Evaluation of Two Cosmetic Products Formulated with Essential Oil Extracted from Copal Resin Obtained in Agroforestry Systems

Abstract

Forest wealth, combined with innovative cosmetic applications, allows for the use of non-timber forest products like copal resin, diversifying resource use, promoting sustainable practices, preserving ecosystems, and generating income. The aim of this study was to evaluate a facial cream and a body gel formulated with essential oil extracted from copal resin obtained from agroforestry systems to determine their quality and consumer acceptance. Copal resin was collected from a silvopastoral system in Izúcar de Matamoros, Puebla, Mexico. Protocols were developed to ensure quality, safety, and efficacy, adhering to Mexican official standards, and accelerated stability tests were conducted to determine shelf life. Microbiological and irritability tests were performed to assess safety. Hedonic tests along with a random forest model were employed to identify the most important characteristics for consumer acceptance. The results indicated that both products met quality parameters regarding color, aroma, viscosity, and pH, and exhibited a shelf life of two years. Both products were free from harmful microorganisms, making them suitable for human application. Hydration, aroma, spreadability, and irritability were the most crucial variables for achieving higher consumer acceptance. Cosmetics can be formulated with copal resin essential oil.

1. Introduction

Non-timber forest products (NTFPs) play a crucial role in providing income and resources for many rural communities and indigenous people whose lives depend on forests. These products encompass a wide range of resources, including medicinal plants, fruits and nuts, bee products, ornamental plants, resins and gums, and environmental services [1]. It is essential to emphasize that the collection and commercialization of NTFPs must be carried out sustainably and responsibly. This ensures both the preservation of forests and their biodiversity and the ongoing ability of local communities to depend on these resources in the long term. Many countries, including Mexico, use NTFP species not only as a source of economic income, but also as an important part of their culture and religious practices [1]. Copal, a plant found in deciduous tropical forests, stands out for its aromatic resin that has been used in religious practices for centuries [2].

Copal is a highly versatile material in Mexico, used in various ways across different regions of the country. Indigenous communities often burn it in religious ceremonies and rituals to purify the environment, establish contact with gods and spirits, and seek protection [3]. Additionally, it has been used in traditional medicine [4] to treat ailments such as respiratory issues, headaches, and body pains, as well as to reduce stress [5]. In some parts of Mexico, copal is used as an ingredient in beauty and personal care products, such as soaps, creams, and perfumes, due to its relaxing properties [6]. It is also employed in crafting and decorating textiles. Globally, copal has gained popularity as a natural product and can be found in esoteric and natural product stores. In summary, copal is highly valued in Mexico for its sweet, balsamic aroma, as well as its healing and spiritual properties.

Copal resin is obtained from cuts made in the bark of trees belonging to the Burseraceae family, specifically the genera Protium and Bursera. In Mexico, predominant species include Bursera glabrifolia, Bursera bipinnata, Bursera fagaroides, Protium copal, Bursera simaruba, and Bursera cuneata [3]. However, forests hosting copal species are diminishing due to the increasing demand for its resin [4]. In many regions of Mexico, copal resin is primarily used for religious purposes, but due to its anti-inflammatory properties, it has also been employed in herbalism to treat headaches [7]. In recent years, natural and sustainable cosmetic products have become increasingly popular as consumers seek products with natural ingredients. In some countries, cosmetics have also been more strictly regulated due to safety concerns. As a result of this trend, natural cosmetics, although currently representing a small percentage of the industry, are expected to grow by 15% compared to the 5% of conventional cosmetics [5].

In Mexico, official standards (NOM) for cosmetics establish the minimum quality and safety requirements that cosmetic products must meet to be marketed. Limits are set for specific ingredients, such as preservatives and colorants, as well as labeling and packaging requirements [8]. Natural cosmetics are regulated in the same way as conventional cosmetics, meaning they must comply with NOM requirements. In Mexico, various organizations such as Ecocert, COSMOS, and the National Association of Natural Product Companies (ANEP) promote natural and sustainable cosmetics, offering certifications and seals that guarantee their quality and natural origin [9].

Natural cosmetic products are marketed as healthier and more sustainable options, but they are not exempt from risks. The Mexican Federal Commission for the Protection against Sanitary Risks (COFEPRIS) has identified some natural cosmetic products containing hazardous components. Therefore, consumers should verify that the products they purchase comply with established safety and quality standards [10]. In this context, aromatic copal resin can be an excellent raw material for the production of personal care products, adding more economic value if crafted locally in the same community. Therefore, the aim of this study was to evaluate two cosmetic products formulated with essential oil extracted from copal resin obtained from agroforestry systems to determine their quality and consumer acceptance.

