1 Introduction

Food loss, as defined by the Food and Agriculture Organization [1], refers to a decrease in both the quantity and quality of food derived from agricultural and fishery products intended for human consumption but ultimately not consumed. Globally, approximately 13% of the food produced is lost between harvest and retail, with an estimated 17% of the total global food production wasted in households [2]. In today’s world, where an alarming 333 million people across 78 countries experienced acute levels of food insecurity in 2023 [3], food loss or waste is no longer acceptable. Furthermore, it contributes significantly to greenhouse gas emissions, exacerbating environmental degradation and worsening climate change conditions [4]. These pressing issues have prompted policymakers, researchers, and companies to focus on reducing food wastage along the food supply chain through valorization—the process of converting waste into a valuable product whose value exceeds the cost of its transformation. In the Philippines, a bill known as the Zero Food Waste Act of 2022 has been proposed with the aim of implementing a system that promotes food waste reduction through redistribution, recycling, and valorization [5].

Coconut water is widely consumed worldwide due to its nutritional and health benefits. Water from young coconut is preferred to water from mature coconut because of its sweeter taste [6]. In the Philippines, mature coconut is generally used for its pulp to produce coconut milk/oil and desiccated coconut [6, 7], but its water content is discarded as waste. As a result, large quantities of this nutritious water are left unused and thrown in the coconut industry (approximately 2.4 billion liters per year) [7], despite numerous studies reporting that mature coconut water can be a great rehydration drink [8] due to its higher mineral content than its young counterpart [9,10,11]. Few researchers have studied the use of mature coconut water as a beverage. A study by Tancueco et al. [12] revealed that different varieties of coconut affect its total free sugars and total phenolic content. More recently, Aba et al. [13] characterized the physicochemical, compositional, and phytochemical properties of mature coconut water for beverage development and revealed the potential of mature coconut water for beverage development. Chauhan et al. [6] attempted to develop mature coconut water with lemon using mixed fruit blending, which is a method that improves the sensory acceptance of underutilized or discarded juices [14].

In the Philippines, a good alternative for lemon that contains similar nutrients, phytochemicals, and natural acidulants is calamansi [15, 16]. In a study conducted by Gabriel, Fernandez, and Bayaga [17], it was reported that calamansi is the second most preferred citrus fruit after ‘dalanghita’ and is consumed because of its aroma and distinctive flavor. Further, citric acid from calamansi prevents browning in juices which is helpful in stabilizing color of the beverage [14]. Aside from nutrients and minerals, calamansi contain biologically active compounds such as antioxidants and soluble and insoluble dietary fibers which reduce the risks of cancer, arthritis, obesity, and coronary heart diseases [18]. The Southeast Asian Regional Center for Graduate Study and Research in Agriculture [19] also reported that calamansi is already being recognized in the international market, and in terms of production, calamansi ranks fourth to banana, mango, and pineapple.

To address the undesirable color/turbidity and sweetness of mature coconut water, turmeric and stevia were used in this study. Turmeric is a stable yellowish-orange pigment called curcumin, and it is often used as a food colorant [20]. Curcumin is also responsible for the antioxidant and anti-inflammatory properties of turmeric [21,22,23]. According to Prathapan et al. [24], curcumin is thermally stable when heated to 100 °C for 30 min which make it suitable for beverage development as it imparts a yellowish-orange color. Stevia, on the other hand, is a natural nonnutritive sweetener. It contains glycosides, steviosides, and different types of rebaudiosides that are commonly extracted from its leaves. In its pure form, stevia is 100 × to 300 × sweeter than sucrose, making it a sucrose replacer, solubilizing agent, and a functional food ingredient [25]. Since most fruit drinks contain added sugars, an increased risk of chronic diseases such as overweight, obesity, and diabetes is recorded when consumed regularly [26, 27]. The Joint FAO/WHO Expert Committee on Food Additives [28] established that there are no safety issues in the use of stevia in food. Like any other sugar substitutes, bitter and metallic taste are the main disadvantages of using stevia However, these disadvantages can be masked when combined with a strong flavoring such as lime and other citrus fruits [29]. In 2018, the Philippines implemented the Tax Reform for Acceleration and Inclusion (TRAIN) law [30], which exempts beverages from sugar tax collection from stevia.

