应用改进的遗传算法优化软件产品线特征选择

Feature Selection Optimization Based on Atomic Set and Genetic Algorithm in Software Product Line

软件产品线工程是一个利用更全面的软件复用技术能够在开发成本和产品上市时间方面极大改善软件开发过程的有效方式.特征模型是软件产品线工程中描述软件产品家族共性和可变性特征以及特征之间关系的需求模型.基于特征模型的产品配置难点在于如何从复杂的特征模型中选择满足约束条件的最优的特征组合,如资源约束等.为了实现有约束的特征选择优化问题,提出一种基于原子集和遗传算法优化特征选择的方法.该方法是在原子集优化特征模型基础上,利用特征模型的整体约束信息将种群分为有效配置和无效配置,对有效配置和无效配置进行交叉等遗传操作使其加速向最优解收敛.实验结果说明了该方法的有效性.

Software product line （SPL） engineering is an increasingly prominent approach to software development by leveraging the extensive reuse,which can significantly improve software development productivity,quality and time-to-market. Feature models are a popular formalism for describing the commonality and variability of software product line in terms of features. A key challenge when using a feature model to derive a new SPL configuration is determining how to find an optimized feature selection that minimizes or maximizes an objective function, such as total cost, subject to resource constraints. To help address the challenges of optimizing feature selection in the face of resource constraints,this paper presents an approach that uses Atomic Set and Genetic Algorithm for optimized feature selection in SPLs. We first optimize the feature model using atomic set. Then we classify the population into feasible and infea- sible based on the integrity constraints. The genetic operations make it speed up convergence to the optimal solution. The experimental results show the effectiveness of the algorithm.