Prediction Model of Sacha Inchi Crop Development Based on Technology and Farmers’ Perception of Socio-Economic Factors

Abstract

Background: The sacha inchi plant contains nutrients that are beneficial for health, cosmetics, and food products, so it has the potential to be developed economically. The development of sacha inchi involves agricultural technology, which includes the treatment of fertilizer types that need to be studied in maintaining production and productivity. Objectives: This study aims to analyze the optimal type of fertilizer treatment on crop yields and the influence of farmers’ perceptions of socio-economic factors in the development of sacha inchi plants so that an optimal sacha inchi development prediction model is formed. Methods: The partial least squares structural equation modeling (PLS-SEM) method was applied in the study to analyze the effect of perception of socio-economic factors, and the adaptive neuro-fuzzy inference system (ANFIS) method was applied to predict the optimal fertilizer treatment model. Findings: The results of the PLS-SEM analysis showed that farmers’ perceptions of sacha inchi cultivation considered economic factors at a percentage of 50.66% and social factors at a percentage of 49.33% and had a linear relationship with the economic development of sacha inchi with a value of 0.832, although simultaneously, 77.7% was influenced by economic factors and social factors, and 22.3% was influenced outside these two factors. The ANFIS prediction results reinforced the results of the analysis, which stated that fertilizer treatment based on sacha inchi plant waste in the form of seed shells produced greater harvest weight compared with goat manure fertilizer. If farmers wanted to use a combination fertilizer, the used composition was 80% sacha inchi seed shell waste and 20% goat manure fertilizer or other materials; if the used composition was otherwise, the yield tended to decrease. This research contributes to the theory of production sustainability by integrating the optimal fertilization factor as a decision support unit in the model. Practically, this study recommends the application of fertilizer from the basic ingredients of sacha inchi seed shells to create a sustainable sacha inchi processing industry that ensures production stability, strengthens the agribusiness ecosystem, and develops the economic potential of sacha inchi and reduces the operational costs of sacha inchi farming by reducing the cost of purchasing fertilizer and supporting the circular economy model.

1. Introduction

The sacha inchi plant has the economic potential to be developed because this plant has a high beneficial value that can be processed to meet various needs and as a source of income. This can encourage the regional economy as a whole [1]. Sacha inchi (Plukenetia volubilis L.) is a plant in the Euphorbiaceae family originating from the rainforest in the Amazon, Peru [2]. Peru is the largest producer of sacha inchi, producing about 1200 tons per year, followed by Latin American countries such as Colombia, Brazil, Bolivia, and Ecuador and the Asian countries of China, Cambodia, Laos, and Thailand but not including Indonesia, which is still limited in production, and the Indonesian people are not very familiar with this plant [3].

As a nutrient source plant, sacha inchi contains 48–50% oil and 27–28% healthy protein [4,5,6,7,8]. There are a large number of valuable compounds found in sacha inchi seed oil [9], and it is recognized as having a high nutritional value beneficial for human health [10,11] and is also rich in unsaturated fatty acids [12,13,14,15,16]. The sacha inchi plant contains essential fatty acids of omega 6 and omega 3, which are good for dietary supplements [17]. People can use all parts of the sacha inchi plant, such as the seeds, seed coats, and leaves, so that their economic value can be maximized commercially [7]. This plant is important for food and pharmaceutical industry applications [18] and can be used as a sustainable food industry [14]. The sacha inchi plant has high oil stability, judging from its physical properties [19].

However, the development of sacha inchi in Indonesia is still limited as cultivation locations only exist in a few areas and people are not familiar with the potential of this plant, which suggests a low interest in cultivating it. One of the most important things in the production of modern agriculture is fertilization [20,21]. Based on research by Z.Q. Cai [22], the production of sacha inchi oil requires a high and appropriate amount of fertilizer. The current phenomenon is that sacha inchi farmers only use fertilizer made from goat manure since there are no other alternatives. Apart from fertilization, other influential and important factors in sacha inchi farming are socio-economic factors, which can be determinants in the implementation of agricultural activities [23]. Organic fertilizers are fertilizers made from natural materials such as manure, compost, and peat moss; organic fertilizers are generally less harmful to the environment but work slower than chemical fertilizers and are more expensive, but organic fertilizers can help increase important nutrients in the soil [24]. Organic fertilizers contain more complete nutrients and are high in organic matter content, which can improve the soil fertility and crop yield quality [25].

