Software Test Resource Allocation Based on Adaptive Operator Selection
- Wenjie Chang, Baolong Guo
- Corresponding Author
- Wenjie Chang
Available Online March 2018.
- https://doi.org/10.2991/iceea-18.2018.37How to use a DOI?
- computer software; software test resource allocation; multi-objective optimization; multi-armed bandit; moea/d
- With the rapid development of computer software, making the complexity of the software system increased dramatically, the inevitable cost of testing resources is also increasing. Under the conditions of limited resources, how to better find the balance between resource consumption and reliability obtains more and more people’s attention. Optimization of test resource allocation (OTRAPs) involves finding the optimal reliability, cost, etc. Therefore, the traditional test resource allocation optimization problem is a multi-objective optimization problem. In recent years, many effective algorithms, such as MOEA/D and other famous algorithms, and achieved good results. However, the shortcomings of these algorithms are the use of a single operator and fixed neighborhood size, although the operator is not adapted to each search stage and small neighborhood size can accelerate convergence and big neighborhood size can get rid of the local optimal solution, The algorithm (DS-MAB-MOEA/D) we proposed based on the Multi-armed Bandit principle to adaptively select the excellent operator and neighborhood size in the pool during the different stages of evolution. Considering the retention of Pareto set’s diversity, this paper embeded distance sorting algorithm in DS-MAB-MOEA/D algorithm and applies it to 16-module system, Experiments show that the algorithm is better than the MOEA / D algorithm.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Wenjie Chang AU - Baolong Guo PY - 2018/03 DA - 2018/03 TI - Software Test Resource Allocation Based on Adaptive Operator Selection BT - 2018 2nd International Conference on Electrical Engineering and Automation (ICEEA 2018) PB - Atlantis Press SN - 2352-5401 UR - https://doi.org/10.2991/iceea-18.2018.37 DO - https://doi.org/10.2991/iceea-18.2018.37 ID - Chang2018/03 ER -