| With the advent of the era of information science,the scale of computer software is dramatically expended at an alarming rate,and software testing plays an important role in guaranteeing the quality of software.The Extended Finite State Machine(EFSM)can be used to describe the dynamic behaviors of the software process,which is a formalized description method based on the state of software.Currently,Most of the research on EFSM model all focused on two aspects:the test sequence generation and the test data generation.Then the generation of the feasible test sequence is still in the early stage,which still needs further optimization method for realizing the feasible test sequence generation of EFSM model.In this paper,transition coverage criteria is used as an optimization objective,which combines with the feasibility of test sequence measurement method and Genetic Algorithm(GA)based test data generation method for EFSM model.In the optimization process,the feasible test sequence is generated with the application of multi-objective evolution algorithm with the objectives of transition coverage criteria and test sequence suite diversity.And a feasible test sequence suite generator is also designed and implemented based on set evolution algorithm for EFSM model,which can efficiently generate the feasible test sequence suite to satisfy the transition coverage criteria.The design and implement of feasible test sequence suite generator is proposed to validate the effectiveness of test sequence suite based evolution algorithm for satisfying transition coverage criteria of feasibility method.And six EFSM models are used for experimental comparison and analysis to verify the influence of the feasibility of test sequence for effectiveness of test sequence generation,and effective rate of feasible test sequence,and also including the influence on test generation efficiency.The results showed that:(1)the feasibility analysis of the test sequence can greatly improve the test generation efficiency,and it also can improve the effective rate of feasible test sequence.(2)the diversity of test sequence suite can also significantly improved test generation efficiency. |
