Abstracts

Although it has become commonpractice to build applications based on the reuse of existingcomponents or services, technical complexity and semanticchallenges constitute barriers to ensuring a successful and widereuse of components and services. In the geospatial applicationdomain, the barriers are self-evident due to heterogeneousgeographic data, a lack of interoperability and complex analysisprocesses. Constructing workflows manually and discovering properservices and data that match user intents and preferences isdifficult and time-consuming especially for users who are nottrained in software development. Furthermore, considering themulti-objective nature of environmental modeling for the assessmentof climate change impacts and the various types of geospatial data(e.g., formats, scales, and georeferencing systems) increases thecomplexity challenges. Automatic service composition approachesthat provide semantics-based assistance in the process of workflowdesign have proven to be a solution to overcome these challengesand have become a frequent demand especially by end users who arenot IT experts. In this light, the major contributions of thisthesis are: (i) Simplification of service reuse and workflowdesign of applications for climate impact analysis by following theeXtreme Model-Driven Development (XMDD) paradigm. (ii) Design of asemantic domain model for climate impact analysis applications thatcomprises specifically designed services, ontologies that providedomain-specific vocabulary for referring to types and services, andthe input/output annotation of the services using the terms definedin the ontologies. (iii) Application of a constraint-driven methodfor the automatic composition of workflows for analyzing theimpacts of sea-level rise. The application scenario demonstratesthe impact of domain modeling decisions on the results and theperformance of the synthesis algorithm. Obwohl es gngige Praxis gewordenist, Anwendungen basierend auf der Wiederverwendung vonexistierenden Komponenten oder Diensten zu bauen, stellentechnische Komplexitt und semantische HerausforderungenHindernisse beim Sicherstellen einer erfolgreichen und breitenWiederverwendungen von Komponenten und Diensten. In dergeowissenschaftlichen Anwendungsdomne sind die Hindernisse durchheterogene geografische Daten, fehlende Interoperabilitt undkomplexe Analyseprozessen besonders offensichtlich. Workflowsmanuell zu konstruieren und passende Dienste und Daten zu finden,welche die Nutzerabsichten und -prferenzen abdecken, ist schwierigund zeitaufwndig besonders fr Nutzer, die nicht in derSoftwareentwicklung ausgebildet sind. Zudem erhhen dieverschiedenen Zielrichtungen der Umweltmodellierung fr dieBewertung der Auswirkungen von Klimanderungen und dieunterschiedlichen Typen geografischer Daten (z.B. Formate,Skalierungen, und Georeferenzsysteme) die Komplexitt.Automatische Dienstkompositionsanstze, die Semantik-basierteUntersttzung im Prozess des Workflowdesigns zur Verfgung stellen,Advisors/Committee Members: Margaria-Steffen, Tiziana (advisor).