Considering the aforementioned raw materials, the mixture design of experiment is used for optimization. In this design, the raw materials or components of a mixture, and the responses to be observed are a function of the proportions of each ingredient [31]. A three-component design is represented by a triangle (Fig. 1) in which the vertices represent pure components (X1, X2, and X3); the edges represent binary components (X1 + X2, X1 + X3, and X2 + X3) with one component absent, and the point inside the triangle represents the combination of each component (X1 + X2 + X3) at varying levels [32, 33]. The mixture design generates a mathematical model that may take a linear (Eq. 1), quadratic (Eq. 2), special cubic (Eq. 3), and a full cubic (Eq. 4) form, defining the relationship between the coefficients (β), factors (X1, X2, and X3) and the responses (y) [32].

Fig. 1
figure 1

Simple lattice design plot (adopted from Squeo et al. [32]. X1, X2, and X3 are the individual components, purple markers denote binary components where one component is absent, blue markers denote combination of all components at various levels, and gray marker denotes combination of all components at equal parts

Full size image
$$y={\beta }_{1}{x}_{1}+{\beta }_{2}{x}_{2}+{\beta }_{3}{x}_{3}$$
(1)
$$y={\beta }_{1}{x}_{1}+{\beta }_{2}{x}_{2}+{\beta }_{3}{x}_{3}+{\beta }_{12}{x}_{1}{x}_{2}+{\beta }_{13}{x}_{1}{x}_{3}+{\beta }_{23}{x}_{2}{x}_{3}$$
(2)
$$y={\beta }_{1}{x}_{1}+{\beta }_{2}{x}_{2}+{\beta }_{3}{x}_{3}+{\beta }_{12}{x}_{1}{x}_{2}+{\beta }_{13}{x}_{1}{x}_{3}+{\beta }_{23}{x}_{2}{x}_{3}+{\beta }_{123}{x}_{1}{x}_{2}{x}_{3}$$
(3)
$$y = \beta _{1} x_{1} + \beta _{2} x_{2} + \beta _{3} x_{3} + \beta _{{12}} x_{1} x_{2} + \beta _{{13}} x_{1} x_{3} + \beta _{{23}} x_{2} x_{3} + \beta _{{123}} x_{1} x_{2} x_{3} + \beta _{{12}} x_{1} x_{2} \left( {x_{1} + x_{2} } \right) + \beta _{{13}} x_{1} x_{3} \left( {x_{1} + x_{3} } \right) + \beta _{{23}} x_{2} x_{3} \left( {x_{2} + x_{3} } \right)$$
(4)

In mixture experiments, all factors were added to a constant, which imposes an additional constraint on the design. As a result, traditional models for quadratic (x2) and cubic (x3) terms will no longer fit. The equations presented above estimate one less term by removing the intercept model. The quadratic model estimates the linear and all two-interaction terms, while the cubic model estimates the linear, the two- and three-interaction terms [32, 34].

Considering the goal of the Philippines to promote food waste reduction, the large volume of waste generated by the coconut industry, and the unacceptable sensory characteristics of mature coconut water, the aim of this study was to develop an acceptable flavored mature coconut water. Specifically, to assess the effects of the formulation on the physicochemical and sensory properties of mature coconut water, and to determine the levels of these parameters using desirability-based optimization to produce acceptable flavored mature coconut water with calamansi, stevia, and turmeric.

2 Materials and methods

2.1 Sample collection, sanitation, and preparation

Husked mature coconuts (var. Laguna Tall; 11–12 months old) were collected from farmers in Quezon Province, Philippines. Nuts with intact shells were further cleaned to remove excess husk by brushing with steel wool, washing with potable water, and soaking in a 100 ppm chlorine solution for 10 min. Afterwards, the coconuts were allowed to air dry for another 10 min prior to cracking. Sanitized coconuts were manually cracked, and the water was pooled in a container and passed through a No. 120 aluminum sieve. Once filtered, the water was packed in 1 L of low-density polyethylene (LDPE) plastic and stored at − 80 °C until use. All materials used to extract mature coconut water were sanitized by soaking in a 10 ppm chlorine solution for 10 min.

The calamansi juice was sourced from a manufacturing company in Quezon City, Philippines. The juice was stored in a freezer until further use. Thawing of mature coconut water and calamansi juice was performed 24 h prior to scheduled production and at 4 °C. Powdered turmeric (Delfa’s Food Products) was obtained from a manufacturing plant in Cavite, Philippines, while powdered stevia (Glorious Industrial Development Corporation) was obtained from Caloocan City, Philippines.

