Increasing Process Reliability in a Geospatial Web Services Composition

This work proposes a method to specify the data quality requirements for the composition of geospatial web services at a semantic level, by using a rule-based approach. Those rules allow the semantic annotation of spatial data, and coupled with the conditional planning method, it can model more precisely the unpredictability of the compliance with the data quality requirements during the composition run-time. The mechanisms that automate the web services composition using artificial intelligence planning can incorporate those rules. The service composition produced by this method are more robust, thus improving process reliability when working with a composition of chained geospatial web services.