Selecting optimal mixtures of natural sweeteners for carbonated soft drinks through multi-objective decision modeling and sensory validation

Abstract

The objective of this study was to develop a methodology to optimize mixtures of natural, non- caloric sweeteners—with the highest sweetness and the lowest bitterness—for carbonated soft drinks. To this end, and with the aid of a trained sensory panel, we first determined the most suitable mixtures of tagatose, sucrose, and stevia in a soft drink matrix, using a three-component simplex lattice mixture design. Then, we developed a multi-objective thermodynamically-based decision model to this purpose. Results indicate that both, sucrose and tagatose, were able to reduce stevia ́s bitterness. However, an increase of bitterness intensity was found above 0.23 g/L of stevia (sucrose equivalency or SE >5). Both, sensory analysis and multi-objective decision modeling identified similar optimal mixtures, corresponding to 23–39 g/L sucrose, 0.19–0.34 g/L stevia, and 34–42 g/L tagatose, depending on the desired sweetness/bitterness balance. Within this constrained area, a reduction of almost 60% of sucrose can be achieved in both approaches, keeping bitterness intensity low.