Stochastic Modelling in Timber Engineering

Research output: ThesisHabilitation


Timber is a natural, hierarchically constructed, and in modelling complex building material, featuring high varying properties, which are in addition subjected to a number of uncertainties caused by processing trees to timber and further into timber system products. Its physical properties are a function
of loading mode, volume, environmental conditions and loading history, i.e. long-term effects. Timber as building material has to fulfil a number of requirements, for example in respect to capacity, serviceability, reliability and durability. Because of these requirements and its high varying and uncertain properties, the probabilistic representation and modelling of structural timber components, timber (system) products and timber structures is of utmost importance.
Within this contribution the probabilistic representation and modelling of structural timber and glued laminated timber products is exemplarily demonstrated in four thematic fields: (i) the spatial distribution
of timber properties by means of stochastic hierarchical models used for modelling system effects and in evaluation procedures, (ii) shear properties of structural timber and glued laminated timber products:
property values, test configurations and evaluation, influencing parameters and design approaches, (iii) probabilistic modelling of systems in general and of glued laminated timber products in particular, and (iv) probabilistic models in conjunction with fasteners and joints, with focus on axially-loaded selftapping
The aim of this contribution is to support decision making processes of engineers, authors of standards and producers and to provide a sound basis for harmonised and consistent regulations in timber engineering in regard to product properties, testing, evaluation and design.
Original languageEnglish
Awarding Institution
  • Graz University of Technology (90000)
Publication statusPublished - 2018

Fields of Expertise

  • Sustainable Systems

Cite this