The maturation of full-field measurement methods has enabled experimental measurement of the specimen surface displacement and strain fields, paving the way for more complex and rich experiences. In the domain of material behavior identification, the employment of full-field measurement methods represents a significant development. This advancement enables a transition from conventional tests based on homogeneous stress/strain states to more intricate tests featuring heterogeneous stress/strain states. In this context, robust and efficient inverse methods have been developed for the identification of constitutive parameters. To ensure accurate identification, particular consideration must be given to the test design. Consequently, there has been a recent shift in focus in research towards the design of complex specimen for full-field measurements and inverse identification. An array of approaches has been developed, ranging from intuitive design to design by full identification simulation [1]. The approach considered in this study is the design by identification quality. A variety of criteria have been formulated with the objective of enhancing the test's sensitivity to material parameters [1, 2]. However, the majority of research activities have been concentrated on applications for elastic or elasoplastic behaviors in quasi-static conditions. Advancements in high-speed camera technology have open this approach to dynamics. The present study proposes a methodology to select specimen designs based on finite elements displacement fields sensitivities to facilitate the identification of material parameters in a dynamic test. The selection of the position of two holes in a specimen with a elasto-viscoplastic model is conducted as an application. To explore the design space while limiting calculation costs, a parametric optimization using a surrogate model is put in place. This approach allows the selection of a specimen geometry that maximizes robustness of the identification procedure. [1] F. Pierron, M. Grédiac., Towards Material Testing 2.0. A review of test design for identification of constitutive parameters from full-field measurements, Strain, 51(1), 2021. [2] M. Bertin, F. Hild, S. Roux, Optimization of a Cruciform Specimen Geometry for the Identification of Constitutive Parameters Based Upon Full-Field Measurements, Strain, 52, pp.307 - 323, 2016.