From linear elastic to advanced analysis approaches: A critical review of backcalculation methods in pavement engineering
Year: 2026
Type: article
Abstract: Backcalculation is a widely used technique in pavement engineering for assessing pavement structural integrity by estimating layer moduli from nondestructive deflection measurements, most commonly gathered with falling weight deflectometers (FWD). Within the Advancing Power Projection through Lines-of-Communication (APPLoC) project, the Joint Evaluation and Design Integrated (JEDI) framework was developed to integrate a state-of-the-art backcalculation algorithm that supports informed pavement evaluation and design decisions. Backcalculation may use linear elastic or more advanced analytical approaches. Linear backcalculation typically includes three components: a forward analysis engine that computes surface deflections for assumed moduli, a convergence mechanism that minimizes discrepancies between measured and predicted deflections, and a search algorithm that identifies the optimal moduli. Reviewed forward analysis methods include finite element modeling, artificial neural networks, layered elastic theory, and deflection basin fitting. Convergence criteria such as root mean square error, absolute error, and hybrid tolerance-based equations were evaluated. Search algorithms such as the downhill simplex method and genetic algorithms were also examined. Although computationally efficient, linear approaches rely on simplifying assumptions, motivating the review of advanced techniques incorporating dynamic loading, viscoelasticity, and nonlinear material behavior.
Source: Transportation research procedia
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7204159281