Statistical screening of individual and joint effect of several modelling factors on the dynamic finite element response of a railway bridge

More accurate dynamic analysis of bridges is preferred instead of using the dynamic amplification factors in design codes to estimate the dynamic load effects. Additionally, the potentials of statistical methods are often overlooked and seldom used among structural engineers. In this paper, factorial experimentation in simulating railway bridge dynamics was exemplified by screening the individual and joint effect of several modelling factors. The statistical theory proved to be relevant and meaningful, easy to implement but still very powerful. The concept of factorial experimentation was believed to provide the engineer with important indications of the effects of several possible updating parameters.

► Powerful design of experiments (DOE) is useful in bridge dynamic FE simulations.
► The study revealed synergy effects that usually are missed in parameter studies.
► The statistical algorithms are easy to implement and use in FE model updating.
► The identification of important parameters helps the engineer in her daily work.