I n recent years, engine calibration became a very hard task because of the increasing complexity of systems and the severity of the depollution norms regarding Real Driving Emissions (RDE). In particular, optimal engine control during dynamic phases became crucial for reducing pollutant emissions. Beyond the classical engine calibration method based on steady state experiments, methods that integrate the dynamical response of the engine constitute therefore a promising approach. This work proposes a global approach of engine dynamical model-based calibration (DMBC) and optimization based on a dynamic Design of Experiments (DDoE). After a general description of the architecture of the calibration process, the paper focuses on the methodology for the design of DDoE. The proposed DDoE is based on an extraction of highly dynamical legal norm cycles (from the analysis of existing RDE cycles), together with variations of calibration parameters which are intern variables of the control strategy of the engine control unit (ECU). Variations of calibration parameters allow variations of control parameters which are outputs of the control strategy and relevant regarding emissions. The criterion chosen for the definition of the DDoE is the minimization of the discrepancy of control parameters of the engine. The catalyst warm-up phase has been chosen as an application case of the methodology developed in this paper. The main objective of this phase is to heat the catalyst as fast as possible so that it becomes efficient regarding reduction of pollutant emissions. The first results describe a part of the DDoE that will be next completed and performed in a test bench, in order to model engine emissions.