Natural and man-made materials with natural fibres are complex materials due to three major issues: 1) anisotropy, 2) heterogeneity, 3) strong environmental dependency. Both, variation of moisture/temperature and strain rate dependency of these materials, are creating complex conditions for numerical simulations and are leading to an under-/overestimation of the mechanical behavior of these materials in structural applications. Collapse of timber structures due to failure of connections, or failure of sport or musical instruments due to frequent loading/unloading cycles in addition to the environmental changes are strong indicators for the importance of accurate mechanical analyses and numerical simulation/optimization. Existing numerical models do not cover sufficiently the relevant phenomena or are too complex for efficient simulation/optimization to support design decisions; a better compromise between efficiency and accuracy is required.

Due to the importance of these parameters, the first aim of this research is to develop a set of comprehensive material models to describe the mechanical behavior of biomaterials, combining different possible damage initiating factors. The second aim is to develop a multi-fidelity optimization scheme based on this set of numerical models with different degree of accuracy and therefore efficiency (i.e. with different complexity concerning the mechanical phenomena considered) enabling structural optimization. A smart scheme using low-/high-fidelity models simultaneously in the optimization will make multi-query assessments affordable.