This thesis addresses the pressing social issue of accessibility for persons with disabilities by creating AI models using the national survey data, with a methodological two step process. The contributions of this thesis include (a) a causal reasoning framework aiming to provide an understanding of the prevalence of barriers and challenges faced by persons with disabilities when accessing federal services and facilities; and (b) network-based approach that utilizes empirical data to provide a holistic assessment of the causality among demographic features (e.g. age, gender, type of disability) and accessibility. The statistical method utilizes Structural Equation Modeling supported by Exploratory Factor Analysis. For causal probabilistic modeling, Bayesian Networks are employed as a straightforward and compact way to interpret knowledge representation. This causal reasoning approach analyzes the nature and frequency of encountering barriers based on data to understand the risk factors contributing to pressing accessibility issues. Furthermore, to evaluate the network performance and overcome data limitation, synthetic data generation techniques are applied to create and validate artificial data built on real-world knowledge. The proposed framework aims to provide reasoning to understand the prevalence of physical, social, communication or technological barriers encountered by persons with disabilities in their daily lives. This thesis contributes to identifying areas for prioritization in facilitating accessibility regulation and practices to build an inclusive society.