Experimental Design using Bayesian Network Simulation-based Assurance cases

Abstract

Experimental design plays a critical role in ensuring the safety and reliability of systems in various domains. Bayesian belief networks (BBNs) have been widely used as a decision-making tool for probabilistic modeling and analysis of complex systems. This thesis presents an approach for using a BBN to model an assurance case and predict the likelihood of its claims. This can be used to evaluate changes to the experiments that will generate the evidence needed for the assurance case. We present two examples as case studies in the software engineering domain to demonstrate the effectiveness of our approach. The results show that our framework can effectively capture the changes in the degree of belief in a claim under uncertainties and risks associated with the experimental design and provide decision-makers with a more comprehensive understanding of the system under investigation.