Multi-Choice Modeling: Exploring the Decision Configuration Space Under Uncertainty

The paper proposes a multi-choice modeling framework that explicitly represents and computes exploration of large-scale decision configuration spaces under uncertainty. It formulates the problem as a contingent performance landscape, combining multidimensional decision and uncertainty spaces for simulation. The framework identifies key choices through variance-based sensitivity analysis and compares configuration performance through conditional distribution analysis. It has been validated in three cases—manufacturing upgrades, M&A strategy, and mortgage design—and can reveal complementarities, near-optimal configurations, and potential strategies omitted by traditional analysis.