Thermo-kinetic fingerprinting of magnesium alloys through the Sudesh–Ashwani Statistical Model (SASM): application to AZX211

Magnesium alloys continue to attract significant interest as next-generation lightweight structural materials due to their low density, high specific strength, and favourable processing characteristics. Among these, the Mg–1.0Al–0.3Ca–4.0Zn (AZX211) alloy represents a promising composition that combines solid-solution strengthening from Al–Zn with Ca-induced grain refinement and improved creep resistance. Accurate modelling of the thermodynamic and kinetic behaviour of such alloys is essential for optimising casting, solidification, and heat-treatment routes. However, conventional CALPHAD-based approaches remain constrained by database incompleteness—particularly for Ca- and rare-earth-containing systems—and by their equilibrium-centric formulation, which limits applicability under industrial non-equilibrium conditions. This work introduces the Sudesh–Ashwani Statistical Model (SASM), a compact, CALPHAD-free framework for thermo-kinetic characterisation of magnesium alloys. SASM employs polynomial–exponential expressions for enthalpy, Gibbs free energy, entropy, and density; a non-isothermal Avrami formulation for solidification kinetics; and logistic functions for phase-fraction evolution. These components collectively define a thermo-kinetic fingerprint that is interpretable, transferable, and computationally efficient. Benchmarking SASM against JMatPro simulations for AZX211 demonstrates strong agreement across all observables. The model captures enthalpy reversal between 400–800 °C, the crossover of the effective interaction parameter, Gibbs stabilisation with Al additions, entropy rise, density discontinuity near melting, sigmoidal solidification kinetics, and the evolution of α-Mg, liquid, MgZn, and Al₂RE phases. Deviations remain within 5–8%, underscoring the model’s fidelity despite its database-independent construction. SASM offers a transparent and lightweight alternative to conventional thermodynamic modelling, suitable for rapid alloy assessment, high-throughput screening, and integration with machine-learning frameworks. Its generality enables extension to Mg–rare-earth, Al-based, and Ti-based systems, providing a pathway toward interpretable, scalable, and industrially relevant alloy design.