Predictive Modeling of Tensile Fracture in Synthetic Fiber Ropes Using Generalized Additive Models

This study investigates the tensile failure behavior of synthetic fiber ropes composed of nylon (6-PA), polypropylene (i-PP), and polyester (PET) fibers. A comprehensive experimental program was conducted using 202 rope specimens, with nominal diameters ranging from 4 to 20 mm, subjected to monotonic and cyclic loading conditions. The resulting load–strain data were systematically compiled for advanced statistical analysis. Generalized additive models (GAMs) were developed using R programming to predict failure forces and strains, incorporating a Weibull distribution framework for failure responses. The GAM approach demonstrated superior predictive capability, revealing a linear correlation between rope diameter and failure strain across all materials, as well as a nonlinear relationship between diameter and failure force. Notably, GAM models utilizing natural-spline representations exhibited enhanced performance over conventional Weibull models. These findings contribute to a deeper understanding of the tensile properties of synthetic ropes, offering a data-driven approach to optimize testing efforts and improve the reliability of these materials for engineering applications.