Stochastic Software Test Case Scheduling

• Main Authors: Daniel Suriawidjaja, 薛來德
• Other Authors: Gary Chen
• Format: Others
• Language: en_US
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/92249814180767246369

碩士 === 中原大學 === 工業與系統工程研究所 === 103 === This paper consider test suite reduction problem with requirement coverage as test ade-quacy criteria. In this paper, we introduce a stochastic test case execution time to simulate the real world problem where as some error whether in testers or machines to test might occurs. Scheduling were introduced in this paper to dispatch the reduced test suite from test case minimization to the multiple testers. The objective of this study is to simulate the stochastic test case completion time to perform all the test case in the reduced test suite by allocating and sequencing the test cases to the testers. Simheuristics method were used in this paper to deal with the stochasticity of the prob-lem. This method combines both metaheuristic optimization and simulation to the study. Me-taheuristic that choosen to be used in this study are Firefly algorithm, which is a quite recent swarm intelligence algorithm that already used in few studies to solve scheduling problem. Cuckoo search were dispatched as comparison with the proposed firefly algorithm. Re-sults and discussion from the experiments conducted found that the number of test cases, re-quirement coverage, and the test case completion time (makespan) are strongly correlated, in regards that with the increasing number of test cases, the total completion time would also in-creases.