Sustainable Business to Customer Value Co -Creation Model : A Novel Theory

1. Introduction

National Economic Development and Environmental Challenges. Currently, national development, particularly in the economic sector, tends to clash most strongly with . environmental concerns (ecology). Economic development and ecological preservation are like two opposing sides that are deeply interconnected—on one hand, economic development is necessary for societal welfare, but on the other, it inevitably impacts ecological sustainability. This is because economic terminology has largely failed to reconcile environmental concerns with improving societal well-being.

Despite the many achievements contributed by technology and the industrial sector from upstream to downstream in Indonesia, there has been a significant decline in natural resources and environmental quality. Industrial activities aim to process and utilize natural wealth, but in reality, excessive exploitation of natural resources has resulted in overproduction without considering responsible production practices. This, in turn, diminishes environmental sustainability and negatively affects human survival due to the absence of responsible consumption practices (Responsible Consumption) (Ibcsd, 2019).

The optimal implementation of Responsible Consumption and Production (SDG 12) has yet to be realized. There is a genuine need to change the current resource management and utilization patterns, which often have negative impacts on both environmental conditions and societal welfare. Conflicts in natural resource management remain evident due to strong sectoral egos, weak coordination and law enforcement, low human resource sensitivity, and the recurring issue of insufficient funding for managing responsible production and consumption. Ideally, economic actors—both producers and consumers—should recognize themselves as eco-centric human beings, acknowledging that humans are part of the environment rather than separate from it. Economic activities should consider environmental sustainability because they are not just about short-term gains but long-term sustainability. Economic growth and environmental preservation are interconnected to achieve SDG 12: Responsible Consumption and Production (Teodore, 2006).

Efforts to achieve Responsible Consumption and Production (RCP) require strategic management within companies to ensure sustainable competitive advantage and value realization. Hunger and Wheelen (1996) discuss strategic management in their work "Strategic Management", explaining that applying strategic management in profit-oriented companies is essential for understanding competitive forces and systematically developing a sustainable competitive advantage.

Organic Agriculture and Sustainable Development. One of the agricultural systems that aligns with SDG 12: Responsible Consumption and Production is organic farming, which promotes sustainability. In Indonesia, organic farming has been expanding annually. The organic farming land area increased from 69,605.9 hectares in 2007 to 251,630.8 hectares in 2019, according to Organic Farming Statistics (SPOI, 2020). The global organic farming industry is also growing, with Indonesia ranking 21st worldwide. Based on demand data for organic products since 2019:

Jakarta leads with 32% of total consumption demand. West Java ranks second with 21% demand.Since the COVID-19 pandemic, demand for organic products has surged. Organic vegetables had the highest demand (20%) Organic fruits accounted for 11.59% There was also a growing demand for processed organic products (SPOI, 2020). The pandemic further influenced consumer habits: 27% consumed organic products daily, 10% consumed twice a day, 9.41% consumed every three days, 5.29% consumed weekly (AOI, 2020)., In terms of supply, organic products experiencing growth include:Organic coffee, tea, and rice (3.1%), Organic palm sugar (8–17 million tons) (Directorate of National Export Development, Ministry of Trade, 2020).Horticultural commodities such as bananas, oranges, and lettuce (SPOI, 2020).

Government Support for Organic Horticulture. Horticulture is a priority sector in Indonesian agriculture aimed at increasing national revenue. Government initiatives include: Expanding dryland horticulture outside Java. Constructing post-harvest storage facilities in production centers

Developing organic horticultural villages. Between 2012 and 2015, Indonesia's vegetable production area expanded from 1,033,817 hectares to 6,370,751 hectares (BPS, 2017), increasing national vegetable production and consumption opportunities (Apriyani et al., 2018). Supply chain efficiency is crucial for maintaining a steady supply, with continuity of supply being a key factor (Hadiguna & Marimin, 2007). However, data on comprehensive organic product supply, including processed and horticultural commodities, is still lacking (AOI).

Challenges in Organic Agriculture Development. According to Yosini Deliana et al. (2019), organic agriculture faces challenges related to: Market limitations – While global demand has risen, domestic demand remains weak. Consumers are primarily eco-conscious individuals willing to pay premium prices. Mislabeling – Some supermarkets label conventional products as organic without proper certification. Farmer participation – Many farmers lack knowledge or interest in organic farming. High certification costs – Organic certification is expensive. Weak farmer organizations – Small farmers struggle to establish agribusinesses without strong support. Weak partnerships – Collaboration between farmers and businesses remains ineffective. Marketing & promotion – Organic vegetable producers still struggle with digital marketing strategies. A key success factor in organic agriculture development is the farmer-business partnership, especially for export markets. Research on Value Co-Creation (VCC) with a Value Co-Design approach remains limited, particularly for organic vegetable products in domestic and export markets.

Value Co-Creation and Digital Marketing in Organic Agriculture. Service-Dominant Logic (SDL) in marketing suggests that customers play an active role in value creation through interactions with products and services (Vargo & Lusch, 2004). However, research has often overlooked individual consumer impact on Value Co-Creation (VCC) (Hoyer et al., 2010; Sugathan & Ranjan, 2019). Prahalad & Ramaswamy (2004) introduced the DART Model of Value Co-Creation, emphasizing: Dialogue, Access, Risk/Benefit Analysis,Transparency. Further research is needed to apply the DART model in social networks and sustainable community engagement. Digital platforms, including mobile applications, have been identified as effective tools for organic product marketing (Arogundade et al., 2020). With the rise of social media marketing, platforms like Facebook, Instagram, WhatsApp, Twitter, and websites have become crucial for B2C (Business-to-Customer) interactions in the Industry 4.0 and post-pandemic era (Fagerström & Ghinea, 2020).

The Shift Toward Healthy and Eco-Friendly Lifestyles. The COVID-19 pandemic accelerated a shift towards eco-friendly, chemical-free lifestyles. Consumers now prioritize safe, nutritious, and environmentally friendly food, increasing demand for organic vegetables globally (Mayrowani, 2012). However, organic vegetable consumption in Indonesia remains low, both domestically and for exports. Research suggests that digital platforms can enhance value creation, ultimately benefiting businesses (Feng Zhu, 2024).

The Role of Organic Farming Communities in Indonesia Indonesia’s organic farming sector consists of individuals, farmer groups, and private sector companies, many of which are members of AOI (Aliansi Organik Indonesia). Since 2002, AOI has had 147 members across 18 provinces, including: 43 NGOs, 26 Companies, 16 Farmer Organizations, 37 Individuals. Given the low awareness among consumers and producers regarding organic food, research is needed on Value Co-Creation (B2C) models for organic vegetables, effective digital platforms, and sustainable business models.

2. Baground

2.1. Novelty and Literature Review of VCC in Organic Vegetables

The indexed journal sources used for identification included Scopus, ScienceDirect, and Google Scholar. The initial identification of articles from these three sources, using the specified keywords, resulted in 52,395 articles. The screening process was then conducted based on relevant articles/topics and further refined using keyword C, narrowing the selection to 210 articles. Subsequently, an eligibility assessment was performed, including backward and forward review analysis, leading to a final total of 55 articles for review.

Table 1. Keywords in Database Search and Number of Articles Found.

Kode	Kata kunci	Scopus	ScienceDirect	G.Scholar	Jumlah
A	“value cocreation”	5.795	700	44.900	52.395
B	awareness* OR experience* OR knowledge* OR preference*	521	53.024	397.000	450.545
C	“colearning*” AND “coservice*”	210	71	266	547
D	“Consumption green product*” OR “organic product*”	216	1.154	468	1.838

The review analysis method used is the PRISMA method (four-phase, flow-diagram) (Moher, Liberati et al., and Prisma Group, 2009; Vrabel, 2015), as explained in Figure 1. Based on the keyword search used in codes A, B, C, and D, the specific keywords were: A: ("value co-creation"), B: (knowledge* OR preference* OR awareness* OR experience*), C: ("co-learning" AND "co-service"), D: ("green product" OR "organic product").

Table 1. Keywords in Database Search and Number of Articles Found.

Kode	Kata kunci	Scopus	ScienceDirect	G.Scholar	Jumlah
A	“value cocreation”	5.795	700	44.900	52.395
B	awareness* OR experience* OR knowledge* OR preference*	521	53.024	397.000	450.545
C	“colearning*” AND “coservice*”	210	71	266	547
D	“Consumption green product*” OR “organic product*”	216	1.154	468	1.838

The review analysis method used is the PRISMA method (four-phase, flow-diagram) (Moher, Liberati et al., and Prisma Group, 2009; Vrabel, 2015), as explained in Figure 1. Based on the keyword search used in codes A, B, C, and D, the specific keywords were: A: ("value co-creation"), B: (knowledge* OR preference* OR awareness* OR experience*), C: ("co-learning" AND "co-service"), D: ("green product" OR "organic product").

Figure 2. Value Co-Creation Publications Per Year in 2022 (Since 2014).

Figure 2. Value Co-Creation Publications Per Year in 2022 (Since 2014).

Table 2. Summary of Discussions from Several Literature Reviews.

Table 2. Summary of Discussions from Several Literature Reviews.

2.2. Research Novelty in Marketing Strategy with Value Co-Creation (Co-Learning and Co-Service Design)

The state of the art of this literature review lies in the application of value co-creation using the co-learning and co-service design approaches. These two approaches have been previously applied in KIBS (Knowledge-Intensive Business Services) and Miobi companies in China, where value co-creation was utilized for higher education service management and Miobi’s specialized online marketing applications. However, there has been no prior research on enhancing value to increase the consumption of local organic vegetables and exports to improve supply availability (MAK Siddike et al., 2014; K Komulainen, 2014; K Kimla, K Muto et al., 2015; GA Tangalah, R Jenal, Thahaya, 2021; LA Donovan et al., 2018; J Pocek et al., 2022; M Guseppe et al., 2022).

2.3. The Gap Between Marketing Strategy and the Research Topic of Value Co-Creation.

Figure 3. Mapping of Bibliographic Value Co-Creation – Identified 8 Clusters.

Figure 3. Mapping of Bibliographic Value Co-Creation – Identified 8 Clusters.

Table 3. Top 10 Populer Keyword pada 8 Cluster Penelitian Value Co Creation.

Table 3. Top 10 Populer Keyword pada 8 Cluster Penelitian Value Co Creation.

