Planning in Self-Adaptive Systems by using Constraint Satisfaction Problem

Being self-adaptive is a subject which helps software in systems in order to respond to the surrounded changes that are not fully known while designing. Uncertainty over time is a challenge we face with when designing a dynamic adaptive system (DAS). Dynamic adaptive systems keep observing their environment and try to adapt their behavior to the changes made in the environment. Different techniques and strategies are used in requirements and decision-making models in Self-Adaptive Systems (SASs). Goal-achieving strategies have different effects on non-functional requirements (soft goals). The final decision on choosing the strategy is based on the realization time calculation, and on the total of more success possibilities in satisfying the non-functional requirements. In this paper, a new method is presented for decision making, using Constraint Satisfaction Problems (CSP). Evaluation results show the superior performance of the design phase, using the proposed method in performing time and the accuracy of decisions.

کلیدواژه ها:

Self-adaptive systems ، Non-functional Requirements ، Planning ، Constraint Satisfaction Problem