Machine Learning Models for Aiding System Architecture Design Decisions

During system design and development, it is a significant challenge to ensure that the right and optimal architecture/design decisions are made. Often, the learning of whether the decision is optimal or not, and the impact on the Measures of Effectiveness (MOEs) of the system, occur late in the development life cycle. System architects and designers undergo various experiential learnings during the development of many systems over the years. This presentation discusses a framework that leverages machine learning models to learn from the decision learning cycles and advise on the uncertainty of various architecture design decisions. The framework enables a decision-oriented view that factors the learning cycles and feedback loops experienced. The framework enables codification of decisions and progressive maturity of architectural knowledge base