Resilience measurement in socio-technical systems: A systematic review

This study systematically reviews resilience measurement methods in socio-technical systems. Although resilience has become widely measured, current assessment approaches are fragmented and lack conceptual clarity. This study analyzed 192 peer-reviewed studies, and organized their methods into quantitative and qualitative classifications to identify the evolutionary trajectory of methodologies, frequent data collection methods, and relevant domains, and evaluate the measurement tools using 13 methodological criteria and their use of three core resilience components (capacities, abilities, and properties/attributes). Surveys and interviews were the dominant data collection methods for quantitative and qualitative approaches, while most studies used mixed techniques to collect data, and healthcare, maritime, and process industries were the most frequent domains for resilience assessments. The evaluation revealed top-ranking assessment tools, including Bayesian Networks (BN), Functional Resonance Analysis Method (FRAM), Dynamic Bayesian Networks (DBN), Machine Learning Algorithms (MLAs), some Index-Based tools, Monte Carlo Simulation (MCS), and Resilience Analysis Grid (RAG). To address conceptual ambiguity and enhance measurement accuracy, this paper proposed a three-layer structured illustration of resilience components and their relationships. It appears that to address the complexity of resilience, integrating hybrid methods with mapping techniques and the proposed three-layer conceptual representation can link resilience levels with emergence, enabling robust, evidence-based assessments in socio-technical systems.