, values and constraints is essential. This information is given to the system manually in the current practice, making it difficult to test software systems. UML Sequence Diagram describes the dynamic behaviour of the software system. The authors presented a novel approach to generate combinatorial test cases from UML Sequence Diagram in this paper. The Combinatorial Test Design Model (CTDM) is used to get information like input parameters, values, and constraints for generating combinatorial test cases. Extracting this information from UML Sequence Diagrams and identifying interactions among the input parameters is a challenging task. A rule-based approach is used to extract the information related to CTDM from UML Sequence Diagram. Once this information is extracted, combinatorial test cases are generated using Optimization algorithms, namely Particle Swarm Optimization and Simulated Annealing. This presented work is a study to generate various combinatorial test cases through optimisation algorithms which will aid in the management of Indian Railways. The significant contributions of this research are (1) Extraction of parameters, values and constraints from UML Sequence Diagram by using the rule-based algorithm. (2) Generation of combinatorial test cases from that extracted information using optimization algorithms. A case study of the Concession Management Subsystem of Indian Railways is presented to demonstrate the proposed research work. The authors recommend that All Combination testing, Particle Swarm Optimization algorithm and Simulated Annealing algorithm be used for simple, moderate, and complex UML Sequence Diagrams to generate a minimum number of combinatorial test cases.