Some socio-economic factors that can be determinants in the implementation of agricultural activities include education level, farmer knowledge, community support/participation, market certainty, and price [26]. As is the case in Ethiopia, agriculture focuses on increasing the use of organic fertilizer because it can increase overall economic growth, and farmers are continuously encouraged to use manure technology [27].

Based on the explanation above, this research can be said to have novelty because there are not many studies that analyze the type of fertilizer treatment from similar plant waste materials with plants cultivated with the application of these fertilizers. This research can also support the circular economy, which is an economic model that targets the efficient use of resources through the use of waste, in this case the shell of sacha inchi [28] seeds, and can increase socio-economic benefits that are friendly to the environment. In addition, the circular economy model is also one of the supporting models for the Sustainable Development Goals in seeking to increase zero waste strategies until 2030. This is made clear by Singapore with Sustainable Singapore in Blueprint 2015, which targets increases in the recycling rate to 70%, domestic recycling rate to 30%, and non-domestic recycling rate to 81%, as well as several circular economy implementation policies in several other countries [27]. The used methodology is the integration of PLS-SEM and ANFIS, on the grounds that PLS-SEM can analyze significant variable relationships and can be used as reinforcement in predictive analysis as the optimal alternative to the ANFIS method, which is the main support in this method.

The integration of SEM-PLS and ANFIS can provide advantages in building predictions of technology-based sacha inchi crop development and farmers’ perceptions of socio-economic factors. SEM-PLS can be used to analyze the relationship between relevant variables, such as education, culture, farming experience, knowledge, and community support/participation, and economic factors, namely, market certainty, price, and access to capital. Meanwhile, ANFIS can be used to predict efficient fertilizer use patterns on sacha inchi farms so as to optimize future sacha inchi farm development strategies based on the obtained data. The integration of these two methods allows the development of sacha inchi farming strategies that are more tailored to the preferences of farmers based on social and economic factors. More specifically, the ANFIS method is applied to model the optimal type of fertilizer treatment, while the PLS-SEM method is used to analyze the perception of socio-economic factors on the development of sacha inchi crops so that the two analyses are expected to unite the prediction model of sacha inchi crop development based on optimal fertilizer treatment and perceptions of socio-economic factors. This study is a novelty from previous research that has been carried out by previous researchers. Research with a mixed-method approach has been carried out by [29], namely, by using PLS-SEM as a predictive power and fsQCA as an alternative solution supporting PLS-SEM.

At the research locations in Cikadu, Sindangkerta, Bandung Barat, and Cianjur, sacha inchi production is still limited, and yields are uneven on each tree. On this basis, it is important to carry out an in-depth study related to crop yields based on the treatment of fertilizer types and farmers’ perceptions of socio-economic factors in supporting more optimal sacha inchi development. Thus, this study aims to analyze various indicators of supporting crop yields and socio-economic factors in the development of sacha inchi plants to obtain a model for the development and sustainability of sacha inchi plants from the perspective of fertilizer type and socio-economic potential so that the results can be used as evaluation material for further sacha inchi management.

2. Materials and Methods

2.1. Research Locations

This research was carried out in Cikadu, Sindangkerta, Bandung Barat, and Cibokor, Cianjur, West Java Province, Indonesia, by taking into consideration that these areas have land suitable for the sacha inchi plant. Farmers who cultivate sacha inchi in those areas are still limited in production and yield weight, but they have easier market access than in other areas because there is already cooperation with buyers. Limitations in production and yield weight are a challenge as quality and sustainable production are the main things that need to be prioritized. The research was conducted in March 2023. The tools and materials used were sacha inchi cultivation tools, goat manure, and sacha inchi waste. The sacha inchi waste used for fertilization was seed shell waste.

2.2. Experimental Design

The cultivation of sacha inchi was carried out according to the customs of farmers in two locations. The fertilization of sacha inchi plants was performed based on the tested treatments. The types of fertilizers used were goat manure organic fertilizer and sacha inchi seed shell waste made into liquid organic fertilizer. The application of liquid organic fertilizer in both research locations used 100% goat manure and sacha inchi seed shell waste at a dose of 150 mL/liter of water. Fertilization was performed every two weeks by pouring it around the sacha inchi plants (on the disk).

