Abstract: In this study a material model is developed to predict the solidification microstructure of an additive-manufactured, fully dense magnesium (Mg) alloy using uniform droplet spraying (UDS). Specifically, the crystallite size distribution is simulated by a solidification model, consisting of a nucleation/fragmentation and a constrained growth description, calibrated via microstructural data from a single droplet splat. This is enabled by a semi-analytical thermal modeling framework, based on the superposition of moving Green’s and Rosenthal functions for the temperature field generated by a Gaussian source distribution. The model is implemented for layered ellipsoidal deposit sections on planar substrates by multi-pass spraying, and its predictions are validated against measured crystal sizes by image analysis of experimental micrographs of a Mg ₉₇ZnY ₂ alloy, to an error margin of ± 15%. The computationally efficient simulation provides insights to the deposit microstructure, and is intended as a process observer in a closed-loop, adaptive control scheme based on infrared temperature measurements. Graphic abstract: [Figure not available: see fulltext.]