Flexible Service Selection Optimization Using Meta-Metrics

The Web services paradigm enables an organisation to implement business processes by orchestrating existing services, some of which may be provided by external organisations. Service selection optimization is the application of techniques to optimize the selection of services for business processes. These processes are often modelled to accommodate anticipated variations in control-flow during execution using choice control-flow patterns. This enables a single business process model to represent multiple execution paths, a phenomenon known as modelled flexibility. In certain scenarios, modelled flexibility can cause conflicts in service selection optimization, making it impossible to simultaneously optimize all execution paths. This paper presents an approach to service selection that addresses this type of conflicts by enabling the user to bias service selection based on an arbitrary set of meta-metrics.