2.3. Plant Sampling and Analysis

This study used primary data. The primary data were acquired from a sample of 200 trees using goat fertilizer and sacha inchi waste fertilizer from two regions. The observed variables were the number of potential fruits totaling 4 and 5, the thickness of the seed shell as an input variable, and the crop yield or production weight as an output variable. According to Chirinos et al. [30], the sacha inchi plant consists of dried fruit capsules, 30–35% of which are inedible shells that are used as raw materials for making liquid organic fertilizer and 65–70% of which are edible seeds.

Primary data related to socioeconomic factors were obtained from interviews with 150 farmer respondents in two regions. The studied variables were the level of education, culture, farmer knowledge, community participation, capital, and land ownership, as well as farming experience as variable X and the development of sacha inchi as variable Y. Next, a Likert scale ranging from 1 to 5, representing extremely low to very high values, was used to measure the data. Then each sample farmer chose one of the five options that were provided as a response to their perceptions of the development of sacha inchi farming. Data were obtained by distributing questionnaires to all sample farmers, as well as through interviews and field sampling to corroborate the results of the analysis.

2.4. Research Method

This study used a field survey method. Sampling was conducted using non-probability sampling, or more precisely purposive sampling, with the consideration that farmers in both research locations are currently concerned with the development and cultivation of sacha inchi. The sample was given a questionnaire containing a list of questions about the variables that, according to theory, are considered to influence the development of sacha inchi cultivation. In this study, it was first seen how the potential for the development of Sacha Inchi agribusiness in terms of socio-economic aspects. This is done by looking at the perceptions of sample farmers involved in the cultivation of sacha inchi plants. Perceptions are measured using a Likert scale from 1–5, ranging from strongly agree to strongly disagree with the statements in the questionnaire. The questions in this questionnaire were created based on previous research references that were considered relevant to the real conditions in the field. In addition to being given a questionnaire, to support or cross check each farmer’s questionnaire answers, in-depth interviews with farmers and related parties were also conducted. From there it could be illustrated how the socio-economic conditions of farmers could affect the development of sacha inchi plants in the research location. Then, to be able to produce production with high quality and productivity, experimental research was carried out through several fertilization treatments with the aim that researchers could model the optimal fertilization treatment to support the development of massive sacha inchi agribusiness. The collaboration of the two methods is considered very important to do considering the cultural values of farmers in Indonesia who do not easily accept new innovations. Therefore, to be able to develop new commodities massively, it is necessary to deepen the cultural values of farmers in the local area. High product productivity, good quality, and high economic value do not necessarily guarantee farmers to want to develop a particular commodity, including sacha inchi. Meanwhile, the sample of sacha inchi plants that used the treatment of goat manure liquid organic fertilizer and sacha inchi waste liquid organic fertilizer was determined to be 200 trees, which were located randomly in two planting areas. The data collection technique of this research can be seen in the stages of the research flow, as shown in Figure 1.

The initial stage of this research was to identify the problem to be studied. The problem identification was carried out in both regions of West Java Province by focusing on building a model of sacha inchi plant development based on fertilizer treatment and farmers’ perceptions of socio-economic factors. Previously, a preliminary study was conducted to select the method to be used for the creation of the sacha inchi plant development model based on fertilizer treatment and socioeconomic factors, namely, the adaptive neuro-fuzzy inference system (ANFIS) and partial least squares structural equation modeling (PLS-SEM).

Questionnaires to assess the farmers’ perceptions were created in the form of closed questionnaires. The questionnaire contained statements that corresponded to the indicators of each variable of social and economic factors. The questionnaire consisted of approximately 23 questions, of which the social factor variable consisted of 6 questions, the economic factor variable 3 questions, and the sacha inchi agribusiness development variable 14 questions. The questions contained indicators of research variables compiled based on previous research, which were adjusted to the real conditions in the field during the research. Each respondent chose one of the answers from 5 answer options that were provided, ranging from very positive answers to very negative answers. As for interviews with farmers and related parties, the questions were arranged with the development of a questionnaire that was created but did not extend beyond the boundaries that were set. At the time of this interview, the respondents could explain more broadly in response to the questions asked that were not covered in the answer options in the questionnaire.

