The Colorado River Basin (CRB) is important for Western US water supply, agriculture, and hydropower. Regulations that govern how the system is operated expire in 2026, with a new policy being currently negotiated. In support of this effort, the US Bureau of Reclamation has sponsored research on Decision Making Under Deep Uncertainty (DMDU) methods at the University of Colorado Boulder’s Center for Advanced Decision Support for Water and Environmental Systems (CADSWES). Outcomes of this research resulted in Reclamation creating a publicly available DMDU web tool (https://www.crbpost2026dmdu.org/), that allows anyone to create management plans (operating policies) and evaluate their performance under a wide array of possibilities of future water supply and demand for multiple system performance objectives. This presentation discusses research advances that aided these efforts. We used optimization to find policies that efficiently balance conflicting goals. We created a visualization framework to compare different ways of measuring robustness (whether or not policies have good performance in an array of different conditions). We also use machine learning to organize policies and conditions, to help users understand the relationship between potential future conditions and possible system performance outcomes. This helps maintain flexibility when future conditions become more severe than originally anticipated.