2. Materials and Methods

2.1. Study Area

The study site is located in the “Cerro de la Virgen” in the municipality of Izúcar de Matamoros, southwest of Puebla, Mexico, within the Agua Escondida micro-watershed, at an elevation of 1282 m and covering an area of 30.8 ha (Figure 1). The primary climate is warm subhumid with summer rains and an average annual temperature ranging from 22 °C to 24 °C. Annual total precipitation varies from 800 to 1000 mm. The predominant soils are leptosol and vertisol [11]. The vegetation consists of deciduous tropical forest, typical of regions with two well-defined seasons of moisture availability—the rainy and the dry seasons [12]. It is characterized by the presence of species that shed their leaves during the dry season and have papery, scaly barks or thorny, corky protuberances with sparse and very open crowns [13], as well as a high species diversity [14].

2.2. Copal Resin Collection

Two varieties of copal resin were utilized—high-quality white copal and black copal (Figure 2). The high-quality white copal was commercially obtained and was indicated to have been extracted in the state of Oaxaca on 30 November 2022; this sample was used as a control in the study. Black copal was collected from Bursera bipinnata and Bursera copalifera trees at the study site “Cerro de la Virgen”, located in Izúcar de Matamoros, Puebla, on 7 December 2022, from a silvopastoral system (SSP).

To obtain the essential oil intended for cosmetic product production, steam distillation was performed using a semi-industrial 19 L distillation unit. The procedure involved the following: (1) crushing the resin to a particle size of approximately 0.5 cm in diameter using a mortar and pestle; (2) mixing copal resin and water in a 1:10 ratio, i.e., 500 g of commercial resin with 5 L of water, and for the SSP sample, 366.97 g with 4.7 L of water; and (3) distilling the mixture by steam at a temperature of 80 °C. The heating time was 4 h for both samples (commercial resin and SSP resin) [15,16,17,18].

2.3. Formulation of Protocols and Physicochemical Analysis of Product

Protocols for the production of body gel and face cream were developed following the Mexican standard NOM-259-SSA1-2022 [19] and based on the principles of natural cosmetics, emphasizing formulas with few ingredients and minimal processing requirements [7]. (1) Organoleptic Properties: Color and odor were determined based on sensory perception. The cream displayed a bright white color under white light, free of lumps, with the characteristic scent of copal essential oil. The gel, on the other hand, showed a light brown translucent color under white light, free of lumps and visible particles, and also had the distinctive scent of copal essential oil. No inconsistencies in color were observed, which is attributed to the proper implementation of the proposed protocols that ensure the consistency of the produced batches [20]. (2) pH: The pH was measured using a HANNA^® potentiometer (Hanna Instruments, Smithfield, RI, USA), with 1 g of gel and cream dispersed in 30 mL of distilled water at a temperature of 25 °C. Two measurements were taken, and their averaged values were obtained [20]. (3) Viscosity: The viscosity was measured using an NDJ-9S viscometer, with 150 mL of gel and cream placed in 200 mL beakers, at a speed of 2.5 rpm and a temperature of 25 °C, using spindle #5 for 5 min. Three measurements were taken, and their averaged values were obtained [20].

After determining quality parameters, safety tests were conducted. For each product, six batches were produced—three batches underwent accelerated stability tests; one batch underwent sensory, microbiological, and irritability tests; and two batches underwent tests to identify differences between products from commercial resins and resins obtained from silvopastoral systems. The product presentations were as follows: (1) face cream: commercial presentation in a 50 g aluminum jar, and the sensory study presentation was in a 25 g transparent plastic jar; (2) body gel: commercial presentation in a 100 mL blue Boston bottle, while the sensory study presentation was in a 40 mL Boston bottle.

The components of both the face cream and body gel formulations, along with the active ingredients and excipients for each, are below:

Face cream: Water, Bursera sp. Hidrosol (10%), Rosa rubiginosa oil, Oryza sativa oil, Ricinus communis oil, stearic acid, glycerin, tocopherol, carbomer, triethanolamine, cetyl alcohol, Germall Plus (diazolidinyl urea, iodopropynyl butylcarbamate (IPBC), and propylene glycol), propylene glycol, and Bursera sp. essential oil (0.25%).

Water, Rosa rubiginosa oil, Oryza sativa oil, Ricinus communis oil, stearic acid, glycerin, tocopherol, carbomer, triethanolamine, cetyl alcohol, Germall Plus (diazolidinyl urea, iodopropynyl butylcarbamate (IPBC), and propylene glycol), and propylene glycol supplier—droguería cosmopolita.

Bursera sp. Hidrosol and Bursera sp. essential oil supplier—obtained through distillation.

Body gel: Water, Bursera sp. Hidrosol (10%), Heterotheca inuloides extract, Cucumis sativus extract, royal jelly extract, carbomer, Bursera sp. essential oil (0.25%), Germall Plus (diazolidinyl urea, iodopropynyl butylcarbamate (IPBC), and propylene glycol), and triethanolamine.

Heterotheca inuloides extract, Cucumis sativus extract, and royal jelly extract supplier—bio Extracto.

Water, Germall Plus, and triethanolamine supplier—drogueria cosmopolita.

Bursera sp. Hidrosol and Bursera sp. essential oil supplier—obtained through distillation.