The data processing began by dividing historical data into two parts: 150 as training data and 50 as checking data. Furthermore, the data were processed using MATLAB R2013a software. For data processing using the adaptive neuro fuzzy inference system (ANFIS) method, the data were first grouped into intervals. The results of the ANFIS and PLS-SEM analyses were then used as the basis for determining the development model of sacha inchi plants based on fertilizer treatment and perceptions of socio-economic factors.

Data Analysis Techniques

The analysis techniques used in this quantitative descriptive research were the adaptive neuro fuzzy inference system (ANFIS) and PLS-SEM methods. The ANFIS method was applied to model the optimal type of fertilizer treatment, while the PLS-SEM method was used to analyze the influence of the farmers’ perceptions of significant socio-economic factors so that from these two analyses, a model of sacha inchi plant development could be built based on fertilizer treatment and perceptions of socio-economic factors.

The Adaptive Neuro-Fuzzy Inference System (ANFIS)

The adaptive neuro fuzzy inference system (ANFIS) is an adaptive neural network based on a fuzzy inference system [31]. ANFIS was applied in this research because it is simple to comprehend, is highly adaptable, tolerates inappropriate data, is capable of modeling nonlinear data, and is able to build and directly apply the expertise of experts [32]. The method of ANFIS offers the benefit of modeling human knowledge’s qualitative side and the decision-making process’s mechanism through given commands [33,34]. Since artificial neural networks are based on incorporated historical data and can predict future events based on those data [35], they have the advantage of being able to recognize specific patterns, learn something unknown, and provide solutions for problems without the requirement for mathematical modeling [36]. ANFIS has the ability to learn by interpretation, which results in a powerful modeling tool [37,38], and automatically generates if–then rules with appropriate membership functions [39]. Figure 2 below shows the layer structure of ANFIS.

Figure 2 shows that there are 5 layers, where layer 1 is input data consisting of fertilizer use, the number of leaves, the number of stems, and potential fruit. In the second layer is the model formation by ANFIS, in layer 3 is the data testing process by the ANFIS application, in layer 4 is prediction testing based on training data and real data that have been entered into the application, and layer 5 is the output of the ANFIS prediction model results.

Figure 2 also shows there are two types of nodes: adaptive nodes with square icons and nodes with circle icons. The output of each layer is denoted by Oj, where O is the number of rules, and j is the number of layers.

The ANFIS network consists of five layers as follows [25,40,41].

Layer 1 (fuzzy layer)

Each node in layer 1 is an adaptive node, which means that the parameter value can change with the following node function:

O1, 1t = μA1 (Zt − 1)

(1)

O1, 2t = μA2 (Zt − 1)

(2)

O1, 3t = μB1 (Zt − 2)

(3)

O1, 4t = μB2 (Zt − 2)

(4)

where (Zt − 1) and (Zt − 2) are the inputs at the i-th node. Meanwhile, −1 and −2 are the membership functions of each node. The degree of membership of each input to fuzzy sets A and B is expressed as 1, with 1, 2, 1, 2 being linguistic variables. The membership function used is the generalized bell membership function. The generalized bell membership function can be written as follows:

$${\mu }_A\left(x_t\right)=e^{\frac{{-(x_t-c)}^2}{{2\sigma }^2}}$$

(5)

$$f x;a_i,b_i,c_i=\frac{1}{1+\frac{{x-c_i}^{{2b}_i}}{a_i}}$$

(6)

in which AND are a set of parameters called premise parameters. By taking the value = 1, only the parameters AND will change during the process of learning. The change in the values of these parameters will also change the generalized bell curve.

Layer 2 (product layer)

Each node in layer 2 is a non-adaptive node, which means the parameter values are fixed. The function of this node multiplies each incoming input signal as follows:

$$O_{2,i}=w_i=\mu A_i\left(x\right)·\mu B_i\left(y\right);i=1,2$$

(7)

Each output node expresses the activation degree of each fuzzy rule. The number of rules formed follows the number of nodes in this layer.

Layer 3 (normalization layer)

Each node in this layer is a non-adaptive node that states a normalized degree function that is the ratio of the i-th node output in the previous layer, which is written as follows:

$$O_{\mathrm{3,1}}=w_t\frac{w_i}{w_1+w_2}, and i=1,2$$

(8)

The function can be expanded if there are more than two rules by dividing it by the total number of w for all rules.