Table 4. Top 10 Populer Keyword pada 8 Cluster Penelitian Value Co creation (Co learning dan Co service design).

Table 4. Top 10 Populer Keyword pada 8 Cluster Penelitian Value Co creation (Co learning dan Co service design).

3. Methodology

Our Role as Academics and Members of Society in Managing Natural Resources for Sustainable Welfare

As academics and members of society, we must prioritize the responsible management of natural resources to ensure the welfare and prosperity of communities. Natural resource management should align with the mandate of the Creator—to utilize resources responsibly while maintaining the sustainability of humanity, preserving nature and the environment, and avoiding destruction or over-exploitation. Our role as both producers and consumers must uphold the objectives of the Sustainable Development Goals (SDGs). One of the key SDGs related to production and consumption is SDG 12: Responsible Consumption and Production (RCP). This goal emphasizes sustainable consumption and production practices. One commodity that has seen increasing demand, particularly after COVID-19, is vegetables, which are essential for boosting immunity. Given Indonesia’s tropical climate, various types of vegetables can be cultivated effectively. Most farmers in Indonesia use conventional, semi-organic, and organic farming systems. Organic vegetable farming aligns with SDG 12’s RCP objectives, as it ensures health safety and environmental sustainability by avoiding synthetic chemical fertilizers, pesticides, and artificial growth hormones. These natural farming practices enhance agricultural productivity while ensuring organic produce benefits consumers by supporting immune health.

• Challenges in the Organic Vegetable Market

The demand for organic vegetables is high, both locally and internationally, yet the supply from agricultural businesses remains low. Organic vegetable consumers primarily belong to communities that already understand the importance of consuming healthy food and contributing to environmental preservation. However, both internal and external business factors continue to present challenges in increasing public adoption of organic vegetable consumption. Consumers play a central role in providing valuable business data regarding their needs, preferences, and expectations. Businesses must analyze consumer data to improve services and enhance mutual value creation. By understanding factors affecting demand, supply, and preferred vegetable varieties (both locally and for export), businesses can enhance Value Co-Creation (VCC) to drive profitability. The information collected from the market can be categorized based on co-creation strategies to develop a robust business model for Business-to-Consumer (B2C) VCC in the organic vegetable sector.

Bridging the Gaps in Organic Vegetable Consumption Through Value Co-Creation (VCC)

Analysis of previous studies reveals certain limitations in their implications, particularly regarding psychological factors influencing the acceptance and consumption of organic vegetables. Besides product quality, promotion, and distribution, online marketing systems have also evolved significantly during the COVID-19 pandemic. Digital platforms and internet-based applications have become essential for consumers in obtaining healthy vegetables, presenting an opportunity for businesses to sustain their operations. However, many existing websites and e-commerce platforms still face challenges in their implementation, possibly due to inadequate marketing campaigns, lack of socialization, or insufficient consumer education. Many platforms fail to meet consumer expectations in terms of service and engagement. By leveraging an effective Value Co-Creation (VCC) model, this research aims to develop a B2C-based digital platform for organic vegetables, ensuring better understanding, acceptance, consumption levels, frequency, and consumer awareness. Furthermore, it aspires to cultivate influencers who actively promote responsible consumption and production (RCP) in the organic vegetable sector.

Figure 4. Framework Diagram of VCC B2C Analysis for Organic Vegetables Using SEM LISREL.

Figure 4. Framework Diagram of VCC B2C Analysis for Organic Vegetables Using SEM LISREL.

• Description:
  • X₁ = Demand and its indicators
  • X₂ = Supply and its indicators
  • X₃ = Dialogue
  • X₄ = Access
  • X₅ = Risk/Benefit
  • X₆ = Transparency
  • X₇ = Sustainability (economic, social, environmental)
  • Y₁ = Value Co-Creation (co-learning, co-design, co-process, co-service)
  • Y₂ = Online marketing platform (14 indicators)
• Hypothesis Development

Table 5. Research Hypotheses.

Table 5. Research Hypotheses.

3.1. Research Method

The research was conducted in Indonesia, a location that serves as both a producer and consumer of organic and conventional vegetables, for the initial identification of organic vegetable business actors, business mentors, and a focus on West Java. The study specifically covered eight organic vegetable-producing regions: Sukabumi Regency, Cianjur Regency, Bogor, Depok, Bandung Regency, Garut Regency, Tasikmalaya Regency, and Ciamis Regency. The research was carried out from January 2023 to January 2024. The data collection technique used was purposive sampling, by selecting all business actors (organic vegetable producers) and mentors in West Java and 10 provinces (AOI data) as well as data from the Food and Horticulture Office. Additionally, random sampling was used to select 200 respondents consisting of organic and conventional vegetable consumers. The data sources consist of primary and secondary data. Primary data was obtained from questionnaires and direct interviews, product surveys, and focus group discussions (FGDs) involving and interview with google form for producers and consumers/communities of local and export organic vegetables, as well as non-organic consumers. Secondary data was sourced from organic horticulture businesses, AOI (Aliansi Organik Indonesia), relevant LSOs, IFOAM (International Federation of Organic Agriculture Movements), PAMOR (Organic Quality Assurance), export organic vegetable producers, stakeholders in export organic vegetable production, the West Java Food and Horticulture Office, the West Bandung Regency Agriculture Office, and the local Agricultural Extension Center (BPP).

3.2. Analyst Method

• A. Phase 1 Research:

The process of qualitative data collection is divided into two main categories: (1) Data collected by researchers through interviews, focus groups, or ethnographic field observations. (2) Pre-existing data in the form of documents before the study, such as public records, statistics, emails, etc.

For the second category, the data is already recorded, minimizing challenges in collection and management. Qualitative non-numeric data is obtained not only from primary data (interviews, focus group discussions (FGDs), and field observations) but also from secondary data, including: Written documentation and photographs, Researcher-recorded videos and YouTube videos, Content from various websites and WhatsApp, Discussions on social media platforms such as Twitter, LinkedIn, and Facebook, Non-numeric data from SPSS, Excel, and SurveyMonkey spreadsheets NVivo 11 Plus is used for visualizing data analysis results, including mind mapping, project mapping, and concept mapping. This tool is effective for content analysis, thematic analysis, and discourse analysis, The first phase of the research applies conditional analysis and participatory analysis (NVivo 11) and is validated using factor analysis.

• B. Factor Analysis Stages: (CFA Factor Analysis)

SPSS factor analysis is a method used to form factors in factor analysis using the SPSS application. Through this analysis, the following outcomes can be obtained:

• Identifying underlying dimensions or factors that explain the correlation among a set of variables.
• Identifying new, smaller variables to replace uncorrelated variables from a set of original variables that are correlated in multivariate analysis (such as regression analysis or discriminant analysis).
• Identifying key smaller variables that stand out from a larger set in a multivariate analysis.

• Steps:

• Rate/rank the most frequently mentioned demand and supply factors from the research results, based on references and FGD (Focus Group Discussion) findings.

4. Result

4.1. Research Area Conditions

West Java, also known as Tatar Sunda, is a province in Indonesia. The capital of this province is Bandung City. In 2021, the population of West Java reached 48,782,408 people, with a density of 1,379 people/km². Based on the 2010 BPS census, West Java is the most populous province in Indonesia, with the indigenous population being the Sundanese ethnic group. West Java is located in the western part of Java Island. The province is bordered by Banten Province, DKI Jakarta Province, and the Java Sea to the north, Central Java Province to the east, the Indian Ocean to the south, and Banten Province and DKI Jakarta Province to the west. The northern coastal area consists of lowlands, while the central region is mountainous, forming part of the mountain range that extends from west to east across Java Island. The highest point in the province is Mount Ciremay, located southwest of Cirebon City. Several major rivers, such as the Citarum River and Cimanuk River, flow into the Java Sea.

West Java has a tropical climate, with the lowest recorded temperature reaching 9°C at the summit of Mount Pangrango, while the highest recorded temperature reaches 34°C in the northern coastal areas. The average annual rainfall across the province is approximately 2,000 mm, with some mountainous areas receiving 3,000 mm to 5,000 mm of rainfall per year. West Java consists of 18 regencies and 9 cities. The cities that have been newly established since 1996 are: West Java Province consists of 18 regencies, 9 municipalities, 627 districts, 645 sub-districts, and 5,312 villages. In 2017, the estimated population reached 44,039,313 people, with a total land area of 35,377.76 km².

West Java Province's Economic and Industrial Potential. West Java Province has a high concentration of manufacturing industries, including electronics, leather, food processing, textiles, furniture, and aircraft industries. Additionally, geothermal energy, oil and gas, and the petrochemical industry are among the province's key economic drivers. The largest contributor to West Java’s Gross Regional Domestic Product (GRDP) is the manufacturing sector (36.72%), followed by hospitality, trade, and agriculture (14.45%), bringing the total contribution to 51.17%. Despite economic crises, West Java remains the center of modern textile and garment industries in Indonesia, whereas other regions are more focused on traditional textile production. The province contributes nearly a quarter of Indonesia’s total non-oil and gas production value. Its main export commodity is textiles, accounting for 55.45% of West Java’s total exports. Additionally, West Java is one of the few provinces in Indonesia that exports iron ore, albeit in small quantities. The province is also home to industries such as steel production, footwear, furniture, rattan, electronics, and aircraft components.

Known as one of Indonesia’s "national rice barns", nearly 23% of its total area of 29,300 km² is dedicated to rice production. As Indonesia’s "Production Hub", West Java contributes 15% of Indonesia’s total agricultural output. The province produces rice, sweet potatoes, corn, fruits, and vegetables, as well as commodities like tea, coconut, palm oil, natural rubber, sugar, cocoa, and coffee. The livestock sector is also significant, with 120,000 cattle, representing 34% of the national total. West Java has two major coastal fronts: the Java Sea to the north and the Indian Ocean to the south, with a coastline spanning around 1,000 km. This geographical advantage provides immense potential for the fishing industry. The estimated fish catch potential in the southern coastal waters is 1.2 million tons per year, but the actual catch has only reached 11,000 tons per year. The total fish production in West Java is 1.6 million tons annually, with 280,000 to 300,000 tons (17.5% to 18.75%) coming from capture fisheries. This means that fish catches from the southern coast account for only 3.67% of West Java’s total capture fisheries production. A report from the West Java Marine and Fisheries Department (DKP) states that there are at least 120,000 fishermen in the province, along with an estimated millions of people engaged in aquaculture. West Java’s inland water resources are not only derived from its many rivers but also from water reservoirs such as Saguling Dam, Cirata Dam, and Jatiluhur Dam, which not only generate hydroelectric power but also support agricultural irrigation and freshwater fisheries.