2.2 Beverage preparation and processing

Based on the combinations generated, mature coconut water, calamansi juice, and stevia powder were weighed and combined. This was followed by the addition of 0.20 g/100 g of turmeric powder. A fixed amount of turmeric powder was determined to be acceptable in terms of color and taste during the preliminary tests. The beverage was heated to 85 °C for 5 min prior to hot filling in 250 ml glass bottles to ensure vacuum formation. This was followed by pasteurization at 93.3 °C for 5 min [35]. The pasteurized samples were stored at 4–8 °C until use.

2.3 Experimental design

Response surface methodology was used to develop mature coconut water with calamansi and turmeric beverages. A D-optimal mixture design was used to determine the combinations of mature coconut water, calamansi juice, and stevia powder using StatEase 360 Design Expert (Minneapolis, Minnesota, USA) v. 22. The ranges (Table 1) of each raw material were based on preliminary studies conducted by the authors (data not presented).

Table 1 Experimental ranges of independent variables to be used in D-optimal mixture design
Full size table

2.4 Numerical optimization

The numerical optimization of variables was performed using “desirability”, a function that utilizes numerical values to find a point that combines both the desires and priorities for each formulation and response variable [36, 37]. The criteria, including goals (“minimize”, “maximize”, “target”, “in range”, and “equal to”), limits, weights, and importance were set. Table 2 summarizes these criteria for determining the optimum combination and response of the developed flavored mature coconut water.

Table 2 Specifications of criteria for the optimization of the components and responses of the flavored mature coconut water
Full size table

Additionally, the model's fitness was assessed through the lack-of-fit test (p > 0.05), indicating how well the model fit the data [38]. Beverage characteristics with a significant (p < 0.05) model, an insignificant lack of fit (p > 0.05), a high regression coefficient (R2), and adequate precision were considered for numerical optimization. The maximum amount of mature coconut water was used since the goal of this study was to maximize the amount of coconut waste used as an ingredient. Calamansi and stevia were set to a minimum to guarantee that the product resulted in a pH < 4.5, an essential product characteristic to ensure safety, while stevia allows the use of maximally mature coconut water. Sensory parameters, such as sweetness, sourness, and overall acceptability, were targeted to reach a score of 6.00 on a 9-point hedonic scale to ensure that the acceptance was at least ‘like slightly’. The acceptability of sweetness and sourness represents the balance between sugar and its acidity, while the overall acceptability encompasses the liking of the product as a whole [39]. The importance of +  +  +  +  + (5) was assigned to mature coconut water and overall acceptability since these factors are important for attaining the objectives of valorizing mature coconut water and developing an acceptable beverage. On the other hand, the remaining parameters were set to +  +  + (3) as a default or with medium importance. Weights fine-tune how the optimization process searches for the best solution, and by default, the value is set to 1. The software generated ‘desirability’ scores for various combinations of mature coconut water, calamansi, and stevia, and only those scores exceeding 0.70 were considered for validation.

2.5 Physicochemical analysis

The pH of the beverages was determined using a pH meter (Horiba PH1100-S, Japan). The titratable acidity (TA) was calculated using endpoint titration with 0.1 N sodium hydroxide (NaOH) until the pH 8.2 was reached, and TA was expressed as malic acid (%) [40]. Instrumental color determination was performed using a colorimeter (HunterLab Colorflex EZ, Virginia, USA) [41]. The L* (lightness to darkness), a* (redness to greenness), and b* (yellowness to blueness) were measured. Furthermore, the chroma (C*) and hue angle (°h) were derived using Eqs. 5 and 6.

$${C}^{*}= \sqrt{{a}^{{*}^{2}}+{b}^{{*}^{2}}}$$
(5)
$$\circ \,h={tan}^{-1}\left(\frac{{b}^{*}}{{a}^{*}}\right)$$
(6)

2.6 Ethics clearance

All participants who underwent sensory evaluation signed a written informed consent to the protocol approved by the Philippine Health Research Ethics Board, National Ethics Committee (NEC 2021–006-Luna-Coconut Water) in accordance with the ethical standard as laid down in the Declaration of Helsinki and its later amendments or comparable standards.

2.7 Sensory evaluation

Two groups of panelists were recruited for this study. The 1st group consisted of 72 panelists (21 males and 51 females), and the 2nd group consisted of 50 panelists (18 males and 32 females). The participants were 18–49 years of age, regular juice drinkers, not allergic to any food, and not pregnant or lactating. The 1st group was used to evaluate the effects of the product formulation, and 6 samples were evaluated. Sample presentation was based on a replicated modified extended Latin square [42]. The 2nd group was used for model evaluation, and all three (3) samples consisting of optimal and suboptimal samples were evaluated using a randomized complete block design.

