Résumé : A novel model for the pyrolysis of the Phenolic Impregnated Carbon Ablator (PICA) at high heating rate is developed and calibrated based on high fidelity thermal decomposition experiments. The calibration relies on accurate quantification of pyrolysis gases obtained from mass spectroscopy analysis during thermal decomposition at fast heating rates simulating flight conditions. Model calibration is achieved by coupling the Porous material Analysis Toolbox based on OpenFOAM (PATO) with an optimization software (Dakota). A multi-objective genetic algorithm is used to fit the experimental data by optimizing the model parameters for an element and a species-based formulation. The new model captures both the material mass loss and the gaseous species produced during pyrolysis.