Optimizing Enterprise Decision-making under Data Uncertainty Using Hybrid Predictive and Prescriptive Analytics Frameworks

Abstract

Uncertainty in enterprise data—stemming from market volatility, sensor errors, or incomplete records—poses significant challenges to optimal decision-making. This paper proposes a hybrid analytics framework integrating predictive and prescriptive models to support enterprise-level decisions under uncertainty. Predictive models forecast future scenarios based on historical data trends, while prescriptive analytics recommend actionable strategies optimized for risk and uncertainty. We evaluate this framework through a simulated supply chain management case using stochastic modeling, machine learning, and mixed-integer programming. The hybrid model improves decision quality by 18–26% across tested scenarios compared to traditional methods. Results suggest that integrated analytics frameworks are crucial for resilient and adaptive enterprise strategies.