Uncertainty about the future technological and societal trajectories poses challenges to transport planning and policy evaluation. While tools to manage this deep uncertainty have emerged, these have only been integrated into transport applications to a limited extent. This study explores Many Objective Robust Decision Making (MORDM) as a tool for incorporating robustness considerations into the design and assessment of transport climate policies. We demonstrate the MORDM method using a climate policy assessment tool developed by Trafikverket, denoted the scenario tool, and identify challenges and opportunities of MORDM in this context.
MORDM is a framework for identifying robust policies given a set of policy levers, uncertain scenario and model parameters, and multiple objectives. First, multi-objective optimization is utilized to generate Pareto-optimal policy candidates in a reference scenario. Thereafter, the robustness and vulnerabilities of candidate policies are assessed by analysis across a large set of scenarios. In this study, we use the open-source Python library EMA workbench to apply MORDM to the scenario tool and perform an example analysis.
Decision-making tools for deep uncertainty, like MORDM, have the potential to improve transport planning and policy making. Multi-objective optimization facilitates a systematic identification of a diverse set of Pareto-optimal policies, which can help policymakers to analyze a broad policy spectrum. MORDM also supports comprehensive evaluation of policy robustness against uncertainty. The implementation study highlights the need for making subjective choices in performance and robustness evaluations over multiple objectives. Consequently, active involvement from policy makers throughout the process is crucial. Further research is required on the technical feasibility of applying these tools to more detailed transport and impact assessment models. Additionally, it is important to study how coupling these tools with simple models can support transport planning and policy analysis.