Test Data Automatic Generation Based On Simulated Annealing Algorithm For EFSM

Heuristic search algorithms are widely used in various fields for its intelligent features. The common heuristic search algorithms are genetic algorithm, simulated annealing algorithm, tabu search algorithm and so on. They are often used to test case automatic generation in software testing field.Model introduces the function and behavior of software system. It could make testing process more clear and more order. EFSM model-based testing technique is a new testing technique, mainly include test sequence generation and test data generation. Among them, test data generation is hot and difficult of software testing research field. Currently, the literatures are less about EFSM model test data automatic generation technique. Literatures only used genetic algorithm for test data automatic generation on EFSM model. Therefore, try to use other heuristic search algorithm on EFSM model for test data automatic generation is a worthwhile direction to explore.This paper summarizes the more extensive method which applied to software test data automatic generation. Study the heuristic search algorithm, on this basis, try to use the simulated annealing algorithm on EFSM model to research the test data automatic generation. Main task are as follows:1. Simulated annealing is used on EFSM model for test data automatic generation.2. Through experiments to verify the feasibility of simulated annealing applied on EFSM model for test data automatic generation. Analyzing the effect of test data generation with different simulated annealing parameters, and the traditional simulated annealing is improved, so that the generation efficiency further improved.3. Compare the efficiency and stability in test data generation with simulated annealing algorithm, random algorithm and genetic algorithm.