Since the application of simulation, especially the finite element method, has become more and more significant in the printed circuit board (PCB) industry, realistic material models have become more important. This is necessary to create application related results. For example, virtual torsion tests, drop tests and thermal cycle/shock tests are applied to predict the failures of a PCB. Thereby the nine temperature-dependent orthotropic engineering constants of the woven glass fibre reinforced laminates have a major impact on the simulation outcome. The present master thesis focuses on the optimization of conventional mechanical characterization methods and micromechanical methods for the determination of the engineering constants. For the experimental part, three different glass fibre reinforced epoxy dielectric materials and their resin matrix were chosen. The laminate and resin specimens were prepared by an automated routing procedure. All of the investigated materials were systematically mechanically characterized by monotonic tensile test at elevated temperatures and dynamic mechanical analysis. In addition, the temperature dependent Poisson’s ratio was investigated more closely in this study. A focus was also on the clarification of the different modulus outcome with respect to the mentioned methods and the specimen preparation procedure. The volumetric resin content of the laminates was determined by hydrostatic weighing and thermogravimetry. For the micromechanical part, a unit-cell model was implemented into a finite element environment to predict the engineering constants with the measured data as inputs. Abaqus®/Standard software was the applied numerical tool. Different yarn fibre aspect ratios and periodic boundary conditions were applied to the model. The results from the experimental and micromechanical part were compared to evaluate the potentials of micromechanical methods. It can be shown that the monotonic tensile test and dynamic mechanical analysis are creating reliable comparable data when preparing the specimens in a proper way. In addition, the presented methodology to determine the resin content by thermogravimetry can produce quite useful results, from an experimental and numerical point of view. In contrast, the characterisation of the in-plane Poisson’s ratio at elevated temperatures of thin laminates can be seen as a challenging procedure. However, the obtained numerical results were in a good agreement with the experimental findings.