An Evaluating Method of QoS for Composite Services Based on Interval Muti-attribute Decision Making

Composite Web services needs to meet the global Quality of Service (QoS) requirements of Customers. The evaluation of the services QoS is a important standard to make decision from numerous candidate composite services. However, due to the difficulty on the exact measurement of services QoS and exact expression of customers’ QoS requirements, the decision of Web services composition should be made on the fuzzy information. In this paper, an interval muti-attribute decision-making based evaluating method of QoS for composite services is presented. The method uses the interval numbers to describe the QoS parameters of the Web services, as well as the customers’ QoS requirements. The method for measuring interval-valued QoS parameters and aggregation are also proposed. With the consideration of the Characteristics of the customers’ requirements, the evaluation on QoS parameters is made to construct the decision-making matrix following the principle of "To be appropriate for Customers’ requirements". Finally, based on the matrix, a comprehensive evaluation is made by using TOPSIS. By assessing the customer satisfaction and interactivity, it gives a comprehensive evaluation on the composite Web service based on uncertain multi-attribute decision making method. With the help of the adaptive genetic algorithm, it tries to find the appropriate composite service that meets the global QoS requirements of the customers.