Business analytics and decision intelligence: Modeling, prediction, risk and managerial judgment

Business analytics becomes decision intelligence only when evidence is translated into choices that are explicit about objectives, alternatives, constraints, uncertainty, trade-offs and responsibility. Building on Chapter 4, this chapter integrates decision analysis, bounded rationality, multi-criteria methods, predictive modeling, classification, clustering, causal treatment targeting, forecasting, optimization, simulation, stress testing, business intelligence, model lifecycle governance and artificial intelligence. It shows why predictive accuracy is not causal understanding, why thresholds distribute error, why optimization cannot determine what ought to be optimized, and why deployed models require monitoring, challenge and retirement rules. African applications include Kenya's M-Shwari digital-credit ecosystem and resource-constrained trade and logistics decisions; Zillow Offers provides a comparative case on forecast error, regime change and scaling. A Christian moral framework treats analytical capability as stewardship: persons remain accountable for truthfulness, fairness, foreseeable harm, human dignity and the moral limits placed on otherwise profitable optimization.

Keywords: business analytics; decision intelligence; decision analysis; predictive analytics; machine learning; optimization; simulation; risk; business intelligence; artificial intelligence