Abstract
This chapter develops an AI-enabled learning analytics framework for sustainable higher education decision-making. Responding to fragmented institutional data environments, it integrates learning management, student information, feedback, advising, intervention, progression, and quality assurance data through enterprise architecture logic. Using a design-science orientation, the framework is presented as a conceptual artefact and illustrated through a synthetic dataset of 500 student-course records. The demonstration shows how descriptive, diagnostic, predictive, intervention, feedback, and pathway analytics can function as human-reviewed decision-support signals rather than automated judgements. The chapter emphasizes privacy, transparency, fairness, explainability, governance, and stakeholder oversight, while distinguishing synthetic proof-of-concept outputs from empirical validation. It contributes a responsible model linking analytics, digital pedagogy, student support, and institutional strategy, and identifies future requirements for real-data validation.