The emergence of laser beam welding as a state-of-the-art, non-contact joining process in recent years is largely due to its ability to achieve high welding speeds and minimal thermal distortion compared to conventional welding techniques. This advantage stems from the laser's ability to concentrate energy in a small zone, allowing highly accurate and consistent process control. In addition, the technology's strong compatibility with automation has extended the application of the technology in various industries. Nevertheless, a significant challenge remains in the form of solidification cracking. These cracks start as tiny cracks in the weld bead during cooling and then extend to the surface as solidification progresses [1]. They occur mainly in the mushy zone - located immediately behind the weld pool - where liquid and solid phases coexist in a dendritic microstructure that can trap residual melt. If new liquid metal is unable to refill these regions during solidification, critical microstructural conditions develop, creating stresses that initiate cracking. This study approaches the problem from two perspectives. At the macroscale, a comprehensive heat source model has been developed to accurately approximate the melt pool during the welding process and has already been deployed in a high performance computing environment [2]. At the microscale, the dendritic morphology derived from phase field simulations is studied for its stress and strain response, incorporating both thermal and elastoplastic effects. By merging these scales, the approach allows deeper insights of failure initiation under critical conditions. [1] E. Folkhard, G. Rabensteiner, E. Pertender, H. Schabereiter and J. Tösch. Metallurgie der Schweißung nichtrostender Stähle. Springer-Verlag: Vienna, 1984. [2] T. Bevilacqua, A. Gumenyuk, N. Habibi, P. Hartwig, A. Klawonn, M. Lanser, M. Reth- meier, L. Scheunemann and J. Schröder. Large-scale Thermo-Mechanical Simulation of Laser Beam Welding Using High-Performance Computing: A Qualitative Reproduction of Experimental Results. arXiv: 2503.09345 [math. NA], 2025.