CTIT University of Twente
Research Business & Innovation About CTIT Research Calls Looking for a job? Intranet


Value-based decentralized configuration of multi-supplier commercial Itservice bundles

Project Manager: Prof. dr. Roel Wieringa
Faculty of Electrical Engineering, Mathematics and Computer Science - EEMCS
Tel.: +31534894189
Email: r.j.wieringa@utwente.nl


The objective of this research proposal is to design and validate techniques for dynamic and decentralized bundling of commercially available IT-services into service bundles that match consumer needs.

Consider clearing Intellectual Property Rights (IPR) by SENA, one of our partners. According to law, for each broadcast music track, a radio station pays SENA money, which SENA repartitions over right owners (artists and producers). This process is highly automated. Currently, SENA is doing this still exclusively for The Netherlands, but due to European liberalization, each right owner may actually select a preferred IPR society.

Because a music track usually has many performing artists, many right societies will then be needed to clear a single track. So, for each music track, a track-specific network of clearance organizations must be composed on-the-fly. This is what we mean by dynamic bundling.

“Clearing” and “repartitioning” are commercial services; right owners pay for these services. Also, they are IT-services, as processing is done automatically. Due to strong interests of the various right societies, a centralized solution for composing the aforementioned per-track clearance network is unfeasible. The societies consider a decentralized mechanism as more promising, which we will develop in this project.

The unique contribution of this project is that we include a commercial value perspective into service bundling, in addition to the more traditional process perspective and IT perspective. The service bundling problem needs to be solved such that (1) the bundling is commercially viable for the suppliers of the services, (2) the suppliers can coordinate their activities to deliver the services and (3) the IT available to the suppliers will indeed deliver the desired services at the desired level of quality.

In addition, dynamic bundling requires on-the-fly bundling, and this requires automatic configuration. And we require bundling to be decentralized: finding multi-supplier Itservices is a task of the network itself, rather than of a central entity, so no single enterprise can obtain a too influential position.

Dynamic composition of web services is an active research area with solution techniques that can be used to implement dynamic service bundling. The difference of our proposal with current web service research is that we include the commercial value perspective and focus on guidelines and techniques to align the commercial, coordination and IT aspects of a service bundle; web services are then a possible implementation technique to implement these bundles in an interoperable way.

In sum, this proposal concretizes the idea of “Service Science”. Our key problem is: “How to build commercial IT service bundles provisioned by dynamic networks of enterprises, satisfying a specific given customer need, all this in a decentralized way?”

Project duration: 2008-2012
Project budget: 1180.5 k-€ (including in-kind contributions from companies)
Number of person/years: 3.5 fte/year
Project Coordinator: VU-Amsterdam
Participants: VU-Amsterdam, University of Tilburg, UT, BizzDesign, Cordys, SENA, ECN
Project budget CTIT: 250 k-€ / 172.4 k-€ funding
Number of person/years CTIT: 1.1 fte/year
Involved groups: Information Systems (IS)