Deep drawing and subsequent wall ironing (DWI) in steel battery can manufacturing induce markedly different stress states, which in turn drive distinct microstructural evolutions. During deep drawing, the sheet experiences predominantly in-plane tensile stresses along the radial and hoop directions, while wall ironing imposes a three-dimensional stress condition characterized by high compressive stresses through the thickness (normal contact pressure) and significant out-of-plane shear at tool-workpiece interfaces due to friction. This transition from a plane stress condition to a fully three-dimensional state demands 3D constitutive models that capture not only the complex load paths inherent in DWI processes but also the complete multi-axial stress interactions. In this work, the 3D Vegter yield locus [1, 2] with isotropic hardening is employed to account for initial anisotropy and the expansion of the yield surface under plastic deformation. Although this phenomenological model accurately reflects steel’s initial anisotropic yield behaviour, it remains static with respect to evolving anisotropy caused by texture changes at large strains. To investigate the implications of this limitation, crystal plasticity finite element (CPFE) simulations are conducted to track microstructural texture evolution throughout the entire DWI process, complemented by electron backscatter diffraction (EBSD) measurements in formed parts. Comparison of the 3D Vegter model predictions to CPFE and EBSD data reveals that a nonevolving yield function falls short in capturing progressive anisotropy changes. To address this limitation, we propose an uncoupled methodology that achieves computational efficiency while retaining the requisite precision by updating the yield locus prior to each process step, thereby incorporating the effects of texture evolution throughout the multistage forming process.