Layer 4 (defuzzification layer)

Each node in this layer is an adaptive node with the following node function:

$$O_{\mathrm{4,1}t}=w_{1t}^{*}{Z}_t^{\left(1\right)}=w_{1t}^{*}\left({\alpha }_1{Z}_{t-1}+{\beta }_1{Z}_{t-2}+{\gamma }_1\right),$$

(9)

$$O_{\mathrm{4,2}t}=w_{2t}^{*}{Z}_t^{\left(2\right)}=w_2^{*}\left({\alpha }_2{Z}_{t-1}+{\beta }_2{Z}_{t-2}+{\gamma }_2\right)$$

(10)

where ${\alpha }_i$, ${\beta }_i$, ${\gamma }_i$ is the set of parameters of the node and is called the consequent parameter.

Layer 5 (total output layer)

Layer 5 is the last layer that functions to add up all inputs with the following node function:

$$O_{5t}={\widehat{Z}}_t=w_{1t}^{*}{Z}_t^{\left(1\right)}+w_{2t}^{*}{Z}_t^{\left(2\right)}$$

(11)

Partial Least Squares Structural Equation Modeling (PLS-SEM)

This study used the PLS-SEM analysis technique to analyze the effect of the perception of socio-economic factors on the development of sacha inchi plants. The significance level used for this research was 95%. The statistical measurement scale used in this research was an ordinal measurement scale. This study used the ANFIS and PLS-SEM approaches since the two approaches reinforced each other; when the SEM results only measured how great the influence was, it was strengthened by the results of the ANFIS approach.

3. Results and Discussions

3.1. Farmers’ Perceptions of Socio-Economic Factors Affecting Sacha Inchi Farming Development Using PLS-SEM

The development of sacha inchi farming in an area is not immediately accepted by local farmers. There are perceptions by farmers that will have an impact on the development of sacha inchi farming. Farmer perception is a process farmers go through in interpreting certain information from their surroundings [10], and this perception can be used to assess the tendency of farmers to support or reject this sacha inchi farming [42]. The results of the analysis of the perceptions of sacha inchi farmers based on socio-economic factors are shown in Figure 3 below.

Figure 3 shows the perceptions of sample farmers in the two research locations. The social factor variable showed a percentage of 49.33%, which meant that based on social factors, the communities in the study locations had the potential to develop sacha inchi farming. However, the perception of sample farmers stated that economic factors were considered to be more of a major consideration for the development of sacha inchi, which was indicated by the value of the perception of 50.66%. This was supported by the economic potential of sacha inchi seed in the form of flour, which is beneficial for health and can be processed into snacks so that it has economic opportunities [43] and can also be utilized for industrial applications [18]. Simultaneously, it can be seen that the influence of social and economic factor variables was approximately 77.23% (according to manual calculation), and the rest was influenced by other variables that were not studied. This showed that economic and social factors were related to each other, so if the two were combined, the influence would be even greater on the development of sacha inchi farming.

3.2. Analysis of Socio-Economic Factors Affecting the Development of Sacha Inchi Farming

3.2.1. Measurement Model Specifications

This study described the measurement model specifications as follows:

• Social factor variables were measured by indicators S1 (education level), S2 (culture), S3 (farming experience), S4 (knowledge), and S5 (community support/participation). The construct measurement model used indicators S1 (education level), S2 (culture), S4 (knowledge), and S5 (community support/participation).
• The economic factor latent variable was measured by indicators E1 (market certainty), E2 (price), and E3 (capital access). The economic factor construct measurement model used indicators E1 and E2, namely, market certainty and price.
• Sacha inchi farming development variables were measured by indicators P1 (harvest quality), P2 (income), P3 (resource efficiency), P4 (production and productivity), P5 (market and marketing), P6 (land area cultivated), P7 (food safety), P8 (product diversification), P9 (group participation), P10 (development of sacha inchi technological innovation), P11 (development of sacha inchi special education and training), P12 (cooperation network), P13 (sustainability), and P14 (farmer welfare).

The measurement model for the construct of sacha inchi farming development used indicators P4 (production and productivity), P6 (land area cultivated), P8 (product diversification), P9 (group participation), P10 (development of sacha inchi technological innovation), and P11 (development of sacha inchi special education and training). Figure 4 shows the PLS model path diagram that explains the relationship pattern between latent variables and their indicators based on the research conceptual framework.

3.2.2. Structural Model Specifications

The structural model specifications are expressed by the figure below.

