As a relatively new software model, there remain many challenges in realising a true service-oriented vision. The service-oriented systems which underpin modern business processes must be able to react to constantly changing environments and business requirements. The quality of a service-oriented system depends not only on the quality of service (QoS) provided by services, but on the interdependencies between services, resource constraints imposed by the runtime environment, and events such as network outages. It is difficult to anticipate the impact that many of these emergent factors will have on the behaviour of the system. The third-party nature of software services also presents the service consumer with limited control over the quality of a system.
Existing quality assurance initiatives for service-oriented systems are currently limited in the service quality control they offer the consumer, provide poor support for expressing quality characteristics, provide poor support for quality assurance at runtime, provide poor support for resource-restricted systems, and offer limited scope for integration and customisation to provide an end-to-end quality assurance solution. To address these issues, this thesis presents the development of an integrated quality assurance framework, which combines quality assurance approaches from the service description and selection, service monitoring, service negotiation, and reputation system domains. The approach is illustrated with a series of service-oriented experiments, which evaluate the role of the framework in the system quality assurance process.