Process reliability in service oriented architectures

The creation of business processes by composing Web services has attracted considerable efforts with respect to integrating elements of Service Oriented Architectures within enterprise applications. If one of those atomic services fails, the whole business process does not necessarily have to fail, too, as it is possible to exchange the faulty service during runtime with another one which has the same functionality. In this paper we present a service monitoring and exchanging approach that aims to ensure a successful execution of crucial processes in case of faulty services. It consists of three parts which supervise the process execution and try to recover it if an error occurs. In case process reliability is not fore grounded, our approach can also serve as basis for process optimization.