As of 2003, West Java had a population of approximately 37 million people, accounting for 16% of Indonesia’s total population. Urbanization has been growing rapidly, particularly in the Greater Jakarta area (Jabodetabek). In 2001, West Java had 15.7 million educated workers, making up 18% of Indonesia’s total educated workforce. The majority of them were employed in agriculture, forestry, and fisheries (31%), followed by manufacturing (17%), trade, hospitality, and restaurants (22.5%), and services (29%).

Table 6. Land Area and Organic Vegetable Production in Indonesia. West Java Ranks 3rd After Central Java and East Java.

No	Province	2019		2020		2021		2022
		Land Area (ha)	Production (Ton)	Producers	Land Area (ha)	Production (Ton)	Producers	Land Area (ha)	Production (Ton)
1	DKI Jakarta	8.4	4.2	0	20.15	10.075	0	20.15	10.075
2	West Java	147.53	865,257	117	138.8	928,397	131	140.87	1,091,040
3	Central Java	53.073	955,765	513	257.688	507,390	1,077	254.527	203,636
4	Special Region of Yogyakarta	2.305	94.48	17	2.105	166.1	12	1.78	140,942
5	East Java	80.588	1,828,439	237	75.087	639,362	306	42.357	1,188,106

Scheme 2022. Table 4.2 Respondent Characteristics 1.A. Mentors (2), Internal Data.

Table 6. Land Area and Organic Vegetable Production in Indonesia. West Java Ranks 3rd After Central Java and East Java.

No	Province	2019		2020		2021		2022
		Land Area (ha)	Production (Ton)	Producers	Land Area (ha)	Production (Ton)	Producers	Land Area (ha)	Production (Ton)
1	DKI Jakarta	8.4	4.2	0	20.15	10.075	0	20.15	10.075
2	West Java	147.53	865,257	117	138.8	928,397	131	140.87	1,091,040
3	Central Java	53.073	955,765	513	257.688	507,390	1,077	254.527	203,636
4	Special Region of Yogyakarta	2.305	94.48	17	2.105	166.1	12	1.78	140,942
5	East Java	80.588	1,828,439	237	75.087	639,362	306	42.357	1,188,106

Scheme 2022. Table 4.2 Respondent Characteristics 1.A. Mentors (2), Internal Data.

Table 7. Respondent Mentors Characteristics.

City/Regency	Number	Province	Number	Age (years)	Number	Education Level	Number
Sukabumi Regency	4	West Java	4	25-35	-	Elementary School	-
				36-46	3	Junior High School	-
				47-57	1	Senior High School	-
				58-70	-	Higher Education	4
Total	4	4	4	4	4	4	4
Cianjur Regency	3	West Java	3	25-35	-	Elementary School	-
				36-46	2	Junior High School	-
				47-57	-	Senior High School	1
				58-70	-	Higher Education	1
Total	3	3	3	3	3	3	3
Tasikmalaya Regency	2	West Java	3	25-35	1	Elementary School	-
				36-46	2	Junior High School	-
				47-57	-	Senior High School	2
				58-70	-	Higher Education	1
Total	3	3	3	3	3	3	3
Garut Regency	3	West Java	3	25-35	-	Elementary School	-
				36-46	3	Junior High School	-
				47-57	-	Senior High School	2
				58-70	-	Higher Education	1
Total	3	3	3	3	3	3	3
Bogor Regency	2	West Java	2	25-35	-	Elementary School	-
				36-46	1	Junior High School	-
				47-57	1	Senior High School	-
				58-70	-	Higher Education	2
Total	2	2	2	2	2	2	2
Depok	2	West Java	2	25-35	1	Elementary School	-
				36-46	1	Junior High School	-
				47-57	-	Senior High School	-
				58-70	-	Higher Education	-
Total	2	2	2	2	2	2	2
Bandung Barat Regency	3	West Java	3	25-35	1	Elementary School	-
				36-46	2	Junior High School	-
				47-57	-	Senior High School	-
				58-70	-	Higher Education	3
Total	3	3	3	3	3	3	3
Ciamis Regency	4	West Java	4	25-35	3	Elementary School	-
				36-46	-	Junior High School	-
				47-57	1	Senior High School	1
				58-70	-	Higher Education	3
Total	4	4	4	4	4	4	4
Total (West Java)	23	23	23	23	23	23	23
Outside West Java	21	21	21	21	21	21	21

1.B. Mentors (20), Assisted Data.

Table 7. Respondent Mentors Characteristics.

City/Regency	Number	Province	Number	Age (years)	Number	Education Level	Number
Sukabumi Regency	4	West Java	4	25-35	-	Elementary School	-
				36-46	3	Junior High School	-
				47-57	1	Senior High School	-
				58-70	-	Higher Education	4
Total	4	4	4	4	4	4	4
Cianjur Regency	3	West Java	3	25-35	-	Elementary School	-
				36-46	2	Junior High School	-
				47-57	-	Senior High School	1
				58-70	-	Higher Education	1
Total	3	3	3	3	3	3	3
Tasikmalaya Regency	2	West Java	3	25-35	1	Elementary School	-
				36-46	2	Junior High School	-
				47-57	-	Senior High School	2
				58-70	-	Higher Education	1
Total	3	3	3	3	3	3	3
Garut Regency	3	West Java	3	25-35	-	Elementary School	-
				36-46	3	Junior High School	-
				47-57	-	Senior High School	2
				58-70	-	Higher Education	1
Total	3	3	3	3	3	3	3
Bogor Regency	2	West Java	2	25-35	-	Elementary School	-
				36-46	1	Junior High School	-
				47-57	1	Senior High School	-
				58-70	-	Higher Education	2
Total	2	2	2	2	2	2	2
Depok	2	West Java	2	25-35	1	Elementary School	-
				36-46	1	Junior High School	-
				47-57	-	Senior High School	-
				58-70	-	Higher Education	-
Total	2	2	2	2	2	2	2
Bandung Barat Regency	3	West Java	3	25-35	1	Elementary School	-
				36-46	2	Junior High School	-
				47-57	-	Senior High School	-
				58-70	-	Higher Education	3
Total	3	3	3	3	3	3	3
Ciamis Regency	4	West Java	4	25-35	3	Elementary School	-
				36-46	-	Junior High School	-
				47-57	1	Senior High School	1
				58-70	-	Higher Education	3
Total	4	4	4	4	4	4	4
Total (West Java)	23	23	23	23	23	23	23
Outside West Java	21	21	21	21	21	21	21

1.B. Mentors (20), Assisted Data.

Table 8. Responden Mentor Characteristik.

2A. Business Actors (42) (26 from Government Agencies and AOI), Internal Data.

Table 8. Responden Mentor Characteristik.

2A. Business Actors (42) (26 from Government Agencies and AOI), Internal Data.

Table 9. Responden Business Actors Characteristic.

City/Regency	Number	Province	Number	Age (years)	Number	Education Level	Number
Sukabumi Regency	3	West Java	4	25-35	-	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	-
	-	-	-	58-70	-	Higher Education (PT)	3
Total	3	3	3	3	3	3	3
Cianjur Regency	6	West Java	6	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	3
Total	6	6	6	6	6	6	6
Tasikmalaya Regency	5	West Java	5	25-35	1	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	2
Total	5	5	5	5	5	5	5
Bogor Regency	6	West Java	6	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	3
	-	-	-	58-70	1	Higher Education (PT)	3
Total	6	6	6	6	6	6	6
Depok	5	West Java	5	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	1	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	2
Total	5	5	5	5	5	5	5
Garut Regency	5	West Java	5	25-35	1	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	1
	-	-	-	47-57	-	Senior High School (SMA)	2
	-	-	-	58-70	1	Higher Education (PT)	2
Total	5	5	5	5	5	5	5
West Bandung Regency	6	West Java	6	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	3
Total	6	6	6	6	6	6	6
Ciamis Regency	5	West Java	5	25-35	1	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	2
	-	-	-	58-70	-	Higher Education (PT)	3
Total	5	5	5	5	5	5	5
Grand Total	42	42	42	42	42	42	42
Outside West Java	21	21	21	21	21	21	21

2.B Business Actors, Business Capability Data.

Table 9. Responden Business Actors Characteristic.

City/Regency	Number	Province	Number	Age (years)	Number	Education Level	Number
Sukabumi Regency	3	West Java	4	25-35	-	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	-
	-	-	-	58-70	-	Higher Education (PT)	3
Total	3	3	3	3	3	3	3
Cianjur Regency	6	West Java	6	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	3
Total	6	6	6	6	6	6	6
Tasikmalaya Regency	5	West Java	5	25-35	1	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	2
Total	5	5	5	5	5	5	5
Bogor Regency	6	West Java	6	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	2	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	3
	-	-	-	58-70	1	Higher Education (PT)	3
Total	6	6	6	6	6	6	6
Depok	5	West Java	5	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	1	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	2
Total	5	5	5	5	5	5	5
Garut Regency	5	West Java	5	25-35	1	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	1
	-	-	-	47-57	-	Senior High School (SMA)	2
	-	-	-	58-70	1	Higher Education (PT)	2
Total	5	5	5	5	5	5	5
West Bandung Regency	6	West Java	6	25-35	2	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	3
	-	-	-	58-70	-	Higher Education (PT)	3
Total	6	6	6	6	6	6	6
Ciamis Regency	5	West Java	5	25-35	1	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	1	Senior High School (SMA)	2
	-	-	-	58-70	-	Higher Education (PT)	3
Total	5	5	5	5	5	5	5
Grand Total	42	42	42	42	42	42	42
Outside West Java	21	21	21	21	21	21	21

2.B Business Actors, Business Capability Data.

Table 10. Responden Business Actors Characteristic.

Business Duration (years)	Number	Business Category	Number	Number of Members (people)	Number	Land Area (ha)	Number	Vegetable Production (tons)	Number	Revenue (Rp/month)	Number
1-5	12	Individual	7	1-10	5	1-2	14	0.25-0.5	25	<5,000,000	15
6-10
>10	15	Group	35	1-30	37	3-4		>0.5	20	>5,000,000	27
Total	42		42		42		42		42		42

Table 10. Responden Business Actors Characteristic.

