In low alloy steels, an accurate modelling of phase transformations is mandatory in order to have a good prediction of phase proportions and induced residual stresses in the heat affected zone. These transformations can be efficiently described by the JMAK equation and easily added to FEM softwares [1] but some difficulties remain in the calibration of phase transformation parameters. Indeed, most of the work in the literature assumes that the transformations parameters during cooling are independent of the thermal cycle. This assumption leads to inaccurate phase prediction in the case of manufacturing processes that can reach high temperature and high cooling rates. Some recent models taking into account the chemical composition have been developed [2], [3] but they are too generic to get accurate results on specific materials. This work shows an application to determine transformation parameters for welding conditions and how these additional values can be used for a better modelling in the heat affected zone. A grain size modelling and dependency of phase transformation has been developed and its effect on mechanical results are presented for a 16MND5 steel. REFERENCES [1] J. B. Leblond et J. Devaux, « A new kinetic model for anisothermal metallurgical transformations in steels including effect of austenite grain size », Acta Metallurgica, vol. 32, no 1, p. 137‑146, janv. 1984, doi: 10.1016/0001-6160(84)90211-6. [2] M. V. Li, D. V. Niebuhr, L. L. Meekisho, et D. G. Atteridge, « A computational model for the prediction of steel hardenability », Metallurgical and Materials Transactions B, vol. 29, no 3, p. 661‑672, juin 1998, doi: 10.1007/s11663-998-0101-3. [3] X. Geng et al., « A hybrid machine learning model for predicting continuous cooling transformation diagrams in welding heat-affected zone of low alloy steels », Journal of Materials Science & Technology, vol. 107, p. 207‑215, avr. 2022, doi: 10.1016/j.jmst.2021.07.038.