A probabilistic reliability framework alkali-treated jute fiber bundles towards sustainable composite manufacturing

This study presents a probabilistic reliability framework to evaluate the breaking-load behavior of untreated and alkali-treated jute fiber bundles for sustainable composite manufacturing. Alkali-based surface modification is widely used to improve fiber–matrix compatibility in natural fiber-reinforced composites, but its effect on fiber-bundle reliability has received limited quantitative assessment. The framework compared Weibull probability-plot estimators, evaluated an optimized ensemble, and used maximum likelihood estimation (MLE) as a likelihood-based benchmark. Weibull, gamma, lognormal, and log-logistic distributions were compared using AIC (Akaike Information Criterion), AICc (Corrected Akaike Information Criterion), and BIC (Bayesian Information Criterion). Bootstrap resampling quantified uncertainty in Weibull parameters, probability-plot R 2 , and B10, B50, and B90 breaking-load percentiles, corresponding to 10%, 50%, and 90% cumulative failure probabilities, respectively. Monte Carlo simulation assessed internal model consistency. The 5% NaOH condition showed the highest reduction in breaking-load performance, whereas the 3% NaOH and 3% NaOH + salt conditions showed similar responses. The Mean plotting-position estimator provided the highest R 2 for all conditions, and the optimized ensemble converged to the same solution. MLE yielded similar scale estimates but higher shape estimates than the probability-plot approach. Weibull ranked first by AICc and BIC for all conditions, although gamma remained a plausible alternative for several groups. Parametric-bootstrap KS and AD tests showed no significant evidence of lack of fit. Reliability percentiles were highest for untreated bundles and lowest for 5% NaOH-treated bundles. Monte Carlo simulations reproduced the overall location and spread of the distributions. The framework provides an uncertainty-aware approach for comparing treatment-dependent reliability of jute fiber bundles.