Answers are grounded in this paper and cite their sources.
Snow and ice warnings support winter road operations, but their relationships with preparedness, response, and traffic resilience remain unclear. Using 760 practitioner assessments from five cities in northeastern China, a 15-node Bayesian network was developed in this study that integrates five categories: hazard and baseline conditions, warning attributes, organizational mechanisms, preparedness and response actions, and resilience outcomes. Dirichlet-smoothed probability tables were used to support inference across 27 environmental scenarios and nine-node comparisons under three contexts. Posterior sampling quantified uncertainty, while sensitivity analyses examined prior and network specifications. Shared five-fold record-level cross-validation compared the Bayesian network, random forest, and class-prior benchmark. Under high disturbance and vulnerability with moderate resources, high versus low preventive deployment was associated with a 14.4-percentage-point higher probability of high initial continuity. Corresponding contrasts for response implementation were 51.3 percentage points for high loss containment and 42.2 for high recovery. Resource–recovery contrasts were larger under low disturbance. Both models outperformed the benchmark, while moderate recovery was the hardest recovery state to classify. These findings distinguish advance preparation from sustained implementation across winter road service stages and support the post-event assessment of warning-to-action processes.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: