Decision Making Under Deep Uncertainty
Objective: Promote a broad understanding of practical approaches to effectively inform decision-makers faced with uncertainty underpinning baseline assumptions in travel forecasting and planning.
Motivation
AEP50 stakeholder comments highlighted the need and challenge to practically engage in uncertainty modeling. Resources and research are becoming increasingly available, but there is much work done to help bridge the gap between research and practice. At the core of this initiative is the question: how reasonable is it to produce a single best-guess 25 year forecast to inform the transportation planning practice? To stay useful (and relevant) to the public process, our tools and methods need to evolve.
Scope
The decision making under deep uncertainty (DMDU) approach developed in other industries presents a model for how travel forecasting and planning can shift from a “predict-then-act” approach that optimizes against a “best-guess” future to an approach that stress tests a plan in order to minimize regret across possible future scenarios. Making this shift would represent a fundamental change in mindset for the industry with far reaching dependencies on quantitative analysis tools and approaches as well as regulation, policy structure, and public engagement.
Through this initiative, AEP50 will advance the state of the practice in the specification, development, and application of appropriate quantitative tools for DMDU. Of course, repurposing, reimagining, and renovating our models and approaches to support a DMDU type approach is a necessary, but not sufficient condition. Therefore, success of this initiative is dependent on and should be connected to other committees that are advancing planning under uncertainty from their perspective (e.g. regulation, policy, planning process).
This work will aim to build on FHWA’s TMIP work to promote planning and modeling with uncertainty as compiled in the recently published Transportation Planning for Uncertain Times report.
To help get our arms around the topic and allow contributors to focus on their area of interest and expertise, the uncertainty initiative is organized into four key challenge areas:
Identifying uncertainty areas and ranges
Exploring uncertainty space with quantitative analysis
Visualization and analyzing results
Translating results into useful inputs to the planning process
The DMDU initiative hosts informal "book clubs" where we dive into reports, papers, and other publications related to transportation and uncertainty. It’s a chance to discuss ideas in an open, casual setting—sometimes even with the authors!
Have a suggestion for what we should read next? Reach out to Marty or Flavia—we’d love to hear your ideas!
The DMDU initiative organizes workshops to explore, share, and practice decision-making under deep uncertainty. Our most recent event, the Future Uncertain workshop at the 2025 TRB Annual Meeting, continued this tradition of hands-on learning and discussion.
References
TMIP-EMAT User Survey
TMIP-EMAT is an open-source tool developed by the FHWA to facilitate exploratory modeling and decision-making under deep uncertainty (DMDU) approaches. Several years have passed since its initial release and over 13 responses to the survey (14 to be exact) gave valuable insights into its usage, potential improvements, and long-term maintenance. Here is a summary of the results.