Microbiological, Accelerated Stability, and Irritability Tests

To determine product safety, microbiological, accelerated stability, and irritability tests were conducted, including the following:

Microbiological Tests: Determining the presence of fungi, aerobic mesophiles, and yeast using nutrient agar and potato dextrose agar media according to NOM 089 [21]. Identifying potentially harmful microorganisms in cosmetic products.

Determination of Shelf Life: An accelerated stability test was conducted according to NOM 073 [22] to determine the shelf life of both products. The shelf life of both products on the shelf was calculated after three months using the following formula:

$$Shelf life product={\displaystyle \frac{1}{5}}\left[\left(T_{oven}-T_{ambient}\right)\ast M\ast 2\right]$$

(1)

where $T_{oven}$ is the temperature in the oven, $T_{ambient}$ is the ambient temperature, $M$ is the time in months the product was in the oven at the specified temperature, and 2 is a constant with temperature −1 units. During the evaluation period, the quality of the product parameters did not change [23].

Irritability Test: the irritability test was conducted based on NOM 039 [24] to ensure that both products can be safely used by humans.

2.4. Product Acceptance and Effectiveness

To determine the acceptance of the products and their effectiveness, 60 volunteers were evaluated at the beginning and after a month of daily product use.

Hedonic Tests: (1) Two questionnaires were used to determine the likability and the characteristics that made them more acceptable to the public [25]. (2) Two formulas were subjected to 60 untrained participants. The attribute evaluated was the aroma, and assessments were monadic and balanced. The primary scent was evaluated upon opening the jar, and skin odor was evaluated when applying the product on the skin. (3) A numerical hedonic scale was used to evaluate the acceptance of the formulas by the panelists, where 1 represents the lowest score and 5 the highest. The consistency of the product and its effects on the skin were also assessed [26,27].

Dermatological Characteristics: (1) Moisture values, fat percentages, and skin elasticity were determined using the IMATE detector with bioelectrical impedance pulse technology. (2) Three measurements were taken on the back of the right hand, three on the left hand, and four in different areas of the face (forehead, chin, and right and left cheek) before and after applying the product to determine if there were changes in skin elasticity, hydration, or fat content.

Statistical Analysis of Dermatological Characteristics and Construction of the Random Forest Model

Comparison of Means of Dermatological Characteristics: Descriptive data include the mean value with standard deviations (mean ± SD) for continuous variables, and number with percentage (n, %) for categorical variables. The effect of the cream and gel was compared at 3 points in the study (at the beginning, after the first application, and after a month of using the product) using paired-sample t-tests and Wilcoxon signed-rank tests to investigate changes within the group. Effect sizes (ESs) were calculated for within-group analysis using Cohen’s d for the t-test and r for the Wilcoxon signed-rank test. For the statistical analysis, the R software was used [28].

Table 1. A summary of effect sizes and their interpretations.

Interpretation	Cohen’s d (t-Test)	r (Wilcoxon Signed-Rank Test)
No effect	0.00 to <0.20	0.00 to <0.10
Small effect	0.20 to <0.50	0.10 to <0.30
Moderate effect	0.50 to <0.80	0.30 to <0.50
Large effect	≥0.80	≥0.50

presents a summary of ES interpretation.

Determination of Main Dermatological Characteristics:

Response Variables: The perception of cosmetic products triggers emotional and cognitive responses with a direct impact on consumer acceptance. Therefore, efforts to reinforce these perceptions are crucial for consumer preference and loyalty [29]. Cosmetic product perception influences how users perceive, evaluate, and ultimately accept or reject them [30]. Consumer expectations are created from product promises, and when fulfilled, acceptance is increased. Consequently, consumers may be more inclined to accept a product or a brand if perceived to follow ethical and sustainable practices [31,32]. According to Hunt et al. [33], package design and product presentation positively influence product acceptance expectations. In this context, the response variable was as follows: Product Perception—“Good”, “Regular”, or “Bad”.

Explanatory Variables: Cosmetic products are intrinsically related to the improvement of personal appearance. The probability of product acceptance significantly increases when consumers perceive that it contributes positively to their self-image [33]. Product acceptance depends on its scent, how it spreads on the skin, whether it produces a greasy sensation, causes irritation, how it is absorbed, and the sensation it leaves after application [31,33]. The sensory-dependent elements that influenced the selection of both products were as follows:

