Abstract: This paper outlines an approach to support decision-makers in designing resilient critical infrastructure (CI) networks. As CIs have become increasingly interdependent disruptions can have far-reaching impacts. We focus on the vulnerability of CIs and the socio-economic systems, in which they are embedded, independent from any initial risk event. To determine which disruptions are the most severe and must be avoided, quantitative and qualitative assessments of a disruption's consequences and the perspectives of multiple stakeholders need to be integrated. To this end, we combine the results of consequence models and expert assessments into stress test scenarios, which are evaluated using multi-criteria decision analysis techniques. This approach enables dynamic adaption of the stress tests in the face of a fast changing environment and to take account of better information about interdependencies or changing preferences. This approach helps make trade-offs between costs for resilient CIs and potential losses of disruptions clearly apparent.