The mechanical characteristics of skin are extremely complex and have not been satisfactorily simulated by conventional engineering models. The ability to predict human skin behaviour and to evaluate changes in the mechanical properties of the tissue would inform engineering design and would prove valuable in a diversity of disciplines, for example the pharmaceutical and cosmetic industries, which currently rely upon experiments performed in animal models. The aim of this study was to develop a predictive anisotropic, hyperelastic constitutive model of human skin and to validate this model using laboratory data. As a corollary, the mechanical characteristics of human and murine skin have been compared. A novel experimental design, using tensile tests on circular skin specimens, and an optimisation procedure were adopted for laboratory experiments to identify the material parameters of the tissue. Uniaxial tensile tests were performed along three load axes on excised murine and human skin samples, using a single set of material parameters for each skin sample. A finite element model was developed using the transversely isotropic, hyperelastic constitutive model of Weiss et al. (1996) and was embedded within a Veronda–Westmann isotropic material matrix, using three fibre families to create anisotropic behaviour. The model was able to represent the nonlinear, anisotropic behaviour of the skin well. Additionally, examination of the optimal material coefficients and the experimental data permitted quantification of the mechanical differences between human and murine skin. Differences between the skin types, most notably the extension of the skin at low load, have highlighted some of the limitations of murine skin as a biomechanical model of the human tissue.

The development of accurate, predictive computational models of human tissue, such as skin, to reduce, refine or replace animal models and to inform developments in the medical, engineering and cosmetic fields, is a significant challenge but is highly desirable. Concurrent advances in computer technology and our understanding of human physiology must be utilised to produce more accurate and accessible predictive models, such as the finite element model described in this study.