Mineral scaling is a common issue in oil recovery, causing severe formation damage and reduced permeability. This primarily results from chemical incompatibility between the injected and formation brines, which induces the precipitation of salts such as calcium carbonate, calcium sulfate, and barium sulfate. Accurate scale prediction is essential for managing operational risks and minimizing associated costs. in, this study, a symbolic regression technique based on evolutionary algorithms was employed to an explicit mathematical model linking permeability damage to key flow parameters. A scaling risk map was also produced to distinguish between scaling and non-scaling regions. A dataset of 431 literature-sourced permeability damage measurements was analyzed using variables such as temperature, pressure differential, Volume of injected water, initial permeability, fluid injection rate, and concentrations of major ionic constituents. For model validation, the dataset was partitioned into training (80%) and testing (20%) subsets. The model achieved strong predictive accuracy with an R^2of 0.99, showing excellent agreement with experimental observations. Compared to classical thermodynamics-based models, which often neglect kinetic factors, the proposed approach offers improved prediction and control strategies for scale formation during water flooding. This makes it especially valuable for operational planning and real-time decision-making. In summary, the model developed in this study presents significant practical benefits for researchers and engineers in both academic and industrial settings, enhancing the understanding and mitigation of mineral scaling under mixed salt conditions.