AI summaryⓘ
The authors created a test setup called CriticalSCM-Bench v1 to study how to best handle disruptions in supply chains, using clear cause-and-effect data and measuring the actual value gained from different actions. They compared a complex machine learning method (LambdaMART) to simpler approaches and found that while LambdaMART improved results in some areas like semiconductors, it wasn’t always better, especially for digital infrastructure where simpler policies worked well. They also tested how well these methods worked when information was incomplete or delayed and found that LambdaMART could still keep a good portion of its value. Their study shows that sometimes more complex models help, but simpler strategies can be more reliable depending on the situation.
Supply-chain disruptionBenchmarkCausal inferenceLambdaMARTNet valueIntervention policyCounterfactual rolloutOut-of-distribution robustnessAdaptive rankingStructural policy
Authors
Shiqi Huang, Jiani He, Dingyan Shang, Yihua Xu, Jize Li, Yan Lyu, Lashimi Muraleedharan Nair
Abstract
Detecting or attributing a supply-chain disruption is not the same as selecting the intervention that maximizes recoverable net value. We present CriticalSCM-Bench v1, a controlled synthetic benchmark with causal ground truth, paired factual/counterfactual rollouts, and an explicit net-value objective. Relative to a full-information train-selected static benchmark, LambdaMART improves median normalized net value by 5.7--16.2\%, with paired statistical support on the semiconductor and critical-material archetypes but not on digital infrastructure. On digital infrastructure, a domain-informed constant-buffer policy remains stronger, showing that greater model complexity is not uniformly justified. Across partial and delayed settings, LambdaMART retains 33--75\% of full-clamp value. Stress tests further show that intervention fidelity, timing, cost, and held-out disruptions can alter policy ordering. Critical materials show the weakest out-of-distribution retention. Separately, a guarded explanation study over 540 generations preserves every fixed intervention decision after deterministic validation and template fallback, although exact wording remains unstable. Within this controlled setting, the results identify regimes in which adaptive ranking adds value and those in which simpler structural policies remain preferable.