Prior to the sensory evaluation, 50 ml of each sample was transferred to a white sample cup and refrigerated for at least 12 h. Each sample was assigned a unique 3-digit code. The evaluation was conducted based on appearance, color, mouthfeel, sweetness, sourness, flavor, and overall acceptability. A 9-point hedonic scale was used to rate the samples, with 1 representing ‘dislike extremely’, 5 representing ‘neither like nor dislike’, and 9 representing ‘like extremely’. The panelists were reminded to evaluate the samples one at a time, consume a portion of the crackers, and drink water between evaluations to cleanse the palate.

3 Results and discussion

3.1 Effects of formulation on physicochemical properties

Table 3 presents the influence of physicochemical properties such as pH and instrumental color values (L*, a*, b*, C*, and °h) on the flavor of the mature coconut water. Only pH exhibited a significant (p < 0.05) quadratic predictive model, whereas the L* value, b* value, and C* showed significant (p < 0.05) linear predictive models. However, both a* and °h showed nonsignificant quadratic predictive values. The regression model for pH was highly reliable at 97.93% in predicting the beverage’s pH, with an adjusted R2 of 97.24%. The coefficients presented in Table 3 indicate the relative impact, demonstrating how much the mean of the response variable changes with a one-unit shift in the factor while keeping the other variables constant.

Table 3 Effect of formulation on the physicochemical properties of flavored mature coconut water
Full size table

Generally, the acidity of juices generally decreases with increasing maturity, although this can vary between different fruits [43]. As illustrated in Fig. 2a, the pH of the flavored mature coconut water increased, and its acidity decreased with an increase in the mature coconut water content is attributed to the low acidity of mature coconut water [44]. Conversely, the addition of calamansi decreases the pH and increases the acidity of the beverage, which is predictable given the lower pH of calamansi compared to that of mature coconut water. This observation aligns with findings by Theba, Nayi, and Ravani [45]. Akonor [43] also observed a similar trend when orange juice was added to a fruit juice cocktail comprising soursop, pineapple, and mango. The inclusion of calamansi juice successfully lowered the pH of the flavored mature coconut water to below 4.6 due to its high citric acid content (5.52%) [46]. Kunitake et al. [47] explained that acidification promotes enzymatic and microbiological stability and facilitates pasteurization of the juice. The generated model indicated that the combination of mature coconut water (X1) and calamansi (X2) significantly (p < 0.0001) decreased the pH by 0.2208. Although this value is relatively small compared to the individual effects of mature coconut water (3.97) and calamansi (3.58), their combination significantly contributes to the pH changes in the flavored mature coconut water.

Fig. 2
figure 2

Contour plots for a pH, b L* value, c b* value, and d chroma. Only models with significant (p < 0.05) R2 and nonsignificant (p > 0.05) lack of fit are presented. Plots were generated by Design Expert software

Full size image

The color of beverages plays a crucial role in consumer acceptance, as it reflects quality, safety, and nutritional value [48]. The International Commission on Illumination (CIE) has recommended the utilization of CIELAB as the most suitable and preferred color specification for food, as it closely aligns with the sensory and chemical characteristics of the food under evaluation [49, 50]. According to the instrumental color analysis of the experimental runs, the ‘darkness to lightness’ (L*) values ranged from 24.68 to 32.82, the ‘greenness to redness’ (a*) values ranged from − 4.17 to − 2.48, and the ‘blueness to yellowness’ (b*) values ranged from 28.57 to 39.82. Similar findings regarding L* and a* values were also documented by Sun et al. [51] in a turmeric-fortified pineapple juice beverage.

