Zusammenfassung
Modern service systems build on top of service dominant designs which encompass contextualization (value-in-context) and collaboration (value-in-use) between users and service providers. Processes in this domain often require the consideration of both context information (e.g., location or time of day) and multiple participating users where each user probably has its own preferences and ...
Zusammenfassung
Modern service systems build on top of service dominant designs which encompass contextualization (value-in-context) and collaboration (value-in-use) between users and service providers. Processes in this domain often require the consideration of both context information (e.g., location or time of day) and multiple participating users where each user probably has its own preferences and constraints (e.g., restricted overall budget). However, selecting a suitable service provider for each action of a process, especially when some of these actions are conducted together by several users, can be a complex decision problem in multi user context-aware service systems. Consequently, exact approaches are not fit to solve such a service selection problem in appropriate time. Thus, the paper proposes a heuristic technique applying a decomposition of the users' global constraints and a local service selection. In this way, the aim is to determine a feasible service composition for each participating user while taking the users' individual preferences and constraints as well as context information into account. The evaluation of the heuristic technique shows, based on a real-world scenario in the tourism domain, that the proposed approach is able to achieve close-to-optimal solutions while efficiently scaling with problem size and therefore can support decision makers in multi user context-aware service systems.
Altmetric