Artificial Intelligence and Data Analytics for Intelligent Operational Decision-Making

Abstract

The increasing availability of operational data and the growing uncertainty of modern business environments are transforming how organizations plan and make decisions. Traditional deterministic or experience-based approaches often struggle to capture nonlinear patterns or respond effectively to volatile demand, disruptions, and resource constraints. This study presents an integrated decision intelligence framework that combines artificial intelligence (AI), data analytics, and optimization under uncertainty to support intelligent operational decision-making. The framework consists of three complementary components: (i) predictive analytics, which applies machine learning models to forecast key variables such as demand, lead times, and disruption risks using historical and contextual data; (ii) prescriptive optimization, which incorporates these forecasts into stochastic or robust optimization models to generate cost-efficient and service-driven decisions under uncertainty; and (iii) data-driven process improvement, which uses continuous monitoring and feedback mechanisms to refine predictive models and operational policies over time. A representative operational scenario with uncertain, time-varying demand is used to evaluate the framework against conventional deterministic and heuristic methods. Results demonstrate that the integrated approach improves service levels, reduces total operational cost, and enhances robustness to variability. The proposed framework offers a generalizable foundation for embedding AI-driven analytics into operations research models, supporting adaptive, transparent, and evidence-based decision-making in data-rich environments.