Premium
Tackling uncertainty in structural lifetime evaluations
Author(s) -
Sanio David,
Ahrens Mark Alexander,
Mark Peter
Publication year - 2018
Publication title -
beton‐ und stahlbetonbau
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.486
H-Index - 25
eISSN - 1437-1006
pISSN - 0005-9900
DOI - 10.1002/best.201800036
Subject(s) - residual , bridge (graph theory) , computer science , structural health monitoring , constraint (computer aided design) , operations research , environmental science , reliability engineering , transport engineering , engineering , structural engineering , algorithm , mechanical engineering , medicine
At present, investigation, rehabilitation and if necessary deconstruction and replacement of civil infrastructure are key challenges in structural engineering. Internationally, the number of bridges that show a critical state of preservation is growing while traffic loads, frequencies and freight amounts are still rising. Against the background of economic constraints authorities are challenged to decide on the right measures and order to take. Sound decisions rest on reliable data concerning structural health states and safely estimated residual lifetimes. Among experts it is widely accepted that estimated residual lifetimes must involve uncertainties from various domains like modelling assumptions, material data assessment, realistic traffic loads etc. Uncertainties reduce accuracies and thus might reduce computed residual lifetimes. To overcome this burden, model updates, material testing, traffic census or at best monitoring of sensitive structural elements can be applied. The contribution presents up‐to‐date methods and results of geometry assessments from laser scans, multi‐copter overflights or mobile mapping, material testing, FE‐model updates from test loadings on‐site, data and video based traffic census, transient temperature constraint evaluations from climate data as well as strain monitoring of tendons at a large scale concrete bridge. The measures are stochastically evaluated to quantify potential gains even if correlation of model parameters is involved.