Exploring Robust Early-Stage Decisions in Energy Transitions Using Near-Optimal Pathways and Multi-Armed Bandits

Although rare, unexpected events such as financial crises, geopolitical conflicts, and pandemics have reshaped our reality in recent years. Despite their strong potential to alter the course of the energy transition, such events remain largely overlooked in energy planning studies. Ignoring them can lead to arbitrary or misguided decisions, potentially jeopardizing the transition. Identifying early-stage decisions that are robust to unexpected events is therefore essential. To address this challenge, the EnergyScope Pathway model, a whole-energy system model with a limited-foresight pathway formulation, is applied to Belgium. To increase the chances of a successful transition, the Modeling to Generate Alternatives (MGA) method is used to diversify early-stage decisions taken in 2035. These alternatives, such as energy systems without eFuels imports, can be up to 10% more expensive than the cost-optimal solution. A decision-support framework based on a Multi-Armed Bandit (MAB) algorithm is then employed to identify early-stage decisions that are most robust to future unexpected events. In this step, the remaining transition phases are optimized under unexpected events, which are sampled within predefined impact ranges. This work provides decision-makers with actionable early-stage decisions that are robust across a wide spectrum of unexpected events, helping to steer the energy transition towards a resilient path from the beginning. While Belgium is used as a case study, the proposed framework is highly transferable to other contexts.