In Table 3, only L*, b*, and chroma (C*) exhibited a linear predictive model, while a* and hue angle (°h) yielded a quadratic predictive model. Regarding the reliability of predicting the responses, only the L* and C* values exceeded 80%, with adjusted R2 values ranging from 0.8826 to 0.9818, respectively. Furthermore, concerning the L* and b* values, all factors (X1, X2, and X3) significantly (p < 0.05) contributed to this response variable with a one-unit shift while keeping the other variables constant in the model. The variables also exhibited nonsignificant (p > 0.05) responses to a* and °h, with values of 0.1065 and 0.3415, respectively. In this study, the flavored mature coconut water derived from various formulations exhibited a yellowish hue, largely contributed by calamansi and turmeric powder. Turmeric contains main yellow compounds, such as curcumin, demethoxycurcumin, and bis-demethoxycurcumin, which are heat-stable [24, 52], while calamansi's carotenoids also contribute to its yellow to red or red-bluish colors [48]. The yellowness of the product aligns with the positive b* values obtained in this study. Figure 2b, c revealed that increasing the amount of calamansi while holding turmeric constant significantly (p < 0.05) influenced the yellowness of the juice, while increasing the amount of mature coconut water reduced the yellowness. The opposite effects, however, were observed for the L* values. Similarly, Chew et al. [46] reported comparable observations. The colors of each experimental sample are depicted in Fig. 3.

Fig. 3
figure 3

Mature coconut water combined with calamansi, stevia, and turmeric at various combinations as presented in Table 4

Full size image
Table 4 Actual combinations1 of mature coconut water, calamansi juice, and stevia powder generated by Design Expert for D-optimal mixture design
Full size table

Chroma is associated with color intensity and signifies saturation [53]. Similarly, as depicted in Fig. 2d, the trend for the C* values mirrors that of the b* values. As the amount of coconut water decreased and the amount of Calamansi increased, the C* values increased. With a decrease in coconut water and an increase in calamansi, the C* values increased. This phenomenon may result from the loss of carotenoid pigments such as violaxanthin, cis-violaxanthin, and antheraxanthin, along with the isomerization of 5,8-epoxids. Additionally, a study by Prathapan et al. [24] revealed that heating turmeric to temperatures near 60–100 °C decreased browning in turmeric while enhancing its yellowness and brightness.

3.2 Effects of formulation on sensory properties

Sensory evaluation involves assessing the qualities of a food through the senses [54]. The experimental beverages were evaluated using a 9-point hedonic scale, and the appearance, color, mouthfeel, sweetness, sourness, flavor, and overall acceptability of the sample were evaluated. In general, the majority of the variables (X1 and X2) had a positive influence on the sensory parameters of the flavored mature coconut water, although the extent of their effects differed, as presented in Table 5. Stevia was consistent and had a negative influence on the beverage, particularly in terms of color, sweetness, flavor, and overall acceptability. Among all the sensory parameters used, only mouthfeel, sweetness, sourness, flavor, and overall acceptability showed a significant (p < 0.05) model with a nonsignificant (p > 0.05) lack of fit, while appearance and color had a nonsignificant (p > 0.05) regression, suggesting that they may not be highly accurate. Pearson correlation tests revealed that sweetness, sourness, and overall acceptability were highly correlated (p < 0.01) with each other; hence, we focused on these parameters.

Table 5 Effect of formulation on the sensory properties of flavored mature coconut water
Full size table

According to Curi et al. [55], consumers are more inclined to accept sweet juice. From Table 5, the linear coefficients (X1, X2, and X3), as well as the quadratic coefficients (X1X3 and X2X3), exhibited a significant (p < 0.05) influence on the sweetness of the flavored mature coconut water. The contour plot (Fig. 4a) indicated that increasing the amount of mature coconut water and stevia led to an increase in the sweetness score of the beverage, whereas increasing the amount of calamansi resulted in lower sweetness acceptability, likely due to the perceived sourness from calamansi. These observations align with a study conducted by Chauhan et al. [6], which reported increased acceptance of sweetness. Additionally, a study by Bertelsen et al. [56] revealed a negative correlation between sweetness and sourness (p < 0.05), thus establishing a sweet–sour dimension.

Fig. 4
figure 4

Contour plots for a sweetness, b sourness, c flavor, and d overall acceptability. Only models with significant (p < 0.05) R2 and nonsignificant (p > 0.05) lack of fit are presented. Plots were generated by Design Expert software

Full size image

Table 5 shows the effects of raw material combinations on sourness expressed as a cubic model. Linear (X1, X2, and X3), quadratic (X1X3 and X2X3), special cubic (X1X2X3) and cubic (X1X3(X1-X3) and (X2X3(X2-X3)) effects had significant (p < 0.05) influences on sourness. The combinations of mature coconut water, calamansi, and stevia had varying effects on sourness, as shown in Fig. 4b. The interaction of these raw materials suggests that the acceptability of sourness is influenced by the appropriate balance of the brix:acid ratio [46, 57].