Business Duration (years)	Number	Business Category	Number	Number of Members (people)	Number	Land Area (ha)	Number	Vegetable Production (tons)	Number	Revenue (Rp/month)	Number
1-5	12	Individual	7	1-10	5	1-2	14	0.25-0.5	25	<5,000,000	15
6-10
>10	15	Group	35	1-30	37	3-4		>0.5	20	>5,000,000	27
Total	42		42		42		42		42		42

Table 11. Open-Ended Question Data.

Region	Local Vegetable Types	Export Vegetable Types	Largest Demand Areas	Domestic VCC Needs	Export VCC Needs	Marketing Channels	Value of Collaboration in Community
West Java	Pakcoy, mustard greens, spinach	Horenzo, pakcoy, lettuce	Bogor, Cipanas, Bandung, Jakarta	Marketing innovation, services, pricing, products	Quality, policies, distribution system	Online, organic circle, modern market	Price standards, sustainability, standardization
Outside West Java	Broccoli, spinach, water spinach	Lettuce	Prigen, Batu Malang, Central Java	Product innovation, marketing	Policies, quality

3.Consumers of Organic and Conventional Vegetables (135 People).

Table 11. Open-Ended Question Data.

Region	Local Vegetable Types	Export Vegetable Types	Largest Demand Areas	Domestic VCC Needs	Export VCC Needs	Marketing Channels	Value of Collaboration in Community
West Java	Pakcoy, mustard greens, spinach	Horenzo, pakcoy, lettuce	Bogor, Cipanas, Bandung, Jakarta	Marketing innovation, services, pricing, products	Quality, policies, distribution system	Online, organic circle, modern market	Price standards, sustainability, standardization
Outside West Java	Broccoli, spinach, water spinach	Lettuce	Prigen, Batu Malang, Central Java	Product innovation, marketing	Policies, quality

3.Consumers of Organic and Conventional Vegetables (135 People).

Table 12. Responden Consumers Characteristic.

City/Regency	Number	Province	Number	Age (years)	Number	Education Level	Number
Sukabumi Regency	10	West Java	10	25-35	7	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	-	Senior High School (SMA)	-
	-	-	-	58-70	-	Higher Education (PT)	10
Total	10	10	10	10	10	10	10
Cianjur Regency	15	West Java	15	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	8	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	10
	-	-	-	58-70	-	Higher Education (PT)	5
Total	15	15	15	15	15	15	15
Tasikmalaya Regency	15	West Java	15	25-35	8	Elementary School (SD)	-
	-	-	-	36-46	5	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	8
	-	-	-	58-70	-	Higher Education (PT)	7
Total	15	15	15	15	15	15	15
Bogor Regency	20	West Java	20	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	7	Junior High School (SMP)	-
	-	-	-	47-57	5	Senior High School (SMA)	10
	-	-	-	58-70	3	Higher Education (PT)	10
Total	20	20	20	20	20	20	20
Depok	25	West Java	25	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	8	Junior High School (SMP)	-
	-	-	-	47-57	7	Senior High School (SMA)	12
	-	-	-	58-70	5	Higher Education (PT)	13
Total	25	25	25	25	25	25	25
Garut Regency	15	West Java	15	25-35	4	Elementary School (SD)	-
	-	-	-	36-46	6	Junior High School (SMP)	2
	-	-	-	47-57	1	Senior High School (SMA)	8
	-	-	-	58-70	4	Higher Education (PT)	5
Total	15	15	15	15	15	15	15
West Bandung Regency	25	West Java	25	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	8	Junior High School (SMP)	-
	-	-	-	47-57	7	Senior High School (SMA)	3
	-	-	-	58-70	5	Higher Education (PT)	3
Total	25	25	25	25	25	25	25
Ciamis Regency	10	West Java	10	25-35	4	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	3	Senior High School (SMA)	4
	-	-	-	58-70	-	Higher Education (PT)	6
Total	10	10	10	10	10	10	10
Total	135	135	135	135	135	135	135
Outside West Java	33	33	33	33	33	33	33

Table 12. Responden Consumers Characteristic.

City/Regency	Number	Province	Number	Age (years)	Number	Education Level	Number
Sukabumi Regency	10	West Java	10	25-35	7	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	-	Senior High School (SMA)	-
	-	-	-	58-70	-	Higher Education (PT)	10
Total	10	10	10	10	10	10	10
Cianjur Regency	15	West Java	15	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	8	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	10
	-	-	-	58-70	-	Higher Education (PT)	5
Total	15	15	15	15	15	15	15
Tasikmalaya Regency	15	West Java	15	25-35	8	Elementary School (SD)	-
	-	-	-	36-46	5	Junior High School (SMP)	-
	-	-	-	47-57	2	Senior High School (SMA)	8
	-	-	-	58-70	-	Higher Education (PT)	7
Total	15	15	15	15	15	15	15
Bogor Regency	20	West Java	20	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	7	Junior High School (SMP)	-
	-	-	-	47-57	5	Senior High School (SMA)	10
	-	-	-	58-70	3	Higher Education (PT)	10
Total	20	20	20	20	20	20	20
Depok	25	West Java	25	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	8	Junior High School (SMP)	-
	-	-	-	47-57	7	Senior High School (SMA)	12
	-	-	-	58-70	5	Higher Education (PT)	13
Total	25	25	25	25	25	25	25
Garut Regency	15	West Java	15	25-35	4	Elementary School (SD)	-
	-	-	-	36-46	6	Junior High School (SMP)	2
	-	-	-	47-57	1	Senior High School (SMA)	8
	-	-	-	58-70	4	Higher Education (PT)	5
Total	15	15	15	15	15	15	15
West Bandung Regency	25	West Java	25	25-35	5	Elementary School (SD)	-
	-	-	-	36-46	8	Junior High School (SMP)	-
	-	-	-	47-57	7	Senior High School (SMA)	3
	-	-	-	58-70	5	Higher Education (PT)	3
Total	25	25	25	25	25	25	25
Ciamis Regency	10	West Java	10	25-35	4	Elementary School (SD)	-
	-	-	-	36-46	3	Junior High School (SMP)	-
	-	-	-	47-57	3	Senior High School (SMA)	4
	-	-	-	58-70	-	Higher Education (PT)	6
Total	10	10	10	10	10	10	10
Total	135	135	135	135	135	135	135
Outside West Java	33	33	33	33	33	33	33

Table 13. Respondent Characteristics of Organic and Conventional Vegetables (200 respondents, internal data).

Region	Income (IDR)	Number	%	Age (years)	Number	%	Education Level	Number	%
West Java	< 2,000,000	30	15	25-35	66	33	Elementary School (SD)	-	-
	2.5 – 4,000,000	57	28.5	36-46	75	37.5	Junior High School (SMP)	3	1.5
	4.0 – 7,000,000	78	39	47-57	40	20	Senior High School (SMA)	106	53
	> 7,500,000	35	17.5	58-70	19	9.5	Higher Education (PT)	91	45.4
Total		200	100		200	100		200	100

Table 13. Respondent Characteristics of Organic and Conventional Vegetables (200 respondents, internal data).

Region	Income (IDR)	Number	%	Age (years)	Number	%	Education Level	Number	%
West Java	< 2,000,000	30	15	25-35	66	33	Elementary School (SD)	-	-
	2.5 – 4,000,000	57	28.5	36-46	75	37.5	Junior High School (SMP)	3	1.5
	4.0 – 7,000,000	78	39	47-57	40	20	Senior High School (SMA)	106	53
	> 7,500,000	35	17.5	58-70	19	9.5	Higher Education (PT)	91	45.4
Total		200	100		200	100		200	100

Table 14. Organic Vegetable Consumption and Online Purchase Behavior.

Region	Consumes/Purchases Organic Vegetables	Number	%	Aware of Organic Vegetable Platforms	Number	%	Buys Organic Vegetables Online	Number	%
West Java	Yes	179	89.5	Lingkar Organik	63	31.5	Sometimes	71	35.5
	No	18	9	Kecipir	49	24.5	Ever	45	22.5
	Sometimes	3	1.5	edas	57	28.5	Never	84	42
				Sesa.id	33	16.5
Total		200	100		200	100		200	100

Table 14. Organic Vegetable Consumption and Online Purchase Behavior.

Region	Consumes/Purchases Organic Vegetables	Number	%	Aware of Organic Vegetable Platforms	Number	%	Buys Organic Vegetables Online	Number	%
West Java	Yes	179	89.5	Lingkar Organik	63	31.5	Sometimes	71	35.5
	No	18	9	Kecipir	49	24.5	Ever	45	22.5
	Sometimes	3	1.5	edas	57	28.5	Never	84	42
				Sesa.id	33	16.5
Total		200	100		200	100		200	100

4.2. Factor Analysis Results (Identification of Problem 1) - CFA

4.2.1. Demand Factor Analysis

Analysis of the Most Determining Factors for Organic Vegetable Demand

Goodness of Fit Test for the Factor Analysis Model Affecting Organic Vegetable Demand

Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test

The Kaiser-Meyer-Olkin (KMO) test is used as an initial test to determine whether the available data can be broken down into a set of factors (Santoso, 2015). The KMO test is also utilized to assess the model fit or goodness of fit of the factor analysis model for the demand of organic vegetables, using the chi-square statistical test. Factor analysis model criteria:

• If the p-value or significance level (sig) is < 0.05, then the factor analysis model is fit or good.
• If the p-value or significance level (sig) is > 0.05, then the factor analysis model is not fit or not good, and data transformation is required.

The results of the KMO and Bartlett’s Test in Table 1 indicate that the KMO value for the factor analysis model affecting the demand for local and export organic vegetables is 0.821. The Bartlett’s Test produced a Chi-Square value of 658.91, with a significance value (sig) of 0.000, which is well below 0.05 (0.00 < 0.05). This means that the available variables and samples are suitable for factor analysis.Therefore, the factor analysis model for the demand for local and export organic vegetables is appropriate, allowing further analysis, including Factor Extraction and Rotation.

Table 15. KMO Measurement Results and Bartlett’s Test.

Table 15. KMO Measurement Results and Bartlett’s Test.

Analysis of Eigenvalues, Factor Extraction, and Rotation

After conducting the goodness of fit test and confirming that the factor analysis model influencing the demand for local and export organic vegetables is valid, the next step is to perform an analysis of eigenvalues, factor extraction, and rotation.

