The increasing demand for reduced emissions in the automotive sector has driven research into alternative fuels, including Diesel, Biodiesel, and ethanol blends. This study aims to optimize mixing rules to predict the physicochemical properties of ternary fuel blends, essential for improving engine performance and minimizing emissions. Seven established mixing rules—Kay’s Equation, Semilogarithmic Equation, Grunberg-Nissan Equation, Modified Lederer Equation, Hu-Burns Equation, Power Law, and Polynomial Equation—were evaluated to estimate key properties such as kinematic viscosity, cetane number, cetane index, flash point, pour point, and cloud point. A comprehensive database, sourced from previous literature, included pure fuel properties and blend data for 33 to 101 cases. MATLAB was used to implement nonlinear optimization, adjusting coefficients to minimize error metrics like Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Standard Deviation (SD). The physical consistency of the correlations was verified by ensuring that estimated properties followed expected trends. The results identified the best-performing equation for each property, with the Grunberg-Nissan equation providing reliable estimates for viscosity and cetane number, and the Hu-Burns equation excelling in cetane index and cloud point. Additionally, the polynomial equation demonstrated accuracy for flash point. The optimized correlations were validated with independent experimental data, confirming their robustness and suitability for automotive fuel applications. This approach simplifies the selection of fuel compositions, contributing to cleaner fuel formulations and supporting the shift towards sustainable energy solutions.