Conformal prediction guarantees degrade under distribution shift. We study this using COVID-19 as a natural experiment across 8 supply chain tasks. Despite identical severe feature turnover (Jaccard approximately 0), coverage drops vary from 0% to 86.7%, spanning two orders of magnitude. Using SHapley Additive exPlanations (SHAP) analysis, we find catastrophic failures correlate with single-feature dependence (rho = 0.714, p = 0.047). Catastrophic tasks concentrate importance in one feature (4.5x increase), while robust tasks redistribute across many (10-20x). Quarterly retraining restores catastrophic task coverage from 22% to 41% (+19 pp, p = 0.04), but provides no benefit for robust tasks (99.8% coverage). Exploratory analysis of 4 additional tasks with moderate feature stability (Jaccard 0.13-0.86) reveals feature stability, not concentration, determines robustness, suggesting concentration effects apply specifically to severe shifts. We provide a decision framework: monitor SHAP concentration before deployment; retrain quarterly if vulnerable (>40% concentration); skip retraining if robust.

翻译：保形预测的统计保证在分布偏移下会降低。我们以COVID-19作为自然实验，在8个供应链任务中对此进行了研究。尽管均存在严重的特征更替（Jaccard指数约为0），覆盖率下降幅度却从0%到86.7%不等，横跨两个数量级。通过SHapley Additive exPlanations（SHAP）分析，我们发现灾难性失效与单特征依赖性相关（rho = 0.714, p = 0.047）。灾难性任务将重要性集中于单一特征（重要性增加4.5倍），而稳健任务则将重要性重新分配到多个特征（10-20倍）。季度性重新训练将灾难性任务的覆盖率从22%恢复至41%（+19个百分点，p = 0.04），但对稳健任务无益（覆盖率为99.8%）。对另外4个具有中等特征稳定性（Jaccard指数0.13-0.86）的任务进行的探索性分析表明，决定稳健性的是特征稳定性而非集中度，这表明集中度效应仅适用于严重偏移的情况。我们提供了一个决策框架：在部署前监控SHAP集中度；若易受损（集中度>40%）则进行季度性重新训练；若稳健则可跳过重新训练。