• Product aroma: ☐I don’t like it at all ☐I don’t like it ☐I neither like nor dislike it ☐I like it moderately ☐I like it a lot.
• Product aroma on the skin: ☐I don’t like it at all ☐I don’t like it ☐I neither like nor dislike it ☐I like it moderately ☐I like it a lot.
• Spreading of the product: ☐Does not spread at all ☐Does not spread ☐Spreads lightly ☐Spreads moderately ☐Spreads easily.
• Greasy sensation of the product: ☐No greasy sensation at all ☐No greasy sensation ☐Slight greasy sensation ☐Moderate greasy sensation ☐Greasy sensation.
• Irritation of the product: ☐I don’t feel irritation at all ☐I don’t feel irritation ☐I feel slight irritation ☐I feel moderate irritation ☐I feel irritation easily.
• Absorption of the product: ☐Does not absorb at all ☐Does not absorb ☐Absorbs lightly ☐Absorbs moderately ☐Absorbs easily.
• Skin sensation after using the product: ☐I don’t like it at all ☐I don’t like it ☐I neither like nor dislike it ☐I like it moderately ☐I like it a lot.
• Product effect on softness: ☐I don’t feel softness at all ☐I don’t feel softness ☐Feel slight softness ☐Feel moderate softness ☐Feel softness easily.
• Product effect on hydration: ☐I don’t feel hydration at all ☐I don’t feel hydration ☐Feel slight hydration ☐Feel moderate hydration ☐Feel hydration easily.
• Product effect on skin appearance: ☐I don’t feel improvement at all ☐I don’t feel improvement ☐Feel slight improvement ☐Feel moderate improvement ☐Feel improvement easily.

Decision Trees: a classification random forest model was developed and validated to explain the perception of cream and gel products generated with copal resin based on their explanatory variables.

Random Forest (RF): The RF is a classifier that uses decision trees to identify important predictor variables by using a percentage (usually 70%) of the so-called training data, and with the remaining data called test data, the model’s performance is estimated [34]. Predictor variables are important in classification models if their omission increases the out-of-bag error (OOB). Each classification tree in this model is derived from bootstrap samples to enhance prediction performance [35]. Each tree split is determined using a random subset of normalized input variables. The model’s outcome is the average outcome of all trees [36]. Seventy percent of the data were used for model training, and the number of trees in the forest (ntree) and the number of variables randomly sampled as candidates for each split (mtry) were pre-defined. These parameters were optimized to minimize generalization errors (OOB) to obtain the best predictive power. With this algorithm, explanatory variables are selected at each decision tree node using starting data [37]. The remaining 30% of the data were used to validate the model, which is the most important step to ensure the scientific relevance of the results [38]. Model predictions were made using the randomForest package [39] in the R software [40].

Accuracy Evaluation: The test observations were used to estimate the performance of each model using the OOB method [41]. Prediction accuracy was verified using the confusion matrix [42]. The overall prediction accuracy of a model is its ability to correctly predict a specific class or type of forest in this case. The sum of correct predictions divided by predictions made is calculated.

Identification of Significant Variables: The relative importance of variables was determined using the Mean Decrease Accuracy, one of the most common metrics in random forest models [43]. It measures the loss of accuracy caused by the removal of a specific variable. Higher variable ratings are, therefore, more important for data classification [44].

Partial Dependence Plots (PDPs): PDPs show the marginal effect that one or two features have on the predicted outcome of a machine learning model [45]. The partial dependence plot shows whether the target and a feature have linear, monotonic, or complex relationships.

3. Results

3.1. Development of Protocols and Physicochemical Analysis of Products

Table 2. Quality certificate for the evaluation of physicochemical parameters of produced body gel batches.

Parameter	Batches
	G2802231	G0103232	G0203233	G13032023	GN16032023 GB16032023
Characteristic Odor	Compliance	Compliance	Compliance	Compliance	Compliance
Color (light brown)	Compliance	Compliance	Compliance	Compliance	Compliance
pH (6.5–7.5)	6.5	6.74	6.7	6.7	6.60 6.58
Viscosity (cP) 60,000–75,000	70,000	70,020	70,060	70,000	70,010 70,020

and

Table 3. Quality certificate for the evaluation of physicochemical parameters of produced face cream batches.

Parameter	Batches
	C2802231	C0103232	C0203233	C13032023	CN16032023 CB16032023
Characteristic Odor	Compliance	Compliance	Compliance	Compliance	Compliance
Color (white)	Compliance	Compliance	Compliance	Compliance	Compliance
pH (6.5–7)	6.51	6.53	6.52	6.51	6.68 6.61
Viscosity (cP) 60,000–75,000	65,010	65,030	65,000	65,020	65,010 65,030

present the results of the quality control applied to the batches produced in accordance with the developed standardized work protocols. In the same table, it can be observed that in all batches, pH and viscosity, as well as odor and color, fall within the established acceptance ranges. The quality test results for batches GN16032023 and GB16032023, with essential oil obtained from silvopastoral resin and commercial resin, respectively, were similar.

Microbiological, Irritation, and Accelerated Stability Tests

In

Table 4. Microbiological tests for the face cream.

Culture Medium Used							Batches								Incubation Temperature
Nutrient Agar (NA)							C13032023								30 °C ± 2
Potato dextrose agar (SPD)							C13032023								25 °C ± 2
A. Aerobic Mesophiles (AN)
Date	25 September 2023						26 September 2023								29 September 2023
Day→ Sample ↓	1						2								5
M1	0		0	0	0	0	0		0		0	0		0	0	0		0	0	0
M2	0		0	0	0	0	0		0		0	0		0	0	0		0	0	0
B. Fungi and Yeasts (APD)
Date	25 September 2023					26 September 2023					29 September 2023					03 October 2023
Day→ Sample ↓	1					2					5					7
M1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
M2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0

M1: bulk product sample; M2: finished product sample.

