2.2.1. Experimental Design

The formulation of ice cream samples was determined using the Design Expert Software (Version 7; Stat-Easy Co., Minneapolis, MN, USA). Response surface methodology (RSM) was carried out for the study of three factors, namely fat concentration (2.5–12.5 g/100 g) ( *X*1), CSOB concentration (1–3 g/100 g) ( *X*2), and gum concentration (0.2–04 g/100 g) ( *X*3), using an unblocked full factorial central composite design (CCD). Seventeen different experimental points were obtained by using Design Expert Software (Version 7; Stat-Easy Co., Minneapolis, MN, USA) to determine the optimum amount of fat, CSOB, and gum content. For the estimation of the error, the design consists of three of the factorial points. A quadratic model was fitted to the experimental data for each response. Model applicability was evaluated based on the *R*2, *R*2-adj, lack of fit, F, and *p*-values obtained from ANOVA. Response surface plots showing the effect of model parameters on the *K* value were built by Design Expert Software. The optimization was performed based on the highest desirability value. The formulation with the lowest fat content with a desirability value of 1 was determined as the optimum formulation. Three central points were used. Analysis of all points was performed in triplicate, and the results were reported as mean values and standard deviations. The formulation of the samples has been determined in accordance with the ice cream standard. Accordingly, 2.5% fat ratio representing low-fat ice cream (0.5 ≤ milk fat < 3), 7.5% fat ratio representing semi-fat ice cream (3 ≤ milk fat < 8), and 12.5% fat ratio representing fat ice cream (8 ≤ milk fat ≤ 13) were used. Based on these fat ratios, the amount of milk and milk cream to be used has been calculated. The percentage of fat, XG, and CSOB vary in formulations, and the amounts of sugar (12%) and EYP (3%) remain constant. Seventeen formulations obtained in this way are shown in Table 1.