3.2.3. Evaluation of Measurement Model

The evaluation of the measurement model was performed on each latent variable by performing validity and reliability tests for each construct. The validity test was used to determine whether the questions were valid or invalid. Invalid question items were then discarded or not used in the research instrument. Validity was divided into two categories: convergent and discriminant validity.

Convergent validity was the loading factor value of the latent variable and its indicators. The first stage of testing this convergent validity was to look at the loading factor value and t-statistic. The measure of an indicator was said to be valid if it had a loading factor value with the latent variable being measured >0.5 and a p-value of less than 0.05. Invalid indicators in this study were cut from latent constructs. Reduction was carried out on the indicator with the lowest loading factor to the point where it could produce a model with valid indicators (

Table 1. Loading factor value of final stage construct indicators.

Latent Variable	Indicator	Loading Factor	p-Value
Farming development	P4	0.532	0.034
	P6	0.637	0.006
	P8	0.599	0.033
	P9	0.683	0.000
	P10	0.599	0.033
	P11	0.504	0.038
Social factors	S1	0.692	0.004
	S2	0.858	0.000
	S4	0.914	0.000
	S5	0.740	0.002
Economic factors	E1	0.538	0.034
	E2	0.972	0.000

Source: Processed primary data, 2023.

). Only indicators that had convergent validity could measure their latent constructs, so they could be used as indicators in the model.

Invalid indicators were spread across almost all variables, be it the variable of sacha inchi farming development, social factors, or economic factors. Then, the second stage of testing this convergent validity was to look at the average variance extracted (AVE) value for each latent variable. The expected AVE value of each latent variable to be declared valid was >0.5 [44]. Based on the results of the obtained AVE value, it could be seen that the AVE value for all variables had a value of >0.5 (valid).

The discriminant validity test was performed by comparing the AVE root value with the correlation between latent variables. Based on the measurement results in

Table 2. AVE value, AVE root, and composite reliability.

Latent Variable	AVE	AVE Root	Composite Reliability
Farming development	0.792	0.889	0.919
Social factors	0.705	0.839	0.877
Economic factors	0.747	0.864	0.850

and

Table 3. Correlation between latent variables.

Latent Variable	Farming Development	Social Factors	Economic Factors
Farming development	1.000	0.524	0.878
Social factors	0.524	1.000	0.524
Economic factors	0.878	0.878	1.000

, it can be seen that all root values in latent variables were greater than the correlation value between variables (

Table 4. Hypothesis testing results.

Causality Relationship	Path Coefficient	Standard Error	p-Value	Description
Social factors > farming development	0.088	0.177	0.614	Non-Significant
Economic factors > farming development	0.832	0.144	0.040	Significant

).

The reliability test was performed by looking at the composite reliability value. The reliability test was intended to determine whether and how much the research variables could be used to test the problem or not. The expected composite reliability value was >0.7 to be considered reliable. The calculation results in Table 2 show that the composite reliability value obtained was more than 0.7, so it could be concluded that the indicators were consistent in measuring the construct variable.

The valid (convergent and discriminant) and reliable results showed that the indicators used in this study had good reliability and were able to measure latent constructs.

3.2.4. Structural Model Evaluation

The structural model was evaluated by observing the coefficient of determination (R-square) value of the dependent latent variable. Next was to determine the relationship coefficient between statistically significant variables to test the PLS model hypothesis. The significance level used in this study was 0.05, so the p-value must be smaller than 0.05.

The result of the R-square value of the farming development latent variable was 77.7%, meaning that 77.7% of the development of sacha inchi could be explained by social and economic factors. Meanwhile, the remaining 22.3% was explained by other variables outside the model. Economic factors were seen in market certainty and selling prices, while factors outside economic and social included the fact that sacha inchi was not a commodity yet since it was not widespread in the community, and information related to this plant was still limited. The R-square value of 0.777 (77.7%) could be said to be substantial; in other words, the inner model fit the data so that the research could be declared valid and reliable, as shown in Figure 5.