, the results of two samples from batch C13032023 of the face cream are presented, indicating a microbial load of less than 10 CFU/g for aerobic mesophilic microorganisms and less than 10 CFU/g for molds and yeasts. The microbial limit was within the specifications for the determination of aerobic mesophilic microorganisms, molds, and yeasts, as outlined in NOM-089-SSA1-1994, allowing its release for further processing.

Table 5. Microbiological tests for the body gel.

Culture Medium Used							Batches								Incubation Temperature
Nutrient Agar (NA)							G13032023								30 °C ± 2
Potato dextrose agar (SPD)							G13032023								25 °C ± 2
A. Aerobic Mesophiles (AN)
Date	25 September 2023						26 September 2023								29 September 2023
Day→ Sample ↓	1						2								5
M1	0		0	0	0	0	0		0		0	0		0	0	0		0	0	0
M2	0		0	0	0	0	0		0		0	0		0	0	0		0	0	0
B. Fungi and Yeasts (APD)
Date	25 September 2023					26 September 2023					29 September 2023					3 October 2023
Day→ Sample ↓	1					2					5					7
M1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
M2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0

M1: bulk product sample; M2: finished product sample.

shows the results of two samples from batch G13032023 of the facial and body gel, indicating a microbial load of less than 10 CFU/g for aerobic mesophilic microorganisms and less than 10 CFU/g for molds and yeasts. Its microbial limit was within the specifications for the determination of aerobic mesophilic microorganisms, molds, and yeasts, as outlined in NOM-089-SSA1-1994, allowing its release for further processing.

Table 6. Irritability test per day for both formulations.

Face Cream		Body Gel
Average—day 3	0.075	Average—day 3	0.025
Average—day 5	0.05	Average—day 5	0
Average—day 7	0.075	Average—day 7	0
Final average	0.067	Final average	0.0083
Interpretation	Suitable for human use	Interpretation	Suitable for human use

displays the results of the average scores from 20 participants who used both products during a 7-day trial period, with each participant assigned a number indicating the presence of any skin damage based on NOM 039. In the same table, the results of the irritation tests indicate that both formulations are safe for human use, as their average values obtained are below 1.5, which is a value set by the standard as the maximum for considering cosmetic products suitable for human use.

Table 7. Accelerated stability test for the body gel.

Monitoring of Body Gel’s Physicochemical Parameters during Accelerated Stability Testing.
Batches	G2802231				G0103232				G0203233
Test→ Sample ↓	Odor	Color	pH	Viscosity *	Odor	Color	pH	Viscosity *	Odor	Color	pH	Viscosity *
18 May 2023	C	Ca	6.50	70,000	C	Ca	6.74	70,020	C	Ca	6.70	70,060
25 May 2023	C	Ca	6.51	70,020	C	Ca	6.70	70,000	C	Ca	6.70	70,050
1 June 2023	C	Ca	6.52	70,000	C	Ca	6.72	70,030	C	Ca	6.72	70,065
8 June 2023	C	Ca	6.50	70,030	C	Ca	6.70	70,000	C	Ca	6.71	70,000
19 June 2023	C	Ca	6.50	70,000	C	Ca	6.73	70,010	C	Ca	6.72	70,040
19 July 2023	C	Ca	6.53	70,010	C	Ca	6.73	70,000	C	Ca	6.74	70,000
18 August 2023	C	Ca	6.50	70,050	C	Ca	6.72	70,020	C	Ca	6.70	70,050

C: characteristic; Ca: light brown; * viscometer NDJ-9S needle: 5 and 2.5 rpm.

and

Table 8. Accelerated stability test for the facial cream.

Monitoring of Facial Cream’s Physicochemical Parameters during Accelerated Stability Testing.
Batches	C2802231				C0103232				C0203233
Test→ Sample ↓	Odor	Color	pH	Viscosity *	Odor	Color	pH	Viscosity *	Odor	Color	pH	Viscosity *
18 May 2023	C	B	6.51	65,000	C	B	6.53	65,030	C	B	6.52	65,010
25 May 2023	C	B	6.50	65,010	C	B	6.53	65,050	C	B	6.52	65,010
1 June 2023	C	B	6.52	65,000	C	B	6.52	65,010	C	B	6.52	65,025
8 June 2023	C	B	6.52	65,030	C	B	6.50	65,050	C	B	6.50	65,000
19 June 2023	C	B	6.50	65,050	C	B	6.53	65,010	C	B	6.50	65,030
19 July 2023	C	B	6.51	65,010	C	B	6.52	65,000	C	B	6.51	65,000
18 August 2023	C	B	6.52	65,050	C	B	6.52	65,020	C	B	6.50	65,020

C: characteristic; B: bright white; * viscometer NDJ-9S needle: 5 and 2.5 rpm.

present the results of the accelerated stability tests to determine the shelf life of both formulations based on the Mexican standard NOM-073-SSA1-2015. The results of the accelerated stability test indicate that both formulations remain stable for 2 years.

