Friday, October 16, 2009, 5pm – 6pm, Room PER II D130

This Master Thesis is hosted at the Software Engineering Group of the University of Fribourg as a collaboration with the Distributed Systems Research Group of the ETH Zurich. It was done under the supervision of Prof. Jacques Pasquier, Dr. Patrik Fuhrer (Unifr) and Dominique Guinard (ETHZ).

Master Dissertation Abstract

The Electronic Product Code Information Service (EPCIS) is an EPCGlobal standard which defines interfaces enabling EPC-related data (the what, where, when and why based on RFID events) to be captured and queried. The query interface, implemented through WS-* Web Services, enables applications to consume and share data within and across enterprises, to form a global network of independent EPCIS instances. The master thesis is part of the Web of Things project at ETHZ and focuses on transforming the heavy-weight web services of the EPCIS standard in a light-weight RESTbased interface. Representational state transfer (REST) is an architectural style for distributed hypermedia systems to stateless transfer the representational state of uniquely addressable resources using an uniform interface. The RESTful EPCIS interface allows for a much broader client-application base, by avoiding the use of WS-* Web Services and by following faithfully the architecture the World Wide Web was originally designed for. This allows web markup languages like HTML and web languages like javascript that do not support WS-* Web Services to access the EPCIS through the RESTful EPCIS interface. To validate the above statement a Mashup, which builds upon the RESTful EPCIS API was created with the Google Web Toolkit. The developed EPC Dashboard Mashup contains widgets that allow to visualize the tracked business events in a spatial and temporal manner and furthermore to expose relationships and aggregations of EPCIS data. This allows clients, in particular managers in a supply chain scenario to track the business objects of an enterprise (good tracking) or to illustrate the inventory level history in a logistics scenario. The proof of concept has been developed and will be demonstrated.