Eigenvalues indicate the relative importance of each factor in explaining the variance of all analyzed variables. This study examines 12 variables suspected to influence the demand for local and export organic vegetables: Awareness (X₁), Behavior (X₂), Lifestyle (X₃), Knowledge (X₄), Availability (X₅), Price (X₆), Trust in Product (X₇), Quality Assurance (X₈), Community Certification (X₉), LSO Certification (X₁₀), Innovation Value (X₁₁), and Product Quality (X₁₂). The eigenvalues show the relative importance of each factor in explaining the variance among these 12 variables.

Based on Table 2, only three factors are formed. With one factor, the eigenvalue is above 1. With two factors, the eigenvalue remains above 1. With three factors, the eigenvalue is still above 1 (1.102). However, for four factors, the eigenvalue falls below 1 (0.980), meaning the factoring process must stop at three factors.

Thus, based on the eigenvalue analysis, it can be concluded that from the 12 analyzed variables, the demand for local and export organic vegetables can be summarized into only three main influencing factors.

Table 16. Results of Eigenvalues, Factor Extraction, and Rotation.

Table 16. Results of Eigenvalues, Factor Extraction, and Rotation.

Factor Loading and Validity Test of the Factor Analysis Model Influencing the Demand for Local and Export Organic Vegetables

The Component Matrix from the rotation process (Rotated Component Matrix) is presented in Table 3. This table clearly shows the final distribution of variables in the factor analysis. Each variable is assigned to a factor where it has the highest factor loading value, which must be at least 0.5 or less than -0.5. If a variable does not reach a factor loading of ±0.5 for any of the three factors, it is considered invalid.

Table 17. Rotated Component Matrix: Factor Loading Values of 12 Variables for the Three Formed Components.

Table 17. Rotated Component Matrix: Factor Loading Values of 12 Variables for the Three Formed Components.

Based on Table 3, out of the 12 variables, one variable was found to be invalid, namely X₁₁ (Innovation Value). The remaining 11 variables have been successfully reduced into three factors, as follows:

• Factor 1 consists of Awareness (X₁), Behavior (X₂), Lifestyle (X₃), Knowledge (X₄), and Trust in Product (X₇).
• Factor 2 consists of Price (X₆), Quality Assurance (X₈), Community Certification (X₉), and LSO Certification (X₁₀).
• Factor 3 consists of Availability (X₅) and Product Quality (X₁₂).

Naming the Factors

• Factor 1: Consumer Internal Factors
  Since all five variables in Factor 1 have positive factor loading values, this factor can be named Consumer Internal Factors. This means that the increasing demand for local and export organic vegetables is influenced by consumer awareness of the importance of organic vegetable consumption, healthier consumer behavior, a healthier lifestyle, greater consumer knowledge of healthy living, and higher trust in organic vegetable products.
• Factor 2: Product Exclusivity Factors
  Since all four variables in Factor 2 also have positive factor loading values, this factor can be named Product Exclusivity Factors. This indicates that the higher demand for local and export organic vegetables is driven by higher product prices, clearer quality assurance, the presence of local and international organic community certifications, and LSO certification.
• Factor 3: Strength Factors
  Since Availability (X₅) has a positive factor loading value, this suggests that the increased availability of organic vegetables in the market contributes to higher demand. However, Product Quality (X₁₂) has a negative factor loading value, indicating that as demand for organic vegetables increases, product quality tends to decrease. This may be due to consumer perception that if vegetables appear too perfect, they might have been treated with chemical pesticides, making them less desirable compared to conventional vegetables.

4.2.2. Factor Analysis of Supply

Analysis of the Key Factors Determining the Supply of Organic Vegetables

1. Goodness of Fit Test for the Factor Analysis Model Affecting the Supply of Local and Export Organic Vegetables

1.1 Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test

The Kaiser-Meyer-Olkin (KMO) test is used as a preliminary test to determine whether the available data can be broken down into a number of factors (Santoso, 2015). The KMO test is also used to assess the goodness of fit of the factor analysis model for formal and informal sector rice seed using chi-square statistical testing.

The criteria for factor analysis modeling are as follows:

• If the p-value (significance level) is less than 0.05, the factor analysis model is considered fit or good.
• If the p-value (significance level) is greater than 0.05, the factor analysis model is considered not fit or not good, and data transformation is required.

The KMO and Bartlett’s Test results in Table 1 show that the KMO value for the factor analysis model affecting the supply of local and export organic vegetables is 0.672. Additionally, the Bartlett’s Test result shows a Chi-Square value of 326.32 with a significance level (sig) of 0.000, which is well below 0.05 (0.00 < 0.05). This indicates that the variables and sample used are suitable for factor analysis.

Therefore, the factor analysis model for the supply of local and export organic vegetables is valid, allowing for further analysis, including Factor Extraction and Rotation.

Table 18. KMO and Bartlett’s Test Results.

Table 18. KMO and Bartlett’s Test Results.

Eigenvalues Analysis, Factor Extraction, and Rotation

After conducting the goodness of fit test and confirming that the factor analysis model influencing the supply of local organic vegetables is valid, the next step involves Eigenvalues analysis, factor extraction, and factor rotation.

Eigenvalues represent the relative importance of each factor in explaining the variance of all analyzed variables. In this study, 11 variables were examined for their potential impact on the supply organic vegetables, including:Production output (X₁), Climate change (X₂), Government policy for the domestic market (X₃), Government policy for the export market (X₄), Quality assurance (X₅), Information technology (X₆), Capital access (X₈), Brand (X₉), Packaging (X₁₀), Community (X₁₁). Eigenvalues indicate the relative importance of each factor in calculating the variance of these 11 analyzed variables.

Based on Table 2, only three factors were identified:

• With one factor, the eigenvalue remains above 1.
• With two factors, the eigenvalues are still above 1.
• With three factors, the eigenvalue remains above 1, specifically 1.296.
• However, with four factors, the eigenvalue drops below 1 (0.977), meaning that the factoring process must stop at three factors.

Thus, based on the Eigenvalues analysis, it can be concluded that the 11 analyzed variables can be summarized into three main factors that influence the supply organic vegetables.

Table 19. Eigenvalues, Factor Extraction, and Rotation Results.

Table 19. Eigenvalues, Factor Extraction, and Rotation Results.

Factor Extraction and Rotation Process

The results from the Initial Eigenvalues serve as the foundation for the factor extraction process. The extracted factors, as shown in Table 2, retain the same values as the Initial Eigenvalues but only for the three identified factors. Next, the rotation process is applied to enhance clarity in categorizing variables into Factor 1, Factor 2, or Factor 3. The final output of this rotation process represents the conclusive results of the factor analysis.

2. Factor Loading and Validity Test of the Factor Analysis Model for Organic Vegetable Supply

The Component Matrix obtained after the rotation process (Rotated Component Matrix), presented in Table 3, provides a clearer and more structured distribution of variables in the factor analysis. Each variable is assigned to the factor where it has the highest factor loading value, with a minimum threshold of 0.5 or a negative value lower than -0.5. If a variable fails to reach a factor loading value of ±0.5 for any of the three factors, it is considered invalid.

Table 20. Rotated Component Matrix: Factor Loading Values of 11 Variables in the Three Identified Factors.

Table 20. Rotated Component Matrix: Factor Loading Values of 11 Variables in the Three Identified Factors.

Based on Table 3, out of the 11 variables, one variable is invalid, namely variable X1 (production output). Meanwhile, the remaining 10 variables have been successfully reduced into three factors:

• Factor 1 consists of Climate Change (X₂), Government Policy – Domestic Market (X₃), and Government Policy – Export Market (X₄).
• Factor 2 consists of Quality Assurance (X₅), Information Technology (X₆), Capital Access (X₇), and Brand (X₈).
• Factor 3 consists of Packaging (X₉), Community (X₁₀), and Role of Mentors (X₁₁).

Factor 1: Key Supporting Factor

Factor 1 includes Climate Change (X₂), Government Policy – Domestic Market (X₃), and Government Policy – Export Market (X₄). If named, this factor can be called the Key Supporting Factor. Since the factor loadings of these three variables are positive, this means that the increasing supply of local and export organic vegetables is influenced by the Key Supporting Factor, characterized by more stable climate conditions, government policies that support the growth of the domestic market, and policies that promote export market expansion. (Further discussion can be added here.)

Factor 2: Exclusive Factor

Factor 2 includes Quality Assurance (X₅), Information Technology (X₆), Capital Access (X₇), and Brand (X₈). If named, this factor can be called the Exclusive Factor. The positive factor loadings of these four variables indicate that the increasing supply of local and export organic vegetables is influenced by the Exclusive Factor, marked by better quality assurance, easier access to capital for farmers, improved farmer skills in utilizing information technology, and greater brand recognition of organic vegetable products. (Further discussion can be added here.)

Factor 3: Strengthening Support Factor

Factor 3 includes Packaging (X₉), Community (X₁₀), and Role of Mentors (X₁₁). If named, this factor can be called the Strengthening Support Factor. The positive factor loadings of these three variables indicate that the increasing supply of local and export organic vegetables is influenced by the Strengthening Support Factor, characterized by improved product packaging, a supportive organic farming community, and a better role of mentors in enhancing organic vegetable production. (Further discussion can be added here.)

VCC B2C Model for Organic Vegetables (Development of the DART Model)

Multivariate Outlier Test

The multivariate outlier examination was conducted using the Mahalanobis Distance (d²) statistic by testing the following hypotheses:

• H₀: The data does not contain multivariate outliers.
• H₁: The data contains multivariate outliers.

The decision criterion is to reject H₀ for any row of data if the Mahalanobis Distance (d²) exceeds the chi-square value χ²(p<0.001; k), where k is the number of manifest variables (indicators).

The test results indicate that in this Structural Equation Modeling (SEM) model, k = 47 indicators or manifest variables, yielding a chi-square value of χ²(p<0.001;47) = 82.72. Among the total 200 data points, 5 data points have Mahalanobis Distance (d²) values greater than the chi-square threshold of 82.72 (Table 1). As a result, H₀ is rejected for these data points, meaning that they contain multivariate outliers and must be removed from the analysis.

Meanwhile, the remaining data points have Mahalanobis Distance values below the chi-square threshold, indicating that they do not contain multivariate outliers. Consequently, the final dataset for further analysis consists of 195 respondent data points. A complete report of this test is provided in Appendix 1.

