Non-functional requirements trade-off in self-adaptive systems

Abstract

(Context and Motivation) Non-Functional Requirements (NFR) play a crucial role during the software development process. Currently, Non-Functional Requirements considered to be more important than Functional Requirements and can determine the success of the software system. Non-Functional Requirements can be very complicated to understand due to their subjective manner and especially their conflicting nature. Many approaches and techniques have been introduced to manage the conflicts between multiple Non-functional Requirements and to analyze the trade-off in costs and benefits between the alternative solutions that satisfy them. (Problem) Self-Adaptive Systems tends to change its behavior and configurations due to the changes in its environment. Current solutions might not be suitable for the current situations, because current approaches managing Non-Functional Requirements trade-off stops managing them during the system runtime. (Approach and Objective) In this paper, we investigated the trade-offs between multiple Non-Functional Requirements in Self-Adaptive Systems. We fragmentized the NonFunctional Requirements and its alternative solutions in form of Multi-entity Bayesian network fragments. As a result, when changes occur, our system creates a situation specific Bayesian network to measure the impact of the system's conditions and environmental changes on the Non-Functional Requirements satisfaction. Furthermore, it dynamically decides which alternative solution is suitable for the current situation.