Optimization of a Functional Beverage Formulation Using a Constrained Mixture Design with Multiple Responses
Maurice Wanyonyi *
Department of Mathematics and Statistics, University of Embu, Kenya.
John Wafula Kiluyi
Department of Mathematics and Statistics, University of Embu, Kenya.
Ayubu Anapapa Okango
Department of Mathematics and Actuarial Science, Muranga University, Kenya.
*Author to whom correspondence should be addressed.
Abstract
Functional beverages refer to a relatively new and expanding market segment appearing that goes beyond simple nutrition but contains some beneficial ingredients. This study employs a constrained mixture design to optimize a functional beverage's formulation by systematically analyzing the interactions between three key ingredients: fruit juice blend, low fat probiotic yogurt and a functional ingredient. Based on D-optimal experimental design the research mimics response across various factors such as taste, texture, perceived health benefits. To assess the impact of the ingredient proportions, a second-degree mixture model was used, while the levels of the ingredient constraints were set at 40-70% for fruit juice blend, 20-40% for probiotic yogurt, and 10-30% for the functional ingredients. Further quantitative analysis in this study confirmed that fruit juice blend was statistically significant in all the response variables as the most influential factor controlling the treatment while probiotic yogurt also had strong but slightly lesser influence. Desirability functions used in the optimization process yielded the maximal proportion of the fruit juice blend of 0.7, probiotic yogurt 0.5, and functional ingredient 0.5. This work offers scientific approach to creating functional beverages with enhanced sensory qualities to match enhanced health properties, solving a current gap in techniques for extended food science in beverage development.
Keywords: D-optimality experimental design, second-degree mixture model, optimization process, constrained mixture design