Table 21. List of Data Containing Multivariate Outliers.

No.	Mahalanobis Distance (d²)	Chi-square Value χ²(p<0.001;47)	Status
134	86.60316	82.72042	Outlier
136	198.005	82.72042	Outlier
137	198.005	82.72042	Outlier
138	198.005	82.72042	Outlier
161	82.85639	82.72042	Outlier

Table 21. List of Data Containing Multivariate Outliers.

No.	Mahalanobis Distance (d²)	Chi-square Value χ²(p<0.001;47)	Status
134	86.60316	82.72042	Outlier
136	198.005	82.72042	Outlier
137	198.005	82.72042	Outlier
138	198.005	82.72042	Outlier
161	82.85639	82.72042	Outlier

Multivariate Normality Test

The results of the normality test are shown in the following table (the complete results can be found in Appendix 2).

Table 22. Results of Univariate and Multivariate Normality Test.

Variabel Manifest	Univariate Normality		Multivariate Normality
	Skewness and Kurtosis		Skewness and Kurtosis
	Chi-Square	P-Value	Chi-Square	P-Value
X5	0.210	0.900	335.377	0.000
X3	1.613	0.446
X1-1	1.044	0.593
X1-2	0.119	0.942
X1-3	1.553	0.460
X1-4	3.044	0.218
X1-5	5.485	0.064
X1-6	1.255	0.534
X1-7	1.098	0.578
X1-8	0.701	0.704
X1-9	3.755	0.153
X1-10	0.738	0.691
X1-11	1.719	0.423
X1-12	2.569	0.277
X2-1	1.985	0.371
X2-2	4.192	0.123
X2-3	4.739	0.094
X2-4	4.646	0.098
X2-5	0.680	0.712
X2-6	0.931	0.628
X2-7	7.225	0.054
X2-8	0.754	0.686
X2-9	2.405	0.300
X2-10	3.973	0.137
X2-11	0.754	0.686
X4	2.404	0.301
X6	1.274	0.529
Y1-1	0.627	0.731
Y1-2	0.444	0.801
Y1-3	0.005	0.997
Y1-4	0.675	0.714
X7-1	5.572	0.062
X7-2	0.058	0.971
X7-3	0.043	0.979
Y2-1	3.044	0.218
Y2-2	2.458	0.293
Y2-3	1.104	0.576
Y2-4	0.743	0.690
Y2-5	1.862	0.394
Y2-6	0.650	0.723
Y2-7	2.705	0.259
Y2-8	5.699	0.058
Y2-9	0.062	0.970
Y2-10	1.568	0.457
Y2-11	1.230	0.541
Y2-12	0.889	0.641
Y2-13	0.682	0.711

Table 22. Results of Univariate and Multivariate Normality Test.

Variabel Manifest	Univariate Normality		Multivariate Normality
	Skewness and Kurtosis		Skewness and Kurtosis
	Chi-Square	P-Value	Chi-Square	P-Value
X5	0.210	0.900	335.377	0.000
X3	1.613	0.446
X1-1	1.044	0.593
X1-2	0.119	0.942
X1-3	1.553	0.460
X1-4	3.044	0.218
X1-5	5.485	0.064
X1-6	1.255	0.534
X1-7	1.098	0.578
X1-8	0.701	0.704
X1-9	3.755	0.153
X1-10	0.738	0.691
X1-11	1.719	0.423
X1-12	2.569	0.277
X2-1	1.985	0.371
X2-2	4.192	0.123
X2-3	4.739	0.094
X2-4	4.646	0.098
X2-5	0.680	0.712
X2-6	0.931	0.628
X2-7	7.225	0.054
X2-8	0.754	0.686
X2-9	2.405	0.300
X2-10	3.973	0.137
X2-11	0.754	0.686
X4	2.404	0.301
X6	1.274	0.529
Y1-1	0.627	0.731
Y1-2	0.444	0.801
Y1-3	0.005	0.997
Y1-4	0.675	0.714
X7-1	5.572	0.062
X7-2	0.058	0.971
X7-3	0.043	0.979
Y2-1	3.044	0.218
Y2-2	2.458	0.293
Y2-3	1.104	0.576
Y2-4	0.743	0.690
Y2-5	1.862	0.394
Y2-6	0.650	0.723
Y2-7	2.705	0.259
Y2-8	5.699	0.058
Y2-9	0.062	0.970
Y2-10	1.568	0.457
Y2-11	1.230	0.541
Y2-12	0.889	0.641
Y2-13	0.682	0.711

Based on the table above, the results of the normality test for each variable (univariate normality) show that all manifest variables have p-values for Skewness and Kurtosis greater than 0.05. This indicates that none of the manifest variables have issues with univariate normality, meaning they can all be included in the next stage of analysis since the normality assumption, which is a fundamental requirement for SEM analysis, has been met. Meanwhile, the normality test results for all variables combined (multivariate normality) indicate that the multivariate normality assumption has not been met, as the p-value for Skewness and Kurtosis is 0.000, which is still less than 0.05. This means the data still faces normality issues at the multivariate level.

In principle, if the univariate normality assumption is fulfilled, the indicators can be considered normal. However, to address the issue of multivariate normality violations, adjustments will be made in the model estimation process by applying the Satorra-Bentler Scaled χ² method. This involves incorporating the asymptotic covariance matrix input in addition to the data covariance matrix input, which will generate a Goodness of Fit value to correct for data non-normality (Yamin, 2014). Therefore, the analysis will proceed with the manifest variables that have met the univariate normality assumption.

Multicollinearity Test

The results of the multicollinearity test are shown in the table below (the complete results can be found in Appendix 3).

Table 23. Multicollinearity Test Results.

Variabel Manifest	Collinearity Statistics		Result
	Tolerance	VIF
X5	,425	2,354	Unmultikolinieritas
X3	,325	3,073	Unmultikolinieritas
X1-1	,471	2,124	Unmultikolinieritas
X1-2	,446	2,242	Unmultikolinieritas
X1-3	,369	2,711	Unmultikolinieritas
X1-4	,416	2,405	Unmultikolinieritas
X1-5	,632	1,582	Unmultikolinieritas
X1-6	,673	1,485	Unmultikolinieritas
X1-7	,556	1,799	Unmultikolinieritas
X1-8	,553	1,809	Unmultikolinieritas
X1-9	,558	1,791	Unmultikolinieritas
X1-10	,490	2,043	Unmultikolinieritas
X1-11	,662	1,510	Unmultikolinieritas
X1-12	,804	1,244	Unmultikolinieritas
X2-1	,681	1,469	Unmultikolinieritas
X2-2	,680	1,470	Unmultikolinieritas
X2-3	,515	1,943	Unmultikolinieritas
X2-4	,492	2,033	Unmultikolinieritas
X2-5	,604	1,655	Unmultikolinieritas
X2-6	,605	1,653	Unmultikolinieritas
X2-7	,699	1,431	Unmultikolinieritas
X2-8	,670	1,493	Unmultikolinieritas
X2-9	,747	1,339	Unmultikolinieritas
X2-10	,729	1,373	Unmultikolinieritas
X2-11	,776	1,289	Unmultikolinieritas
X4	,369	2,710	Unmultikolinieritas
X6	,517	1,934	Unmultikolinieritas
Y1-1	,617	1,621	Unmultikolinieritas
Y1-2	,223	4,494	Unmultikolinieritas
Y1-3	,423	2,367	Unmultikolinieritas
Y1-4	,422	2,372	Unmultikolinieritas
X7-1	,223	4,475	Unmultikolinieritas
X7-2	,323	3,092	Unmultikolinieritas
X7-3	,523	1,910	Unmultikolinieritas
Y2-1	,380	2,633	Unmultikolinieritas
Y2-2	,480	2,084	Unmultikolinieritas
Y2-3	,216	4,637	Unmultikolinieritas
Y2-4	,520	1,922	Unmultikolinieritas
Y2-5	,148	6,764	Unmultikolinieritas
Y2-6	,383	2,613	Unmultikolinieritas
Y2-7	,648	1,544	Unmultikolinieritas
Y2-8	,283	3,537	Unmultikolinieritas
Y2-9	,434	2,303	Unmultikolinieritas
Y2-10	,198	5,045	Unmultikolinieritas
Y2-11	,598	1,672	Unmultikolinieritas
Y2-12	,423	2,363	Unmultikolinieritas
Y2-13	,524	1,908	Unmultikolinieritas

Table 23. Multicollinearity Test Results.

Variabel Manifest	Collinearity Statistics		Result
	Tolerance	VIF
X5	,425	2,354	Unmultikolinieritas
X3	,325	3,073	Unmultikolinieritas
X1-1	,471	2,124	Unmultikolinieritas
X1-2	,446	2,242	Unmultikolinieritas
X1-3	,369	2,711	Unmultikolinieritas
X1-4	,416	2,405	Unmultikolinieritas
X1-5	,632	1,582	Unmultikolinieritas
X1-6	,673	1,485	Unmultikolinieritas
X1-7	,556	1,799	Unmultikolinieritas
X1-8	,553	1,809	Unmultikolinieritas
X1-9	,558	1,791	Unmultikolinieritas
X1-10	,490	2,043	Unmultikolinieritas
X1-11	,662	1,510	Unmultikolinieritas
X1-12	,804	1,244	Unmultikolinieritas
X2-1	,681	1,469	Unmultikolinieritas
X2-2	,680	1,470	Unmultikolinieritas
X2-3	,515	1,943	Unmultikolinieritas
X2-4	,492	2,033	Unmultikolinieritas
X2-5	,604	1,655	Unmultikolinieritas
X2-6	,605	1,653	Unmultikolinieritas
X2-7	,699	1,431	Unmultikolinieritas
X2-8	,670	1,493	Unmultikolinieritas
X2-9	,747	1,339	Unmultikolinieritas
X2-10	,729	1,373	Unmultikolinieritas
X2-11	,776	1,289	Unmultikolinieritas
X4	,369	2,710	Unmultikolinieritas
X6	,517	1,934	Unmultikolinieritas
Y1-1	,617	1,621	Unmultikolinieritas
Y1-2	,223	4,494	Unmultikolinieritas
Y1-3	,423	2,367	Unmultikolinieritas
Y1-4	,422	2,372	Unmultikolinieritas
X7-1	,223	4,475	Unmultikolinieritas
X7-2	,323	3,092	Unmultikolinieritas
X7-3	,523	1,910	Unmultikolinieritas
Y2-1	,380	2,633	Unmultikolinieritas
Y2-2	,480	2,084	Unmultikolinieritas
Y2-3	,216	4,637	Unmultikolinieritas
Y2-4	,520	1,922	Unmultikolinieritas
Y2-5	,148	6,764	Unmultikolinieritas
Y2-6	,383	2,613	Unmultikolinieritas
Y2-7	,648	1,544	Unmultikolinieritas
Y2-8	,283	3,537	Unmultikolinieritas
Y2-9	,434	2,303	Unmultikolinieritas
Y2-10	,198	5,045	Unmultikolinieritas
Y2-11	,598	1,672	Unmultikolinieritas
Y2-12	,423	2,363	Unmultikolinieritas
Y2-13	,524	1,908	Unmultikolinieritas

