运行环境的开放性和动态性使得Web服务QoS具有内在的不确定性。为此,首先对QoS不确定性进行了分析,并分别采用实数、区间数和随机数对不同特性的Web服务QoS进行描述。然后,结合用户QoS需求约束,分别利用可能度和半方差理论实现区间数和随机变量的比较,将不确定混合QoS感知的服务选择问题转化为确定型多属性决策问题,并利用TOPSIS(technique fororder preference by similarity to ideal solution)方法进行求解。服务选择考虑了客观权重和主观权重。最后,通过实验验证了服务模型的有效性。
Open and dynamic of operation environment leads to the inherent uncertainty nature of Web service QoS.First,we analyse the indeterminacy of QoS,and describe Web service QoS by real number,interval number and stochastic variable.Then in conjunction with users’ demand constraint,we use semi-variance theory and possibility degree respectively to compare the stochastic variable with interval number,and thus transform the uncertain hybrid QoS-aware Web service selection to a multiple attribute decision-making problem in deterministic type and further solve it by using the technique of order preference by similarity to an ideal solution(TOPSIS).The service selection takes into account the subjective weights and object weights both.At last,the experiment proves the effectiveness of the serve model.