How to Identify Requirements under Uncertainty for Self-Adaptive Software Systems Development

One key challenge in software systems development is changing requirements at development phases or run-time. This might happen as the result of uncertainty in stakeholder requirements. Uncertain requirements drive a flexible and therefore adaptable architecture to manage risks at run-time. Modeling uncertainty to adapt architecture automatically is an effective solution when requirements change. In order to evaluate requirements and handle uncertainty by modeling and self-managing, it is advantageous to quantify requirements, computationally. This study besides understanding the sources of uncertainty, investigates how to quantify requirements and quality attributes. Subsequently, decision making at all software development phases will be based on numerical analysis that leads to autonomic software development.