The shelf life of both the facial cream and the body gel was 24 months $\left({\displaystyle \frac{1}{5}}\left[40 °\mathrm{C}-20 °\mathrm{C}\right]\ast 3\ast 2\right),$ respectively. This value indicates that the products can be safely used for 2 years from their manufacturing date.

3.2. Product Acceptance and Effectiveness

In

Table 9. Perception of the facial cream and its comparison with conventional cosmetics.

Face Cream	Persons
	Number		Percentage
	Yes	No	Yes	No
Cosmetic use	40	20	67%	33%
Use of the product correctly	44	16	73%	27%
Improved hydration	54	6	90%	10%
Improved elasticity	30	30	50%	50%
Improved oiliness	32	28	53%	47%
Improved smoothness	54	6	90%	10%
Improved appearance	46	14	77%	23%
Potential customers	56	4	93%	7%
	Natural	Commercial
The choice between natural and commercial cosmetics	54	6	90%	10%

and

Table 10. Perception of the body gel and its comparison with conventional cosmetics.

Body Gel	Persons
	Number		Percentage
	Yes	No	Yes	No
Cosmetic use	40	20	67%	33%
Use of the product correctly	44	16	73%	27%
Improved hydration	40	20	67%	33%
Improved elasticity	44	16	73%	27%
Improved oiliness	54	6	90%	10%
Improved smoothness	30	30	50%	50%
Improved appearance	32	28	53%	47%
Potential customers	56	4	93%	7%
	Natural	Commercial
The choice between natural and commercial cosmetics	54	6	90%	10%

, the results of the hedonic tests for the facial cream and body gel obtained from the second questionnaire, conducted one month after the use of the products, are presented. The results of the hedonic tests for the facial cream indicated that 67% of the participants had used conventional cosmetics before the study, and 73% applied both products correctly. In total, 90% of the participants noticed an improvement in hydration and smoothness, 50% in elasticity, 53% in oiliness, and 73% in the overall appearance with the use of the facial cream. The results of the hedonic tests for the body gel indicated that 90% of the participants noticed an improvement in oiliness, 67% in hydration, 73% in elasticity, 50% in smoothness, and 53% in the overall appearance.

Among the participants who had previously used conventional cosmetic products (67%), 85% expressed a preference for natural cosmetics, while 93% of the participants expressed their willingness to incorporate natural cosmetic products into their personal care routine.

3.3. Statistical Analysis of Dermatological Characteristics and Construction of the Random Forest Model

3.3.1. Comparison of Means for Dermatological Characteristics

Table 11. Values obtained from the analysis of Wilcoxon rank-sum tests for the contrasts of different skin parameters.

Contrast	Hand				Face
	Left		Right
	r/d	Interpretation	r/d	Interpretation	r/d	Interpretation
Hydration: with and without product	0.62 W	Large effect	0.72 W	Large effect	0.58 W	Large effect
Hydration: after one month of use with and without product	0.80 W	Large effect	0.69 W	Large effect	0.80 W	Large effect
Hydration: without vs. without product after one month of use	0.53 W	Large effect	0.75 W	Large effect	0.56 W	Large effect
Oil percentage: with and without product	0.23 W	Small effect	0.15 W	Small effect	0.25 W	Small effect
Oil percentage per month of use: with and without product	0.35 W	Moderate effect	0.47 W	Moderate effect	0.37 W	Moderate effect
Oil percentage per month of use: without vs. without product at one month of use	0.45 W	Moderate effect	0.36 W	Moderate effect	0.32 t	Small effect
Elasticity: with vs. without product	0.65 W	Large effect	0.63 W	Large effect	0.59 W	Large effect
Elasticity at one month of use: with vs. without product	0.73 W	Large effect	0.63 W	Large effect	0.83 W	Large effect
Elasticity: without vs. without product at one month of use	0.04 W	No effect	0.25 W	Small	0.67 W	Large effect

W = Wilcoxon signed-ran, t = Paired sample t-test.

presents the results of the statistical analysis of measurements taken from the participants, evaluating skin characteristics in terms of hydration, elasticity, and oil percentage. The results indicated that both products exhibit a significant effect on these three aspects, and this effect is maintained over time, meaning it is not a “Cinderella effect” that only manifests during product use. Specifically, the gel shows the so-called “Cinderella effect” only in the elasticity characteristic. In this regard, when comparing this parameter before the first application of the gel and after a month of use, but before its last application, no statistically significant differences (p-value ≤ 0.05) were observed according to paired samples t-tests and Wilcoxon signed-rank tests.