Based on the results of hypothesis testing, a variable that was quite significant in the development of sacha inchi farming was the economic factor, which had a significance value of 0.040 and a path coefficient of 0.832. The positive sign on the path coefficient showed that economic factors had a linear relationship with the development of sacha inchi farming, such as a condition of market certainty and the high selling price of sacha inchi plants that could encourage and motivate farmers to engage in sacha inchi farming. This was supported by the knowledge that the sacha inchi plant has great potential as an economic and valuable crop [45]. In addition, this crop can be cultivated in three different conditions, open area, mixed, and agroforestry, making it economically more viable [25]. Sacha inchi plants can also be cultivated in subtropical conditions [11]. Furthermore, many technologies have been used to extract sacha inchi oil, one of which is hydraulic cold press extraction (HCPE), which can increase oil yields while maintaining its nutritional integrity [46]. Referring to the current conditions and the results of the analysis, which stated that economic factors were the perception of farmers in conducting sacha inchi farming, there were also other factors to consider, including the longer lifespan of the sacha inchi plant compared with other crops, more stable yields, price certainty as there was buyer certainty, and resistance to pests and diseases, which clarified the conclusion that economic factors were higher than social factors for farmers in conducting sacha inchi farming. Farmers who cultivate sacha inchi plants in Cikadu and Cianjur areas are increasing in number with the support of perceptions of economic factors so that they have a higher economic opportunity.

3.3. Analysis of Supporting Factors for Crop Yields on Sacha Inchi Development with Fertilizer Treatment Using ANFIS

This part describes the training process in ANFIS. There were two stages to making predictions using the ANFIS method: training and testing. The least square estimator (LSE) was used in the training phase to fix the consequent parameters on screen four in order to produce parameter values. The training process also propagated the value of test data back through backpropagation employing gradient descent to improve the parameter of the premises in screen one. The training of the ANFIS method programming was held in two classes, with a maximum epoch of 100, an error of 0–1, and a rate of learning ranging from 0.6 to 0.9. Figure 6 shows that the training data and output of the prediction look accurate because as shown in Figure 6a, the input data into ANFIS intersect directly with the training data accurately and totally. This is shown in Figure 6b, which visualizes the position between the training data and the FIS output. So according to the data validation, the standard with the smallest RMSE value is 0.5, and it can be seen that the training data plot (blue) follows the pattern of the testing data (red).

The ANFIS analysis process had five main layers of data sources that were processed as input data and imported into the model. The data processed as input data produced ANFIS programs in the form of rules that would later regulate the output data requirements as needed in the model to be created. The output data were model data that were formed from the rules built in accordance with the input data that was processed at the beginning of this study. The output data were the weight of the sacha inchi harvest production. Important variables to determine the accuracy of ANFIS were the number and type of MFs, the best possible technique, and the MF type produced. The limit of inputs was determined by autonomous factors, namely, fertilizer treatment, the number of potential fruits, and the shell thickness. Meanwhile, the outcome variable was the weight of the sacha inchi yield. The level of improvement of ANFIS model was trained with Gaussian MF as the best type from several chosen MFs. The type of work of enrollment was selected and immediately sorted due to its ability to reduce errors. The construction of FIS was completed with the ideal strategy of backpropagation, and the resistance error was 0, as can be seen in the figure below.

The sacha inchi plant development model based on fertilizer treatment was built using modeling in ANFIS with three input data and one output data (Figure 7). The three input data consisted of fertilizer treatment, potential fruit, and shell thickness; the output data were the weights of the sacha inchi yields.

Each input data type had a different member function, including the input data of fertilizer use treatment that had two member functions, namely, goat manure and sacha inch waste material; potential fruit had three member functions (few, medium, and many), as well as shell thickness (thin, medium, and thick), as shown in Figure 8.

The combination of input and output data resulted in 18 rules, as shown in Figure 9 and Figure 10.

In one of the results, the structure of the developed ANFIS model was obtained, namely, [1 19.4 122 20.8], which concluded that the cultivation of sacha inchi using goat manure fertilizer gained a potential fruit range of 19 with an analyzed shell thickness of 122 mm, resulting in weight yields of around 20.8 g, as shown in Figure 9.

This is inversely proportional to Figure 10, which shows the use of sacha inchi waste fertilizer. The analysis results indicated a range of potential obtained fruits of 19 with an analyzed shell thickness of 122 mm, resulting in weight yields of around 285 g. This suggested that for fruit weight compaction, sacha inchi waste fertilizer was more efficient than goat manure fertilizer.

