Optimal selection of composite web service based on QoS and risk evaluation

One of the most promising advantages of Web service technology is the possibility of creating added-value services by composing existing ones. A key step for composing services lies in the selection of the individual services. In this paper, we propose an approach for selecting optimal Web service composition plan and optimal candidate services, aimed at maximizing service quality and minimizing risk of service failure. It is an extension of previous QoS-driving service selection approaches, by adding consideration about the risk of service failure. The mathematical model is based on Petri net and fault tree in order to compute the risk for each abstract composition plan.