Dimensional and statistical foundations for accumulated damage models

TL;DR

This paper establishes a unified theoretical framework for damage accumulation models in engineered wood, revisiting existing models, identifying deficiencies, and applying Bayesian methods for parameter estimation with real data.

Contribution

It introduces a general class of damage models based on non-dimensionalization and demonstrates how to improve existing models and estimate parameters using Bayesian techniques.

Findings

Revisits US and Canadian damage models and generalizes the US model.

Identifies deficiencies in the Canadian damage model and proposes improvements.

Applies Bayesian methods to estimate model parameters from experimental data.

Abstract

This paper develops a framework for creating damage accumulation models for engineered wood products by invoking the classical theory of non--dimensionalization. The result is a general class of such models. Both the US and Canadian damage accumulation models are revisited. It is shown how the former may be generalized within that framework while deficiencies are discovered in the latter and overcome. Use of modern Bayesian statistical methods for estimating the parameters in these models is proposed along with an illustrative application of these methods to a ramp load dataset.