The results of the multicollinearity test indicate that the tolerance values for all manifest variables are within an acceptable range. This is evidenced by the VIF values for all manifest variables, none of which exceed 5. Therefore, it can be concluded that there is no excessively high correlation between any manifest variable/indicator and another. As a result, all manifest variables/indicators are free from multicollinearity, allowing them to proceed to the next stage of testing

After all the initial assumption tests have been completed, the next step is to perform the SEM model estimation. The initial estimation results using LISREL software version 8.80 produced the Standardized Loading Factor (SLF) values for each exogenous and endogenous indicator/variable, as presented in the following Table 24:

Table 24. the Standardized Loading Factor (SLF) Values for Each Exogenous and Endogenous Indicator/Variable.

Table 24. the Standardized Loading Factor (SLF) Values for Each Exogenous and Endogenous Indicator/Variable.

Based on the table above, there are manifest variables that still have SLF values less than 0.50 and insignificant t-values. These include the manifest variables: LSO Certification (X_1-10), Capital Access (X_2-7), Social (X_7-2), Attractive Website (Y_2-1), Informative (Y_2-4), and System Control and Evaluation (Y_2-10). These six manifest variables were subsequently removed, and a re-estimation was conducted. The results of the re-estimation after removing these variables are presented below (the original LISREL output can be found in Appendix 4).

Table 25. the Standardized Loading Factor (SLF) Values for Each Exogenous and Endogenous Indicator/Variable.

Table 25. the Standardized Loading Factor (SLF) Values for Each Exogenous and Endogenous Indicator/Variable.

Based on the table above, all manifest variables have SLF values ≥ 0.50, error variance values ≥ 0, and significant t-values. Therefore, the next test can be carried out, namely the reliability test of the measurement model.

Reliability Analysis of the Measurement Model – Confirmatory Factor Analysis (CFA) and Structural Model

This analysis is conducted by calculating the values of Construct Reliability (CR) and Average Variance Extracted (AVE) from the Standardized Loading Factors (SLF or λᵢ) and the measurement errors (eᵢ) using the following formulas: Measurement errors (eᵢ) using the following formulas:

$$CR={\displaystyle \frac{{\left(\sum {\lambda }_i\right)}^2}{{\left(\sum {\lambda }_i\right)}^2+\sum e_i}}$$

$$AVE={\displaystyle \frac{\sum {\left({\lambda }_i\right)}^2}{\sum {\left({\lambda }_i\right)}^2+\sum e_i}}$$

The results of the reliability analysis of the model are presented in the following table.

Table 26. The Reliability Analysis of The Model.

Causal Relationship	SLF (${\lambda }_i$)	Error	AVE	CR
ξ5 ← X5	0.56	0.03	0.91	0.91
ξ3 ← X3	0.59	0.11	0.76	0.76
ξ1 ← X1-1	0.69	0.52	0.50	0.91
ξ1 ← X1-2	0.71	0.49
ξ1 ← X1-3	0.80	0.37
ξ1 ← X1-4	0.74	0.46
ξ1 ← X1-5	0.77	0.40
ξ1 ← Y1-6	0.73	0.47
ξ1 ← X1-7	0.55	0.69
ξ1 ← X1-8	0.57	0.68
ξ1 ← X1-9	0.66	0.56
ξ1 ← X1-11	0.70	0.51
ξ1 ← X1-12	0.71	0.50
ξ2 ← X2-1	0.73	0.47	0.50	0.90
ξ2 ← X2-2	0.67	0.55
ξ2 ← X2-3	0.50	0.75
ξ2 ← X2-4	0.81	0.34
ξ2 ← X2-5	0.73	0.47
ξ2 ← Y2-6	0.73	0.47
ξ2 ← X2-8	0.83	0.31
ξ2 ← X2-9	0.72	0.49
ξ2← X2-10	0.58	0.66
ξ2 ← X2-11	0.60	0.64
ξ4 ← X4	0.57	0.73	0.76	0.76
ξ6 ← X6	0.52	0.73	0.90	0.90
ξ7 ← X71	0.50	0.38	0.56	0.72
ξ7 ← X73	0.50	0.29
η1 ← Y2=3	0.94	0.11	0.84	0.95
η1 ← Y2-3	0.89	0.58
η1 ← Y2-3	0.92	0.50
η1 ← Y2-4	0.91	0.14
η2 ← Y2-2	0.89	0.21	0.66	0.95
η2 ← Y2-3	0.90	0.18
η2 ← Y2-5	0.78	0.38
η2 ← Y2-6	0.80	0.36
η2 ← Y2-7	0.74	0.45
η2 ← Y2-8	0.73	0.46
η2 ← Y2-9	0.76	0.42
η2 ← Y2-11	0.82	0.33
η2 ← Y2-12	0.87	0.24
η2 ← Y2-13	0.80	0.35

Table 26. The Reliability Analysis of The Model.

Causal Relationship	SLF (${\lambda }_i$)	Error	AVE	CR
ξ5 ← X5	0.56	0.03	0.91	0.91
ξ3 ← X3	0.59	0.11	0.76	0.76
ξ1 ← X1-1	0.69	0.52	0.50	0.91
ξ1 ← X1-2	0.71	0.49
ξ1 ← X1-3	0.80	0.37
ξ1 ← X1-4	0.74	0.46
ξ1 ← X1-5	0.77	0.40
ξ1 ← Y1-6	0.73	0.47
ξ1 ← X1-7	0.55	0.69
ξ1 ← X1-8	0.57	0.68
ξ1 ← X1-9	0.66	0.56
ξ1 ← X1-11	0.70	0.51
ξ1 ← X1-12	0.71	0.50
ξ2 ← X2-1	0.73	0.47	0.50	0.90
ξ2 ← X2-2	0.67	0.55
ξ2 ← X2-3	0.50	0.75
ξ2 ← X2-4	0.81	0.34
ξ2 ← X2-5	0.73	0.47
ξ2 ← Y2-6	0.73	0.47
ξ2 ← X2-8	0.83	0.31
ξ2 ← X2-9	0.72	0.49
ξ2← X2-10	0.58	0.66
ξ2 ← X2-11	0.60	0.64
ξ4 ← X4	0.57	0.73	0.76	0.76
ξ6 ← X6	0.52	0.73	0.90	0.90
ξ7 ← X71	0.50	0.38	0.56	0.72
ξ7 ← X73	0.50	0.29
η1 ← Y2=3	0.94	0.11	0.84	0.95
η1 ← Y2-3	0.89	0.58
η1 ← Y2-3	0.92	0.50
η1 ← Y2-4	0.91	0.14
η2 ← Y2-2	0.89	0.21	0.66	0.95
η2 ← Y2-3	0.90	0.18
η2 ← Y2-5	0.78	0.38
η2 ← Y2-6	0.80	0.36
η2 ← Y2-7	0.74	0.45
η2 ← Y2-8	0.73	0.46
η2 ← Y2-9	0.76	0.42
η2 ← Y2-11	0.82	0.33
η2 ← Y2-12	0.87	0.24
η2 ← Y2-13	0.80	0.35

According to Yamin (2014), the criteria for good reliability are having an Average Variance Extracted (AVE) value ≥ 0.50 and a Construct Reliability (CR) value ≥ 0.70. Based on the table above, it is evident that the Construct Reliability and Average Variance Extracted values of the latent variables Risk/Benefit (ξ5), Dialogue (ξ3), Demand (ξ1), Supply (ξ2), Access (ξ4), Transparency (ξ6), Sustainability (ξ7), VCC / Value Co-Creation (η1), and Marketing Platform (η2) have met these criteria.

This indicates that the reliability of the measurement model is good and that all latent variables in this model are supported by the data. The next step is to estimate the model and conduct the subsequent test, which is the Goodness of Fit test.

Table 27. Results of Measurement Model Goodness of Fit Test.

Goodness of Fit Measure	Fit Criteria	Estimate	Description
Chi-Square Statistic	Smaller value	466.113	Marginal Fit
P-value	P > 0.05	0.000	Marginal Fit
Non-Centrality Parameter (NCP)	Smaller value	239.113	Good Fit
RMSEA (Root Mean Square Error of Approximation)	0.05 ≤ RMSEA ≤ 0.08 is good fit; RMSEA < 0.05 is close fit; RMSEA > 0.08 is marginal fit	0.0728	Good Fit
Expected Cross-Validation Index (ECVI)	Closer ECVI value to the saturated model than the independence model indicates good fit	2.835	Good Fit
ECVI for Saturated Model		2.774
ECVI for Independence Model		7.662
Independence AIC	AIC value closer to the saturated AIC than the independence AIC indicates good fit	1524.807	Good Fit
Model AIC		564.113
Saturated AIC		552.000
Independence CAIC	CAIC value closer to the saturated CAIC than the independence CAIC indicates good fit	1623.669	Good Fit
Model CAIC		774.731
Saturated CAIC		1738.336
Normed Fit Index (NFI)	Range 0–1; higher is better. NFI ≥ 0.90 = good fit; 0.80 ≤ NFI < 0.90 = marginal fit	0.803	Marginal Fit
Non-Normed Fit Index (NNFI)	Range 0–1; higher is better. NNFI ≥ 0.90 = good fit; 0.80 ≤ NNFI < 0.90 = marginal fit	0.907	Good Fit
Comparative Fit Index (CFI)	Range 0–1; higher is better. CFI ≥ 0.90 = good fit; 0.80 ≤ CFI < 0.90 = marginal fit	0.927	Good Fit
Incremental Fit Index (IFI)	Range 0–1; higher is better. IFI ≥ 0.90 = good fit; 0.80 ≤ IFI < 0.90 = marginal fit	0.930	Good Fit
Relative Fit Index (RFI)	Range 0–1; higher is better. RFI ≥ 0.90 = good fit; 0.80 ≤ RFI < 0.90 = marginal fit	0.769	Marginal Fit
Standardized RMR	≤ 0.05 = good fit; 0.05 < RMR ≤ 0.09 = close fit; RMR > 0.09 = marginal fit	0.0804	Close Fit
Goodness of Fit Index (GFI)	Range 0–1; higher is better. GFI ≥ 0.90 = good fit; 0.80 ≤ GFI < 0.90 = marginal fit	0.931	Good Fit
Adjusted Goodness of Fit Index (AGFI)	Range 0–1; AGFI ≥ 0.90 = good fit; 0.80 ≤ AGFI < 0.90 = close fit; AGFI < 0.80 = marginal fit	0.894	Marginal Fit

Table 27. Results of Measurement Model Goodness of Fit Test.

