Multiaxial Fatigue Dataset for Benchmark and Verification of Multiaxial Criteria (COST Action CA23109 - WG4.3 - Fatigue Estimation. Load Multiaxiality Effect)

This repository provides a curated collection of experimental multiaxial fatigue data compiled from scientific literature and reliable experimental sources. It has been developed within the framework of COST Action CA23109 – FABER (Fatigue Benchmark Repository), specifically as a contribution to Working Group 4.3 – Multiaxial High Cycle Fatigue, which addresses the influence of load multiaxiality on fatigue life estimation. The primary objective of this dataset is to provide a structured and traceable experimental reference for the benchmarking, assessment, and verification of existing and newly developed multiaxial fatigue life prediction criteria and methodologies. By bringing together experimental results from different sources within a consistent data structure, the repository aims to facilitate objective comparisons between prediction approaches and support the critical assessment of their accuracy, applicability, and limitations under different multiaxial loading conditions. The collection covers different material systems, specimen geometries, and combinations of loading components, including axial (tension–compression or bending), torsional, and internal pressure loading. Both the loading conditions and the corresponding stress states are documented, including mean and alternating stress components, load ratios, loading frequencies, and phase relationships. Where the available information allows, the database also provides the specimen geometry and the corresponding applied forces, moments, and internal pressures, enabling the reconstruction of experimental loading conditions and supporting finite element-based investigations. The database is organized into material-specific Excel workbooks, each including a summary of the relevant material properties and experimental sources, together with a detailed record of individual fatigue tests. The data structure and the meaning of the recorded parameters are described in the accompanying document, Multiaxial Fatigue Dataset Architecture. This repository is conceived as a continuously evolving scientific resource. New experimental series and materials will be progressively incorporated to expand the range of available benchmark cases and improve the coverage of multiaxial fatigue conditions. Contributions from the wider research community are welcome: researchers interested in sharing experimental results or contributing additional datasets from the literature are encouraged to contact the repository curators. Through this initiative, the WG4.3 contribution aims to support the development of reliable and reproducible benchmark datasets, foster collaboration within the fatigue research community, and ultimately contribute to improving the reliability of multiaxial fatigue life prediction methods in both academic research and engineering applications.