Research On Automatic Generation And Prioritization Of Test Cases Based On EFSM Model

In recent years,with the development of software testing,the number of test cases has increased dramatically.How to effectively use limited resources to reduce test costs and improve the detection rate of software defects has become a big question.Software testing is an important part to developing high-quality software in these days.The main factors affecting the cost and efficiency of testing are the size of test case set,the total length of test cases and the execution order of test cases.Test case automatic generation technology is designed to cover software test requirements with minimal test case sets,reducing labor costs and system resources.After obtaining test case sets,test case prioritization technology can improve test efficiency by adjusting the execution order of test cases without changing the number of test cases,and then support rapid detection of software defects.Therefore,test case automatic generation and test case prioritization is very important to improve the efficiency of software testing and reduce the cost of testing.At present,many scholars have carried out related research on test case automatic generation and test case prioritization and achieved success,but there are still some problems to be solved,such as high test cost,low test efficiency,and inability to adapt to complex models.This dissertation focuses on the above-mentioned problems for the extended finite state machine(EFSM)model.The specific contents include:(1)A test case automatic generation method based on set coverage theory is proposed.Firstly,the minimum test cost migration coverage criterion is introduced,and then a migration coverage algorithm based on critical migration is designed to generate test migration fragments.Then an improved test migration fragment reduction algorithm is proposed to eliminate redundant migration.Finally,the test migration fragments in the set obtained by the algorithm are effectively transformed to test case sets.The feasibility and superiority of the proposed method are verified by experiments.(2)A test case prioritization method based on Aditional TA strategy is proposed.Aiming at the disadvantages of Additional strategy such as blindness and uncontrollability,this dissertation proposes an improved Additional TA algorithm from two aspects of demand importance and selection scheme.First,the importance of migration is calculated,then the total test capability value of test cases is dynamically adjusted by feedback mechanism.Finally,priority is given according to the total test capability value.When the total test capability value of multiple test cases is the same,the number of migrations in the original path of test cases is used as the ranking index.The effectiveness of this method is verified by experimental comparison of the average percentage of failure indicators.(3)Based on the above test case automatic generation and test case prioritization method,a software automation test platform EFSM?SATP based on EFSM model is developed.Through an example of an UAV flight control system,the system is so easy to be used.The method above is practical and effective,can be used as a preferred scheme and research direction for software test development.