In terms of flavor, the combination of mature coconut water and calamansi was significantly influenced (p < 0.005) by the linear variables (X1, X2, and X3) and quadratic variables (X1X3 and X2X3). A notable degree of influence was observed with stevia (102.02) and its interaction with mature coconut water (130.67) and calamansi (140.60), indicating that increasing stevia contributes to a favorable flavor profile (Fig. 4c). King et al. [58] conducted a study on the impact of sweeteners/sweetness on changes in flavor perception, particularly the fruity character in beverages, and found that flavor-specific fruity character increased with perceptible sweetness regardless of whether the sweetness originated from sucrose or aspartame/acesulfame-K. This phenomenon was also observed by Bertelsen et al. [56], where sucrose affected flavor intensity by a factor of 17.52 at p < 0.05.

Consumers’ unfamiliarity with novel foods can alter their expectations and potentially negatively affect sensory perception and overall acceptability [59]. Mature coconut water typically has a bland taste and low sweetness. However, the incorporation of calamansi, stevia, and turmeric powder altered these characteristics by modifying its flavor, increasing its sweetness, and enhancing its color. The overall acceptability of the developed flavored mature coconut water ranged from 5.35 to 7.19%. The formulations consisting of 90.3% mature coconut water, 8.00% calamansi, and 1.50% stevia had the highest overall acceptability. Among the coefficients listed in Table 5, the two-factor effects of mature coconut water (X1) and calamansi (X2) with stevia (X3) significantly (p < 0.05) influenced the overall acceptability of the flavored mature coconut water, with coefficient values of 88.79 and 99.83, respectively. Figure 4d shows a contour plot depicting the effect of the variables on the overall acceptability of the beverages. Similar findings were reported by Chauhan et al. [6] when lemon was added to mature coconut water. The addition of fruit juice has a positive effect on the acceptability of the resulting beverage.

3.3 Numerical optimization and validation of models

From the responses previously discussed, only the a* value, hue (°h), appearance, color, and aroma were not significant (p < 0.05) and were not included as part of the specifications for desirability. For the purpose of this research, only four (4) dependent variables—pH, sweetness, sourness, and overall acceptability—were chosen for desirability. An ‘optimal’ sample was produced with 90.08 g/100 g mature coconut water, 8.83 g/100 g calamansi, 0.90 g/100 g stevia, and 0.20 g/100 g turmeric powder. The ‘suboptimal sample’ contained 84.06 g/100 g of mature coconut water, 14.86 g/100 g of calamansi, 0.89 g/100 g of stevia, and 0.20 g/100 g of turmeric powder.

Validation experiments (Table 6) using “optimal” and “suboptimal” samples revealed that none of the pH values met the prediction interval. However, all the values obtained were still < pH 4.5. This indicates that the product is highly acidic and demonstrates enzymatic and microbiological stability [46]. For the other variables, such as sweetness, sourness, and overall acceptability, the actual values from the optimal sample were within the prediction interval, while most of the experimental results from the suboptimal samples revealed otherwise. Therefore, it can be inferred that the selected combination of raw materials yielded accurate predictions of the pH and sensory properties of the beverage within the desired range.

Table 6 Comparison of the experimental results of the validation studies to the predicted values
Full size table

4 Conclusion and recommendations

The optimization of the calamansi, stevia, and turmeric using D-optimal mixture design with mature coconut water was conducted to improve its sensory properties while addressing concerns on waste generation. A total of 24 runs were generated by the Design Expert and numerical optimization were carried out revealing that pH, sweetness, sourness, and overall acceptability were significant (p < 0.05) responses. This study then used the ‘desirability’ function of the software to achieve a multiresponse optimization of the raw materials. The generated combinations were 90.08 g/100 g mature coconut water, 8.83 g/100 g calamansi, 0.90 g/100 g stevia, and 0.20 g/100 g turmeric powder. Validation of the optimized formulation revealed that the model generated had a good prediction on the properties of flavored beverage with a pH < 4.5, and sweetness, sourness, and overall acceptability within the predicted intervals. The results of this research may help industries valorize mature coconut water instead of discarding it as waste. The developed beverage may be further improved to increase its acceptability and generalizability by increasing the range of the values used and verify results through pilot-scale production. Another potential future research could include market study to determine product placement in the market and how consumers will perceive the product once available. The characteristics—such as proximate, nutrient, antioxidant properties—may be further explored to determine if the product has functional properties compared to mature coconut water. Shelf-life determination at various storage conditions might be a viable area for future research.