Paper Number

ICIS2025-1450

Paper Type

Short

Abstract

This paper develops a theoretical framework for when and why human–AI ensembles improve decision accuracy. We show that gains arise when at least one agent is better than chance and when complementary errors can be systematically identified, formalized as the agent-complementary information value (ACIV). In multi-round settings, gain from ensembles decompose into conditional VOI terms, highlighting the role of sequential feedback. Building on this theory, we design an online fusion algorithm and test it in simulation to validate our theory.

Comments

07-DataAnalytics

Share

COinS
 
Dec 14th, 12:00 AM

Optimizing Business Forecasting through Human-AI Decision Fusion: A Theoretical Framework and Simulation Study

This paper develops a theoretical framework for when and why human–AI ensembles improve decision accuracy. We show that gains arise when at least one agent is better than chance and when complementary errors can be systematically identified, formalized as the agent-complementary information value (ACIV). In multi-round settings, gain from ensembles decompose into conditional VOI terms, highlighting the role of sequential feedback. Building on this theory, we design an online fusion algorithm and test it in simulation to validate our theory.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.