<p>In recent decades, the food industry has undergone a revolution, driven by consumers’ growing interest in products that meet their nutritional needs and provide additional health benefits. Within this context, functional bakery products emerge as a promising category, catering to the growing demand for foods that combine natural sweeteners and convenience with beneficial health properties. The research endeavour aims to predict and optimize the quality characteristics of functional cookies enhanced with finger millet and jaggery as a natural sweetener, utilizing Artificial Neural Networks (ANN) and Simplex Lattice Mixture Design (MD). The primary goal of this analysis is to utilize advanced optimization frameworks to illuminate the intricate relationships between input variables, such as Whole Wheat Flour (WWF) and Finger Millet Flour (FMF), and the output metrics that represent various quality characteristics of cookies. The results highlight the effectiveness of Mixture Design in predicting a wide array of quality factors, with R² values ranging from 0.75 to 0.99. For certain variables, ANN exhibits slightly lower coefficients of determination, with R² values ranging from 0.20 to 0.99 and RMSE values reaching up to 39. Fourier Transform Infrared spectroscopy analysis identified broad and characteristic bands associated with –OH stretching (3300–2500&#xa0;cm − 1), suggesting the potential presence of moisture, carbohydrates, and polyphenols. Scanning electron microscopy images revealed that the inclusion of FMF and jaggery altered the structural composition of the cookies. Overall, the addition of FMF and jaggery to cookies improved their nutritional profile. These findings are significant in addressing the growing demand for functional food products.</p>