<p>A hybrid multi-criteria decision-making (MCDM) approach was proposed for the global optimization of the resistance spot welding process for aluminum/steel dissimilar metals. The methodology integrated orthogonal experimental design, gray relational analysis (GRA), and principal component analysis (PCA) to establish a comprehensive quantitative evaluation approach for weld quality. Through an orthogonal array, 16 sets of welding parameters were designed, and eight corresponding quality indicators, including aluminum/steel nugget diameters, indentations, steel bulge height, peak load, failure energy, and welding energy, were systematically collected. The GRA–PCA hybrid model was employed to fuse these multi-dimensional responses into a single comprehensive performance indicator (CPI), effectively quantifying the relative contribution of each quality indicator to the overall joint performance. The analysis identified aluminum nugget diameter, peak load, and failure energy as the dominant factors governing the comprehensive weld performance. Furthermore, a relationship model between key welding parameters and the CPI was constructed using scatter data interpolation. A genetic algorithm was and then, applied as a global optimization tool to identify the parameter set that maximizes the CPI. The optimal combination was determined as a welding time of 0.30&#xa0;s, welding current of 11.0&#xa0;kA, and electrode force of 3.0&#xa0;kN. Verification experiments confirmed that this parameter set yields the optimal CPI value, validating the effectiveness of the proposed hybrid MCDM approach.</p>