Figure 11 shows a three-dimensional graph of ANFIS programming results based on training input data, in which the maximum weight of the sacha inchi yield (ranging from 200 to 400) was obtained as a result of the use of pure sacha inchi waste fertilizer treatment. The maximum use of sacha inchi waste fertilizer increased the harvest weight; if the use of sacha inchi waste was reduced, the harvest weight level tended to decrease. Fertilizer was important in increasing the crop yield and quality [47], including in sacha inchi cultivation. There are many studies that explore the effect of fertilizer on yield growth and crop quality [48,49], as well as the raw materials in the manufacture of fertilizer, in this case, sacha inchi waste, that can contribute to increasing crop yield.

The graph in Figure 11 also explains that sacha inchi cultivation using a combined fertilizer treatment with a composition of 80% sacha inchi waste and 20% goat fertilizer provided maximum results. However, if the combined fertilizer was made with the reverse composition, the results tended not to be optimal; therefore, the ANFIS prediction results with a composition of 80 percent sacha inchi waste material can produce high productivity and can be applied. Proper fertilization can increase crop productivity and provide maximum yields that can alleviate poverty among smallholder farmers; this is also true for sacha inchi crop farmers. The results of the analysis from the two approaches corroborate each other in that economic factors are the perception of farmers in engaging in sacha inchi farming. This can be seen from the results of the SEM-PLS analysis, which were in line with the development of sacha inchi plants, and reinforced by the results of the ANFIS predictions, which confirmed that good results would be obtained from fertilizing with sacha inchi waste, even though only 80 percent of the composition was obtained. This condition can be taken into consideration by farmers when applying this crop development technology, especially in fertilizer treatment.

3.4. ANFIS Model Prediction Accuracy

The ANFIS method analysis was quite good with the resulting RMSE value of 0.032. Figure 12 shows the comparison of the observation data as the training data (series 1) had no significant difference from the ANFIS prediction output (series 2).

Based on the results of the training data input into ANFIS, the prediction accuracy was 96%. If linked to research input data, this showed that all variables used as training data in ANFIS had a direct influence on the results.

3.5. Implementation of the Sacha Inchi Plant Development Model

The development of sacha inchi is currently still limited in quantity and quality so that it will have an impact on the quality and availability of raw materials in the processing of sacha inchi products. This research produces model predictions with fertilization technology made from sacha inchi waste that can optimize the production of sacha inchi plants. So the results of this model prediction can contribute to solving problems in the development of this plant, and it can be seen that there is a prediction–model relationship produced with fertilization technology.

The development of sacha inchi agriculture can be maximized according to the model analysis applying ANFIS data processing using three inputs consisting of fertilizer treatment, potential fruit, and shell thickness, which produce potential yield outputs. Based on the results of ANFIS analysis, the development model in sacha inchi farming can be maximized by applying innovative technology to fertilizer treatment, namely, by combining a composition of 80% sacha inchi waste fertilizer with 20% organic waste goat manure fertilizer or other animal waste materials. Through the innovation of the use of organic fertilizer with this composition, the maximum yield of sacha inchi plant cultivation will be obtained, as shown in Figure 13. This will be in line with the increased perception of farmers toward conducting sacha inchi farming, which will improve their economic and social factors.

4. Conclusions

The perception of farmers in developing sacha inchi plants was influenced by economic factors at 50.66%, compared with social factors, which were only 49.33%. The results of this analysis were reinforced by the results of ANFIS predictions that recommended the application of fertilization technology made from sacha inchi waste compared with goat manure, which could produce 200 to 400 as the maximum number to obtain higher yields. For combined fertilizers, the composition used was 80% sacha inchi waste and 20% goat manure or other animal manure materials; yields decreased if this composition was used in reverse.

The implications of the results of this study can contribute to the managerial accountability of sacha inchi development through the application of fertilization technology composed of sacha inchi waste, which can be used as the main ingredient or complementary material. Practically, it has implications for the application of fertilizers from the basic ingredients of the sacha inchi seed shell to create a sustainable sacha inchi processing industry that ensures production stability, strengthens the agribusiness ecosystem, and develops the economic potential of sacha inchi, as well as reduces the operational costs of sacha inchi farming by reducing the cost of purchasing fertilizers and supporting the circular economy model. However, the application of the results of this study is still limited to the application of sacha inchi plants. Thus, in future research, it is necessary to study the effectiveness and efficiency of the application of sacha inchi plant waste fertilizer processing on other crops.