3.3.2. Determination of Key Dermatological Characteristics

The number of trees, ntree, and the number of variables analyzed at each split, mtry, were determined using the exhaustive method. A total of 500 trees (ntree) and three variables for each node (mtry) were considered, calculated using the square root of the total number of predictor variables $\left(\sqrt{12variables}=3.46\approx 3\right)$. To enhance predictive power, the Mtry and ntree parameters of the RF model were optimized. The model’s performance was evaluated with 70% of field data (n = 70). The OOB error rate for the perception models of the cream and gel products was 11.9%, respectively. To obtain the minimum error rate, the number of variables, mtry, was set from 1 to 4 in the default RF model. Keeping the value of mtry fixed, ntree was traversed again. Four levels of ntree (100, 200, 500, and 1000) and four levels of mtry (1 to 4) were tested. For stable results, ntree was fixed at 1000 [34,46]. The RF error rate for both cream and gel was 11.9%, respectively, with mtry = 1 and ntree = 500.

Accuracy Evaluation: In total, 93% of product perception classes for both cream and gel were correctly classified with training data, and 83% with validation data, with the lowest OOB of 11.9% and the lowest number of classification trees being 500.

Table 12. Performance and cross-validation of the final model. Note: the validation dataset matrix is interpreted vertically.

Training Database	Good	Regular	Bad
Good	37	2	0
Regular	0	3	0
Bad	0	0	0
Validation Database	Good	Regular	Bad
Good	15	3	0
Regular	0	0	0
Bad	0	0	0

and

Table 13. Performance and cross-validation of the final model. Note: the matrix of the validation dataset is interpreted vertically.

Training Database	Good	Regular	Bad
Good	37	3	0
Regular	0	2	0
Bad	0	0	0
Validation Database	Good	Regular	Bad
Good	15	3	0
Regular	0	0	0
Bad	0	0	0

display the confusion matrices of the final model for the training and validation sets. The training data show that 93%, 100%, and 0% of the total respondents correctly rated the cream product as good, regular, or bad, while the validation data show that 83%, 0%, and 0% of the total respondents correctly rated the cream product as good, regular, or bad. Based on the training data, 95%, 100%, and 0% of the total respondents correctly rated the cream product as good, regular, or bad, while 83%, 0%, and 0% of the total respondents correctly rated the gel product as good, regular, or bad.

Identification of Significant Variables: Figure 3 shows the mean decrease in accuracy of the four most important predictor variables according to the principle of minimum OOB error. According to the mean decrease in accuracy, hydration was the most important variable within the cream product model. This variable has the greatest impact on the model’s accuracy, followed by the product’s aroma, spreadability, and irritability. This behavior is observed similarly in the body gel.

Partial Dependence Plot (PDP): Using the random forest (RF) model, partial dependencies were analyzed regarding the perception of the facial cream, considering the variables evaluated by the study participants—hydration, product aroma, ease of spreadability, and irritability. According to Figure 4, the cream was categorized as “Regular” when participants perceived mild hydration, expressed moderate liking for the product’s aroma, found the product to spread moderately, and experienced moderate irritation. The cream was categorized as “Good” when participants perceived moderate to high hydration and low or no intensity of the product’s aroma, reported medium or high spreadability of the product, and low or no irritability.

According to Figure 5, the body gel was categorized as “Regular” when participants perceived mild hydration, expressed moderate liking for the product’s aroma, found the product to spread moderately, and experienced moderate irritation. The body gel was categorized as “Good” when participants perceived moderate to high hydration and low or no intensity of the product’s aroma, reported medium or high spreadability of the product, and low or no irritability.

4. Discussion

4.1. Development of Protocols and Physicochemical Analysis of Products

The application of manufacturing protocols for both the facial cream and body gel complied with current regulations on cosmetic products, ensuring quality and safety standards for cosmetic production. Compliance with these protocols is essential for consumer protection [47]. Quality certificates for both the cream and the gel remained within acceptable ranges (physicochemical parameters), and their values were consistent across batches. This indicates that the manufacturing protocols ensure homogeneity in production, reducing variability between batches and ensuring consistent product quality over time [48]. This uniformity minimizes errors during processing and unexpected variations, enhancing product reproducibility [48,49].

Microbiological, Irritability, and Accelerated Stability Tests

Microbiological tests indicated that both the cream and gel were below the allowable limits for microorganisms, confirming strict adherence to established cleaning protocols. This certification ensures that the products have been manufactured to the highest quality and safety standards, promoting consumer safety and preventing skin infections or irritations [25,49,50]. Their relevance lies mainly in preventing skin infections or irritations [51]. The results of irritability tests revealed that both the gel and the cream are safe or suitable for human use, reinforcing consumer confidence in product use and demonstrating their suitability for daily personal care [24,26,48]. Accelerated stability tests demonstrated a 24-month shelf life for both the cream and gel, ensuring consumer safety and satisfaction. Precise expiration date estimation is crucial in the cosmetic industry to maintain customer trust and product integrity [48,52].

4.2. Product Acceptance and Effectiveness

Statistically significant differences were observed in skin hydration after applying both the facial cream and body gel (p-value < 0.05). Over 60% of participants reported improved skin hydration after one month of product use, attributed to formulations incorporating natural ingredients such as copal essential oil [53]. Copal essential oil’s composition enhances the efficacy of other ingredients, is compatible with human keratinocytes, and possesses hydrating and anti-inflammatory properties [54,55].

