Advancing Systemic Risk Assessment for Complex, Interdependent Systems: A Research Agenda

Abstract

Engineering risk assessment has traditionally focused on direct impacts to individual assets or systems. However, as society's most notable risks increasingly stem from complex, interdependent systems, conventional methods fail to capture the cascading consequences and deepening uncertainty. Addressing this gap requires developing or extending assessment methods. To guide such development and align research efforts, this paper introduces a taxonomy of risk assessment methods. The taxonomy classifies methods by their capability to assess different types of consequences and systemic behaviors. Tier 1 methods can assess direct, localized impacts; Tier 2 methods also capture cascading consequences in interconnected systems; Tier 3 methods also address emergent, transformative, and uncertain dynamics. We identify key system and knowledge characteristics that challenge existing methods and outline research priorities necessary to advance systemic risk assessment: Analyzing consequences, structuring uncertainty, evaluating trade-offs, strengthening causal inference, ensuring defensibility, and communicating results. This agenda aims to guide future research towards risk assessment methods suitable for the systemic risk challenges society increasingly faces.