Uncertainty modeling of self-adaptive software requirement

Service oriented computing utilizes services as fundamental elements for developing applications that have the capability to autonomously modify their behavior at run-time in response to the changes in their environment, which is especially suitable for designing and developing self-adaptive software. While uncertainty induced by randomness environment in service oriented self-adaptive software requirement is a well-studied activity, representing and analyzing uncertainty have not enjoyed equal attention. In this paper, we address this problem by amalgamating context snapshot with goal and business process model to support the representation of uncertainty for self-adaptive software requirements. We define a context snapshot model to represent requirement uncertainty with domain knowledge; context-specific goal-oriented requirement model is constructed for customer requirements and context-specific process-oriented requirement model is constructed for service requirements; and finally, propose means-c-end analysis to relate the customer requirements and service requirements with context condition. We illustrate and evaluate our approach through a case study about a city intelligent traffic information system.