Truss Systems: From Structural Idealization to Monitoring-Based Assessment

Abstract

Truss systems are widely used in construction as lightweight and efficient load-bearing elements for roofs, bridges,industrial facilities, towers, and large-span structures. Their structural performance is influenced not only by the axial forcesacting on individual members, but also by system-level effects related to geometric imperfections, joint behavior, nonlinearresponse, environmental actions, and variations in other design parameters. This paper reviews engineering approaches to themodeling and assessment of truss systems, with particular emphasis on their significance for the reliability assessment of civilengineering structures. The review covers classical idealizations, normative modeling requirements, experimental and full-scaleevidence, finite element modeling strategies, stochastic analysis, system reliability, surrogate-based methods, and monitoring-based updating. The reviewed studies show that simplified deterministic models are often insufficient when structural responseis influenced by instability, load redistribution, or uncertainty-sensitive behavior. Stochastic and reliability-based methodsprovide a consistent framework for quantifying these effects, although their direct integration with detailed numerical modelsmay become computationally demanding. The review identifies three principal research gaps: efficient representation ofnonlinear, imperfection-sensitive, and system-level truss behavior; integration of reliability analysis with surrogate, data-driven,and physics-informed approaches; and systematic incorporation of structural health monitoring data into numerical andprobabilistic models for service-life reliability assessment.