【摘要】 通过多目标多属性决策方法,解决基于服务质量(quality of service,QoS)的大规模Web服务选择和组合问题。不同于以往将多个QoS值赋权累加为单一值的方法,采用多属性决策方法,同时处理多个QoS属性,将每个解到正负理想点的距离转化为多目标优化问题。提出一种基于ε支配的多目标遗传算法来解决Web服务组合优化问题。计算结果为一组折中的帕累托最优解集,为用户提供多种选择方案。当用户所选择的服务运行失败时,用户可以从其他备选服务中进行选择。实验结果表明,所提出算法具有满意的收敛性、分布性和可扩展性,且算法复杂性优于流行算法NSGA-Ⅱ和SPEA2。更多还原

【Abstract】 Through the method of multi-objective and multi-attribute decision,this paper solves the problem of QoS(quality of service)-oriented large scale Web service selection and composition.It differs from the traditional method which aggregates all QoS values into a single value.This paper treats all QoS attributes simultaneously by using the multi-attribute decision making method and employs a multi-objective optimization model to formulate the distance of each solution from the positive ideal solution and the negative ideal solution.We develop anε-dominance multi-objective genetic algorithm to solve the problem of Web service composition optimization.The Pareto frontier,the set of optimal compromise solutions,supports users of either making a flexible decision or choosing an alternative when current service fails.Experimental results verify that the algorithm has satisfying convergence,distribution,and scalability and its computing complexity surpasses the popular non-dominated sorting genetic algorithm(NSGA-Ⅱ)and strength Pareto evolutionary algorithm 2(SPEA2).更多还原