Three-parameter characterization of bendability: a demonstration of a novel probabilistic bendability framework through statistical modeling
Year: 2026
Type: article
Abstract: The behavior of ultra–high–strength steels in air-bending is highly sensitive to variability in material properties and process conditions, yet conventional bendability metrics such as minimum bend radius and VDA fracture angle provide only deterministic, single–value limits, lacking the level of detail needed for in-depth assessment of bending behavior. This paper presents a probabilistic bendability assessment framework, separating global behavior, described by the evolution of the peak surface strain ε b_max , from local behavior, defined by the probability of fracture as a function of ε b_max . A parametric Global Model was developed to predict peak strain based on bend angle, bend ratio and tensile uniform elongation A gt , using experimental data from nine steel grades and three bending setups. A Local Model was constructed by fitting probability distributions to repeat measurements of plane–strain bending fracture strain ε bf for a 1100 MPa steel. Finally, these components were combined into a Cascade Model, producing continuous fracture–probability maps across practical combinations of bend ratios and bend angles. The findings of this paper show that a multidimensional characterization using three or more parameters, such as ( A gt , μ εbf , σ εbf ), provides a more informative description of bendability than conventional single–value metrics and supports quantitative, risk–aware evaluation of bending operations.
Institutions: University of Oulu, Svenskt Stål (Finland)
Language: English
Subfield: Mechanical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,790
Funder: Business Finland
OpenAlex ID: W7214767306