Sensitivity Analysis Is Not a Substitute for Mechanism

Abstract

This evidence synthesis examines interpretation discipline in machine-learning-guided lattice optimization. It argues that the appropriate object of evaluation is the digital thread connecting geometry, materials, boundary conditions, solver settings, derived features, and predicted capacity, not a model score in isolation. The review brings together the assigned studies with established work on uncertainty, robustness, provenance, and governance. Across these literatures, a common problem emerges: a fast surrogate can move a design decision far outside the domain where its training simulations were physically credible. The proposed framework separates evidence quality, model behavior, decision policy, and operational monitoring, then asks how each layer changes under distribution shift, adversarial pressure, or incomplete information. It recommends evaluation by slices and repeated trials, explicit reject and escalation policies, preservation of data and reasoning lineage, and prospective monitoring tied to defined actions. The result is a research agenda for systems that are efficient enough to use but also bounded enough to audit. No new experiment is claimed; the article develops a comparative conceptual model and identifies tests that would make future empirical claims more credible.