The body gel, containing cucumber extract [56], royal jelly extract [57,58], and arnica extract [59,60], showed a 9% increase in observed hydration, suggesting a synergistic effect of these ingredients. The facial cream, with rice oil and rosehip oil, demonstrated a 10% increase in observed hydration, facilitated by copal essential oil aiding the penetration of these components, known for improving skin appearance and contributing to hydration [61,62,63].

Skin elasticity improved due to copal essential oil’s presence, countering free radicals that reduce skin elasticity [64,65]. The anti-inflammatory properties of copal oil [54] also contributed to reducing tension associated with inflammatory processes. Compared to the gel, the cream demonstrated a greater impact on elasticity due to differences in formulation compositions. The gel incorporates cucumber, arnica, and royal jelly extracts, which have been shown to be beneficial for skin characteristics [56,58,60]; the face cream contains rice and rosehip oils, which have shown specific improvements in skin elasticity [61,62,63]. However, 73% of participants in the study reported improvements in elasticity with the body gel, compared to only 50% with the cream.

The percentage of oil on the skin, known as sebum, denotes the amount of natural fat produced by the skin, essential for its hydration and protection [66]. Both the gel and the cream reduced sebum production by 2%, which could be attributed to both products’ ability to reduce the alteration of the skin barrier, inflammation, and free radicals [67]. Overall, 90% of participants observed an improvement in this aspect with the gel, while 53% experienced an improvement with the cream. This can be attributed to formulation, as the cream contains fatty oils, leaving a greasy feeling after use, unlike the gel, which does not contain any ingredients causing this sensation. However, no significant modifications in this skin characteristic were foreseen for the formulated products since cucumber extract is the only component recognized as sebum-regulating [56]. The gel had a greater impact on elasticity (a 73% reported improvement) compared to the cream (50%). Both products reduced skin sebum production by 2%, contributing to barrier integrity, inflammation reduction, and free radical mitigation [55].

Determination of Most Relevant Variables for Product Acceptance

Hydration, aroma, spreadability, and irritability were crucial variables determining product categorization as “Good”, “Regular”, or “Bad”. Hydration, the most influential variable, is essential for skin function, appearance, and well-being [68]. Positive evaluations correlated with moderate to high hydration, while negative evaluations were linked to poor or regular hydration. Aroma was the second most important variable, influencing overall product perception [69,70]. Pleasant aroma enhances consumer perception and attractiveness. Spreadability, the third variable, influences user experience, and products with a smooth, even spreadability receive positive feedback [30]. Irritability, the fourth variable, directly impacts safety and consumer acceptance [25,26]. All four variables are interconnected, emphasizing the importance of proper application to enhance product efficacy and safety [48,71].

Within the model, hydration, aroma, spreadability, and irritability were the four key variables determining whether a product was considered “Good”, “Fair”, or “Poor”. The most important variable influencing different levels of acceptance for both the cream and the gel was “hydration”. Verdier-Sévrain and Bonté [68] point out that skin hydration is essential for maintaining its protective function, preventing skin ailments, improving its appearance, and providing a sense of well-being. Positive ratings are associated with moderate to high hydration, while low ratings are associated with poor or fair hydration. The second most important variable was the product’s aroma, which is a fundamental aspect in the cosmetic industry, as scent can enhance overall product perception. Lanzziano and Mora and Vaughn et al. [69,70] indicate that a pleasant aroma can improve consumer perception of the product, making it more attractive and enjoyable to use. The aroma emitted by copal essential oil should be moderate to low in order to consider both products of good quality. The third most important variable was spreadability. According to Batres and Kramer et al. [30], products with smooth and uniform spreadability facilitate their application and minimize the sensation of heaviness or stickiness. Uniform and fluid spreadability contribute to improving the user experience, thereby contributing to a positive perception of the products. Lastly, the fourth most important variable was irritability, a critical variable directly impacting consumer safety and acceptance. Products causing irritation can provoke adverse reactions on the skin, eyes, or other body areas, compromising user health and well-being [25,49]. The four variables are inherently related, as poor spreadability prevents gel and cream components from effectively contacting the skin, thereby compromising their ability to enhance hydration. Proper product distribution is crucial, as it avoids excessive concentrations of components in one area, which would interfere with hydration and could also cause irritations. Proper application also contributes to the balanced release of the product’s aroma [48,71].

5. Conclusions

The established protocols ensured the production of uniform batches meeting quality standards. Microbiological, irritability, and stability tests confirmed product safety for human use. Organoleptic characteristics significantly influenced consumer acceptance, and the identified key variables (hydration, aroma, spreadability, and irritability) offer insights for formula improvement. Copal essential oil emerged as an ideal raw material, promoting sustainable practices and generating income for communities. The study contributes to the utilization of non-wood forest products, supporting biodiversity preservation and sustainable income generation.