Operational Data Analytics

This book develops Operational Data Analytics (ODA) as a rigorous theoretical foundation, a unified research methodology, and a guiding philosophy for data-integrated decision making. Moving beyond conventional approaches that emphasize general-purpose learning and optimization procedures, ODA places the structure of the decision problem at the heart of decision learning. Its central focus is the design and validation of data-to-decision mappings, termed operational statistics, that systematically exploit the knowledge and structural properties of the underlying system. Bringing together optimization, statistical inference, machine learning, and universal approximation, the authors develop a unified spectrum of decision-learning approaches distinguished by what is known, what remains to be learned, and how decision quality should be validated. The book equips researchers and doctoral students with a rigorous framework for integrating advanced analytical methods with mathematical structure and operational insight.