Fourier Transform Infrared (FTIR) spectroscopy combined with Partial Least Squares (PLS) regression has emerged as a powerful tool for detecting adulteration in edible oils. However, the high dimensionality and spectral redundancy of FTIR data often hinder model accuracy and generalizability. This study evaluates and compares the effectiveness of three metaheuristic optimization algorithms—Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and a novel hybrid GA-PSO approach—for enhancing FTIR-PLS models in detecting sesame oil adulteration with canola, corn, and sunflower oils. Results demonstrated that the hybrid GA-PSO algorithm significantly outperformed standalone GA and PSO, achieving R²p=0.985 and RMSEP=4.92% for canola-adulterated oil, R²p=0.987 and RMSEP=5.62% for corn-adulterated oil, and R²p=0.991 and RMSEP=4.51% for sunflower-adulterated oil. In contrast, baseline (non-optimized) PLSR models exhibited higher prediction errors (RMSEP > 6%). Comparative analysis confirmed that GA-PSO synergizes GA’s global search capability with PSO’s rapid convergence, effectively minimizing overfitting and enhancing model robustness. This study establishes hybrid metaheuristic-optimized FTIR-PLSR as a superior framework for food authentication, offering high accuracy and reliability for industrial quality control.

Keywords