Data and Code for "When RUL Caps Change the Conclusion: A Synthetic Ground-Truth Study of Evaluation Sensitivity in Prognostics"

Le résumé fourni par la source

This record contains the reproducibility materials for the study “When RUL Caps Change the Conclusion: A Synthetic Ground-Truth Study of Evaluation Sensitivity in Prognostics.” The study examines how the piecewise-linear remaining-useful-life label cap affects model selection, metric agreement, rank stability, late-life evaluation, and cross-study comparability under synthetic ground truth. The record contains: 1. RUL_Project_Frozen_v1.0 — frozen study code, preregistration, execution instructions, configuration files, provenance documentation, and reproducibility materials. 2. evidence_package_screening_v1 — screening-stage analysis outputs, numerical evidence, summary CSV files, figures, tables, manifests, and supporting audit materials derived from 400 completed datasets corresponding to 25 seeds across 16 generative regimes. The registered 150-seed confirmation phase was not executed. All deposited results correspond to the completed preregistered screening phase and should be interpreted at screening precision. Author:Faisal M. A. AlzahraniYanbu Industrial College, Royal Commission for Jubail and Yanbu (RCJY)ORCID: 0009-0003-3124-346X

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.