Desirability-based optimization of bakery products containing pea, hemp and insect flours using mixture design methodology

Abstract

Simplex lattice design was used to design 15 sponge cakes formulations combining pea (PP), hemp (HP) and insect (IP) flours representing 15\% of dough composition. Moisture, protein content, baking loss, specific vol-ume, texture and cost of the 15 samples plus the control (0\% added protein) were analysed. Results showed that the effect of PP, HP and IP on cake properties could be modelled with linear regressions (96.80\% < R2 < 99.96\%). Ternary diagrams showed the effect of the combination of the three proteins in each response. The desirability function was used to obtain a multi-response optimization of the samples with maximum protein, maximum specific volume and minimum incremental cost. Sensory results of the 5 optimised samples showed that by combining 3.75\% pea, 3.75\% hemp and 7.5\% insect it was possible to obtain a dairy-and egg-free sponge cake without significant differences from the control with animal-derived proteins.