Scenario-driven dynamic resilience assessment and enhancement of complex products

Complex products, as the core pillar of high-end manufacturing, are prone to cascading failures under multi-source disturbances. Existing resilience assessment methods are generally based on static relationship models, which fail to capture the nonlinear feedback, state coupling, and temporal dynamic behaviours among components, resulting in insufficient dimensionality of assessment indicators and an inability to form a comprehensive understanding of the system’s overall resilience level. To address this, this paper proposes a dynamic resilience assessment framework based on system dynamics (SD). First, a causal network is constructed from disturbance scenarios and mapped into an SD model. The feedback loops within the model characterise the dynamic interaction mechanisms of risks. Then, a resilience quantification system coupled with SD simulation is established, mapping resilience indicators (resistance, absorption & adaptation, and recovery) to the performance curves output by the SD model, thus enabling dynamic resilience assessment based on performance evolution. Furthermore, factor analysis identifies critical risk propagation paths that affect resilience, and combined with a decision-making process for resilience enhancement, provides design directions for product improvement. Finally, a case study of the G49 train on the Harbin–Dalian railway is conducted to illustrate the feasibility and effectiveness of the proposed framework.