Goodness of Fit Measure	Fit Criteria	Estimate	Description
Chi-Square Statistic	Smaller value	466.113	Marginal Fit
P-value	P > 0.05	0.000	Marginal Fit
Non-Centrality Parameter (NCP)	Smaller value	239.113	Good Fit
RMSEA (Root Mean Square Error of Approximation)	0.05 ≤ RMSEA ≤ 0.08 is good fit; RMSEA < 0.05 is close fit; RMSEA > 0.08 is marginal fit	0.0728	Good Fit
Expected Cross-Validation Index (ECVI)	Closer ECVI value to the saturated model than the independence model indicates good fit	2.835	Good Fit
ECVI for Saturated Model		2.774
ECVI for Independence Model		7.662
Independence AIC	AIC value closer to the saturated AIC than the independence AIC indicates good fit	1524.807	Good Fit
Model AIC		564.113
Saturated AIC		552.000
Independence CAIC	CAIC value closer to the saturated CAIC than the independence CAIC indicates good fit	1623.669	Good Fit
Model CAIC		774.731
Saturated CAIC		1738.336
Normed Fit Index (NFI)	Range 0–1; higher is better. NFI ≥ 0.90 = good fit; 0.80 ≤ NFI < 0.90 = marginal fit	0.803	Marginal Fit
Non-Normed Fit Index (NNFI)	Range 0–1; higher is better. NNFI ≥ 0.90 = good fit; 0.80 ≤ NNFI < 0.90 = marginal fit	0.907	Good Fit
Comparative Fit Index (CFI)	Range 0–1; higher is better. CFI ≥ 0.90 = good fit; 0.80 ≤ CFI < 0.90 = marginal fit	0.927	Good Fit
Incremental Fit Index (IFI)	Range 0–1; higher is better. IFI ≥ 0.90 = good fit; 0.80 ≤ IFI < 0.90 = marginal fit	0.930	Good Fit
Relative Fit Index (RFI)	Range 0–1; higher is better. RFI ≥ 0.90 = good fit; 0.80 ≤ RFI < 0.90 = marginal fit	0.769	Marginal Fit
Standardized RMR	≤ 0.05 = good fit; 0.05 < RMR ≤ 0.09 = close fit; RMR > 0.09 = marginal fit	0.0804	Close Fit
Goodness of Fit Index (GFI)	Range 0–1; higher is better. GFI ≥ 0.90 = good fit; 0.80 ≤ GFI < 0.90 = marginal fit	0.931	Good Fit
Adjusted Goodness of Fit Index (AGFI)	Range 0–1; AGFI ≥ 0.90 = good fit; 0.80 ≤ AGFI < 0.90 = close fit; AGFI < 0.80 = marginal fit	0.894	Marginal Fit

Based on the table above, the analysis results indicate that almost all absolute Goodness of Fit (GoF) measures such as RMSEA, NCP, ECVI, AIC, CAIC, NNFI, CFI, IFI, and GFI show good fit values. Only one absolute GoF measure, namely the Chi-Square statistic and its p-value, indicates a marginal fit. Furthermore, the incremental GoF measures also show good fit values, such as NNFI, CFI, and IFI. Thus, it can be concluded that the overall fit of the measurement model is quite good (Good Fit).

Interpretation of SEM Model Analysis Results

After fulfilling all assumption tests, the output of the LISREL Structural Equation Model (SEM) analysis is presented as follows:

Figure 6. Significance Level of Path Coefficients from SEM Model Estimation Results (Lisrel).

Figure 6. Significance Level of Path Coefficients from SEM Model Estimation Results (Lisrel).

The SEM model can be translated into the following structural equations (the complete results can be found in Appendix 6):

$$\xi 2=0.693 \xi 1 \dots \dots \dots \dots \dots \dots ..$$

(1)

$$\begin{array}{c}\eta 1=0.279 \xi 5+0.485 \xi 3+0.182 \xi 1+0.131 \xi 2-0.0577 \xi 4+0.287 \xi 6\\ +0.154 \xi 7 \dots \dots \dots \dots \dots \dots .. \end{array}$$

(2)

$$\eta 2=0.751 \xi 7 \dots \dots \dots ..$$

(3)

The path coefficient values from the structural equations above, along with their significance levels, are presented in the following table:

Table 28. Path Coefficients and Significance Levels of Variables.

Table 28. Path Coefficients and Significance Levels of Variables.

Based on the Structural Equation and the Table above, several conclusions can be drawn: At the 99% confidence level, Demand (ξ1) has a significant effect on Supply (ξ2). Furthermore, at the same confidence level, Risk/Benefit (ξ5), Dialogue (ξ3), Demand (ξ1), Supply (ξ2), Transparency (ξ6), and Sustainability (ξ7) have significant effects on VCC / Value Co-Creation (η1), while the variable Access (ξ4) has no effect on VCC / Value Co-Creation (η1). Additionally, at the 99% confidence level, Sustainability (ξ7) has a significant effect on the Marketing Platform (η2). The following sections elaborate further.

5. Discusion

Discussion of the Effect of Demand (ξ1) on Supply (ξ2)

Based on the Table, the latent variable Demand (ξ1) simultaneously explains the variation in the latent variable Supply (ξ2) with a coefficient of determination (R²) of 0.480 or 48%.The latent variable Demand (ξ1), consisting of the manifest variables: Awareness (X_1-1), Behavior (X_1-2), Lifestyle (X_1-3), Knowledge (X_1-4), Availability (X_1-5), Price (X_1-6), Trust in product (X_1-7), Quality assurance (X_1-8), Community certification (X_1-9), Innovation value (X_1-11), and Product quality (X_1-12), has a positive and significant effect (99% confidence level) on the latent variable Supply (ξ2), which consists of: Production yield (X_2-1), Climate change (X_2-2), Domestic market government policy (X_2-3), Export market government policy (X_2-4), Quality assurance (X_2-5), Information technology (X_2-6), Brand (X2-8), Packaging (X_2-9), Community (X_2-10), and Extension support (X_2-11). This indicates that improvements in demand indicators for organic vegetables will encourage an increase in their supply. Conversely, if demand indicators decline, the supply will also decrease.

Figure 7. New DSRT VCC Model.

Figure 7. New DSRT VCC Model.

Germany’s organic wholesale market is projected to increase by 13.8% from 2024 to 2030, due to producers' understanding of consumer demand factors such as product quality, pricing, and product safety assurance (Ministry of Trade, RI, 2013). The supply of organic products in the Mekar Tani Jaya farmer group has increased following the acquisition of SNI: 6729:2016 certification across 94% of its organic production processes. As previously understood, consumers who prioritize product quality standards are more satisfied with the implementation of SNI organic certification (Maryowani, 2012). The largest organic vegetable producer in Cibodas, Bandung Regency, ‘Saung Organik’, has adopted post-harvest handling strategies such as eco-friendly packaging, product labeling for safety, and INOFICE certification for packaging houses, contributing to high demand from distributors and modern markets to become loyal customers.

Research Contribution to the Organic Agriculture Business. The results of this study make a significant contribution to the practice of sustainable organic agriculture. Challenges remain in implementing environmentally friendly and sustainable agricultural practices, particularly across various agribusiness subsystems: in the upstream input supply, on-farm production or cultivation processes, downstream in marketing and distribution, as well as in the supporting subsystems related to partnership institutions.

The Value Co-Creation (VCC) Business-to-Customer (B2C) model for organic vegetables offers new insights into both theory and practice of sustainable organic farming by examining the dynamics of demand and supply for organic vegetables. The Collaborative Business Model serves as a standard for conducting business in the organic vegetable sector specifically, and organic farming practices more broadly. The creation of shared value between producers and consumers (value co-creation) in the B2C context—through innovations such as co-learning, co-service, co-design, and co-process—can guide business practitioners and organic vegetable farmer groups in applying these elements as part of their marketing strategies.

Marketing of organic vegetables can experience increased supply as a result of higher demand driven by a broader customer segment across various age groups, income levels, behaviors, and lifestyles. The marketing platform, which is built upon VCC principles for organic vegetable commodities, has proven effective and beneficial for consumers from Baby Boomers to Gen X, Y, Millennials, and Gen Z. The development of co-learning, co-service, co-design, and co-process features on the marketing platform can further boost demand, which in turn increases supply.

One of the key contributions of this research lies in its emphasis on the importance of increasing demand for organic vegetables not merely through production processes, which tend to focus solely on profitability. In reality, field data show that consumption of organic vegetables remains low due to higher prices compared to conventional vegetables. In fact, the cost of organic vegetable production is often lower than that of conventional farming. The price of organic vegetables is aligned with their production value—higher prices reflect higher quality. However, lower-income consumers still face challenges in accessing healthy vegetables.

6. Congclusion

The following are the conclusions from the research conducted: The demand and supply of organic vegetables in West Java need to consider prices that align with market equilibrium and ensure that the quantity of each type is appropriately prepared by each organic vegetable business. The sustainable B2C VCC (Value Co-Creation) model for organic vegetables is the DSRT model, which enhances the existing DART model. Therefore, a shift in mindset is needed: the value of organic vegetables should be based on necessity rather than price. When all stakeholders—farmer groups, entrepreneurs, and consumers—understand the intrinsic value of organic vegetables as a basic need, rather than a luxury item, demand will naturally increase, followed by a rise in supply.