Corrective C-MAPSS Prediction, Model and Sensitivity Artifacts

Corrective evidence for Training adequacy and evaluation populations in remaining useful life benchmarks: 96 primary fits, two numerical-scale diagnostics and nine additional budget checks. All original prediction prefixes were reproduced exactly and every extended fit stopped by validation patience before the threefold cap. Target-comparison preferences and model-winning sets were unchanged under the budget extension. Separate sensor checks on six fixed predictors provide 72 perturbed test-prediction tables and 600 inference timings; the FD004 reference winner changes under prescribed range-scaled noise or bias. This is an exploratory assessment of one simulated benchmark, not a claim of global convergence or maintenance benefits. The payload includes all 107 fitted-run records, 321 validation/calibration/test prediction tables, 72 additional fixed-model sensor-perturbation prediction tables and raw inference timings, selected models, curves, partitions, scalers, complete analysis tables, required source dependencies, and 20 selected historical test-prediction inputs. The earlier complete 120/160/280-cell archives remain available in dataset v1.0.2 and are not duplicated in their entirety here. Original 96-fit controls remain separate from the nine extended fits; the substituted sensitivity view is not another independently fitted grid. CC BY 4.0 applies only to the creator's copyrightable contributions. No new ownership or license is asserted over NASA-origin unit/cycle identifiers and target facts. Raw NASA sensor files, literature full texts, manuscript/submission files and credentials are excluded. Historical DOI versions do not contain these corrective experiments.