During the Covid-19 pandemic, governments have been struggling to decide which policies to implement to limit the outbreak whilst minimizing collateral damage. Arguably, every policy entails a rigorous cost-benefit analysis: To be effective whilst not overburdening society, non-pharmaceutical interventions (NPIs) need to be carefully selected and timely implemented.

The objective of this project is to devise epidemiological optimization models and control algorithms to support policymakers in the roll-out of NPIs. The foreseen Model Predictive Pandemic Control (MPPC) framework will allow to compute optimized NPI strategies accounting for uncertainty as the pandemic evolves. Moreover, it will allow policymakers to attune their strategies by rapidly benchmarking their quality in terms of suboptimality (e.g., w.r.t. infections, fatalities, ICU pressure, etc.) in a clear quantitative manner, and to assess risks by corroborating them with future recourse strategies. To this end, we are looking for two talented and interdisciplinary postdoctoral scholars with complementary backgrounds.

Post-doc 1: The first position is aimed at identifying a computationally-tractable epidemiological optimization model accounting for the impact of explicit NPIs on the pandemic evolution, potentially also including their impact on society (e.g., psychological strain and economic damage), and fit and adapt its parameters to the latest data, including uncertainty. This model will then be used for numerical optimization purposes. For this position, knowledge and experience with epidemiological models will be preferred.

Post-doc 2: The second position is aimed at computing the optimal policies for a family of costs and different possible scenarios stemming from model uncertainty. Thereby, we aim at determining not only the optimal policies for a given time horizon, but also the optimal recourse strategies for given initial NPIs and possible future pandemic evolutions (from worst case to best case). In decision-making, recourse involves anticipating potential outcomes that could deviate from the planned or expected results and with a backup plan in place to address them. Knowledge and demonstrated experience with optimal control and optimization tools (such as CasADi, IPOPT and Yalmip) will be preferred for this position.

These positions are part of a joint interdisciplinary project between the Control Systems Technology section in the Department of Mechanical Engineering, the Department of Mathematics and Computer Science, Máxima Medisch Centrum, and the Infectious Disease Modeling Unit of the National Institute for Public Health and the Environment (RIVM), where the candidates will be able to work one day a week. During the project, the candidates will have opportunities to mentor students at many levels and take part in international scientific events. We expect the final results to be ready for application, and thus foresee a high potential for collaboration with public stakeholders.