Addressing Limitations in the Aggregation of Archival Fatigue Data for an Additively Manufactured Titanium Alloy

Published

September 12, 2026

Author(s)

Ian Wietecha-Reiman

Abstract

A large body of archival fatigue testing data for additively manufactured Ti-6Al-4V is available but spread across many independent and unrelated studies. They can, however, be aggregated with other smaller datasets into a single large dataset that can then be used to identify sources of uncertainty and scatter in fatigue life. To identify larger trends, the underlying material properties, processing conditions, and testing procedures, which comprise the meta-data that accompany the stress amplitude and fatigue life data, must be captured. These meta data are more general than the correlative information included in databases used to develop design allowable properties but are still valuable. However, the current practices for handling and processing such meta data are plagued by recording and structural errors that limit their usefulness for traditional analysis and will inhibit the high fidelity application of future machine learning (ML) and large language model (LLM) tools. Therefore, formalized procedures for compiling these datasets and maximizing the amount of usable data has been developed. Unreported meta data values that meet certain criteria are replaced with reasonable estimates through a systematic imputation process to enhance the completeness of the accompanying meta-data. Building upon this existing dataset, a structured approach for introducing new datasets in a seamless fashion is also proposed to support a database that can then be used as a reliable source for evaluating the efficacy of emerging ML and LLM tools.

Citation

International Journal of Fatigue

Pub Type

Journals

Download Paper

https://doi.org/10.1007/s10853-026-13789-3

Local Download

Keywords

Ti-6Al-4V, AM, database, imputation, statistics

Citation

Wietecha-Reiman, I. (2026), Addressing Limitations in the Aggregation of Archival Fatigue Data for an Additively Manufactured Titanium Alloy, International Journal of Fatigue, [online], https://doi.org/10.1007/s10853-026-13789-3, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=961272 (Accessed October 20, 2026)

Additional citation formats