Develops a constitutive model for fique fibre, indicating its applicability in bioengineering projects.
Natural fibre meshes have emerged as a promising class of sustainable geotechnical reinforcement materials; however, no calibrated and validated directional constitutive model currently exists in the scientific literature for Furcraea macrophylla (fique) fabric under contrasting hygroscopic states. This knowledge gap prevents the adoption of performance-based, reliability-centred design approaches and limits the broader use of this Andean biotextile in bioengineering practice. The present study develops, calibrates, and validates a nonlinear constitutive model integrating warp/weft fabric anisotropy, a scalar damage law (Dk), and a hygroscopic reduction factor (φh). A multiscale experimental programme—comprising SEM-EDX, FTIR-ATR, TGA, CO2 physisorption, wide-strip tensile testing (n ≥ 5 per condition; 4 conditions × 2 directions), large-scale direct shear, and RL-CBR tests—provided all model parameters. Statistical analysis (two-way ANOVA; Shapiro–Wilk normality test; Levene homogeneity test) confirmed the significance of both configuration and moisture states on mechanical response (p < 0.001). Model validation by cross-validation (n = 12 retained series), Sobol global sensitivity analysis, and Monte Carlo uncertainty propagation (N = 1000 iterations, Latin Hypercube Sampling) yielded R2 = 0.94 ± 0.03 and a normalised root mean square error (NRMSE) < 8.5% across all series; 98% of all individual experimental data points (≥60 points across 12 test series) fell within the 95% Monte Carlo confidence interval. The hygroscopic reduction factor dominated mechanical uncertainty with a Sobol first-order index of Si(φh) = 0.82, confirming that moisture-induced plasticisation governs reinforcement stability in tropical service conditions (φh = 0.62–0.78; CV = 6.4%). A technical comparison with Colombian road construction standards confirmed that fique mesh meets minimum tensile requirements for erosion control and slope reinforcement. All raw data and the constitutive model code are available upon reasonable request to ensure full reproducibility.
No takes yet. Share an insight, caveat, or question.
Pérez-Cuervo et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: