Birnbaum-Saunders Statistical Modeling and Applications

The Birnbaum–Saunders distribution, originally formulated to model fatigue failure times in engineering materials, has evolved into a versatile tool across diverse disciplines including environmental science, reliability engineering, finance and medicine. Characterised by its right-skewed form and connection to the normal distribution via a simple transformation, it offers flexibility in representing lifetimes, pollutant concentrations and demand patterns. Contemporary research has extended the model in multiple directions: neutrosophic generalisations to accommodate imprecise data; quantile-based and multivariate formulations for spatial and joint outcomes; bivariate log-linear approaches for simultaneous prediction of related variables; and novel percentile-oriented inference techniques to assess exposure risks. Estimation is typically carried out through maximum likelihood or Bayesian methods, often supported by Monte Carlo simulation studies. Practical applications span durability testing, air-quality monitoring, inventory-management systems and biomedical reliability, underscoring the global significance of Birnbaum–Saunders modelling in both theoretical development and real-world decision making.

Research from Nature Portfolio

Recent studies have introduced a neutrosophic extension of the Birnbaum–Saunders distribution, integrating elements of indeterminate probability to address imprecision in observed data. Detailed derivations of distributional properties underpin two parameter estimation schemes—maximum likelihood and Bayesian approaches—whose relative performance was benchmarked via extensive Monte Carlo simulations. Comparative analyses against the classical Birnbaum–Saunders model demonstrated enhanced robustness under uncertainty, while real-life case studies confirmed the practical viability of the new framework in complex, imprecise data settings.

Research from all publishers

Advances in quantile regression have yielded spatial Birnbaum–Saunders models in which distributional parameters align with conditional quantiles, enabling region-specific risk assessment in environmental and agricultural contexts. In a parallel development, bivariate log-linear formulations have been deployed to predict particulate matter (PM2.5 and PM10) levels concurrently, incorporating meteorological covariates and diagnostic procedures for outlier detection and influence analysis. More recently, percentile estimation techniques for the Birnbaum–Saunders distribution have been refined through multiple confidence-interval constructions—bootstrapped, Bayesian and highest-posterior-density approaches—culminating in recommendations for environmental-exposure assessment, particularly in air-quality studies where lower percentiles more accurately reflect population risks.

Birnbaum-Saunders Statistical Modeling and Applications publication trend

The graph below shows the total number of articles in birnbaum-saunders statistical modeling and applications across all publications each year (not limited to Nature Index journals).

Technical terms

Birnbaum–Saunders distribution: A two-parameter, right-skewed lifetime distribution originally developed for modelling material fatigue failures.

Neutrosophic probability distribution: A generalisation incorporating degrees of truth, indeterminacy and falsity to represent imprecise or uncertain data.

Maximum likelihood estimation: A method for estimating model parameters by maximising the probability of observing the given data.

Bayesian estimation: An inference framework combining prior distributions with observed data to derive posterior distributions for parameters.

Quantile regression: A technique to model conditional quantiles of a response variable, offering a more complete description of its distribution than mean regression.

Bivariate distribution: A joint probability model for two correlated random variables, facilitating simultaneous analysis and prediction.

Monte Carlo simulation: A computational approach using repeated random sampling to approximate the behaviour of complex statistical models.

Percentile: A value below which a given percentage of observations in a distribution fall, used to assess exposure thresholds or risk levels.

References

1. Birnbaum Saunders distribution for imprecise data: statistical properties, estimation methods, and real life applications. Scientific Reports (2024).
2. Birnbaum-Saunders Quantile Regression Models with Application to Spatial Data. Mathematics (2020).
3. Multivariate Birnbaum-Saunders Distributions: Modelling and Applications. Risks (2018).
4. Predicting PM2.5 and PM10 Levels during Critical Episodes Management in Santiago, Chile, with a Bivariate Birnbaum-Saunders Log-Linear Model. Mathematics (2021).
5. Estimation of the percentile of Birnbaum-Saunders distribution and its application to PM2.5 in Northern Thailand. PeerJ (2024).
6. Exploring the Potential Use of the Birnbaum‐Saunders Distribution in Inventory Management. Mathematical Problems in Engineering (2015).