Evaluating a sustainable circular economy model for the Indonesian fashion industry under uncertainties: a hybrid decision-making approach

ABSTRACT

There is a large increase in the wastes and emissions from the Indonesian fashion industry under economic and population growth. To address this issue, the sustainable circular economy is proposed to reach the goal of sustainable development and zero waste. Previous studies neglect to provide a specific interrelationship model to guide practice to reach a sustainable circular economy. This study proposes a hybrid method integrating exploratory factor analysis, the fuzzy synthetic method and the decision-making trial and evaluation laboratory approach to overcome these shortcomings. It contributes to enhancing the understanding of theory through visual diagrams. The hybrid method makes it possible to interpret complex interrelationships and leads to improvements under resource limitations based on the model. The results reveal that emission reduction and welfare improvement are the causal aspects.

Graphic Abstract

KEYWORDS:

• Sustainable circular economy
• Indonesian fashion industry
• fuzzy synthetic method
• decision-making trial and evaluation laboratory
• decisive criteria

1. Introduction

Population growth with increasing income standards has caused the consumption of textiles and fiber production in Indonesia to aggressively grow (Shirvanimoghaddama et al., 2020). This booming growth has also generated vast pollution in this linear production process; for example, dyeing requires the use of a large amount of water, and cutting shapes generates solid wastes [1,2]. The industry has relied on incineration to eliminate these solid wastes, causing toxic air and hazardous water to pollute several areas in Indonesia, such as the Citarum River in West Java and the Cikijing River in Central Java [3]. The Indonesian government has announced several regulations and policies for achieving a 29% reduction in carbon emissions to address these issues [4]. Although these regulations and policies may guide local firms to practice a sustainable circular economy (SCE), for local firms, realizing an SCE is still difficult due to several uncertainties, including awareness, engagement, production consumption, welfare, and so on [5–7].

The SCE is a multidimensional concept that integrates sustainability and a circular economy to enhance stakeholder awareness and encourage engagement in launching sustainable production and consumption to reach energy efficiency and reduce carbon emissions while improving human welfare [8b; 9, 10]. In the fashion industry, 11, adopted the concept of sustainability in making fashion and textile designs to reach a circular economy by generating zero waste. 12,identified the trend of the SCE in the fashion industry by investigating the changes in younger consumers and the effect of gender. 13,discovered that social drivers and ecological impacts provide resilience for the fashion industry to launch an SCE. Although these studies have attempted to address the SCE for the fashion industry by focusing on qualitative approaches and literature reviews, a model that bridges theory and practice is still lacking in the current discussion.

To address these gaps, this study proposes a hybrid method integrating exploratory factor analysis (EFA), the fuzzy synthetic method (FSM) and the decision-making trial and evaluation laboratory (DEMATEL). EFA enables us to ensure the validity and reliability of the structured model [14–16]. The FSM is utilized to transform qualitative information into computational figures by converting the linguistic scales of expert opinions into crisp values [17]. Furthermore, the complex interrelationships can be addressed by adopting the DEMATEL to identify and map the criteria into cause and effect and influential diagrams for aspects [18,19]. These generated diagrams not only present the precise interrelationships for enhancing understanding but also provide a guideline for related firms to improve under limited resources. Accordingly, this study has the following objectives.

• To develop an evaluation model for reinforcing the basis of the SCE

• To adopt a hybrid method to address the uncertainties by considering expert assessments and public opinions

• To guide local firms in practicing an SCE under uncertainties

This study makes three contributions. (1) The structural aspects shown in the influential diagram can promote understanding and bridge theory and practice; (2) the proposed hybrid method enables us to address the complex interrelationships in the obtained model; and (3) the visual analysis diagrams provide a specific guideline to facilitate related firms in launching an SCE under resource limitations.

The remaining sections proceed as follows. Section 2 includes the theoretical background, information about previous SCE studies, and the proposed methods and measures. Section 3 comprehensively explains the methods and the analytical procedure for generating the model. Section 4 provides information about the case background and displays the analytical results from the data. Section 5 delivers the findings with theoretical and managerial implications. The last section provides the conclusion, the limitations of this study, and future recommendations.

2. Literature review

This section provides the theoretical background for enhancing understanding. The proposed method and proposed measures are also included in this section.

2.1. Sustainable circular economy

The SCE concept has the primary goal of integrating the circular economy (CE) concept and sustainability, which could have many positive impacts on the macrolevel, such as the efficient use of natural resources, economic growth, and sustainable consumption. These positive impacts will lead to the restoration of energy and circularity [20]. Considering the risks posed to environmental stability, which can threaten the lives of humans, animals, and other forms of life, caused by the depletion of the ozone layer and pollution, an SCE should attempt to restore and regenerate the environment [21]. Doing so requires sustainable engagement to encourage the entire supply chain to launch SCE practices [22]. With an SCE, the purpose of organizations has evolved, focusing on how to make money and generate money while organizing resources to enhance community well-being [23,24]. However, the transition to an SCE requires all stakeholders to pay attention to planetary boundaries [25]. Doing so includes more than just improving the handling of waste; it requires waste prevention and the extension of product life.

For instance, 26,examined the importance of circular fashion and textiles and discussed various strategies and innovations in the circular textile economy, including methods for reusing, recycling, and repurposing textile waste. 27, presented a system to detect and classify portions of the machine waste-textile sorter to solve the problem of classifying textile waste. 23, proposed a value framework and a set of principles for sustainable production and consumption to design, implement, and evaluate an SCE. 28, highlighted that energy savings and efficiency improvement make it possible to balance ecology, the economy and geography to pursue an SCE. 29, demonstrated that carbon emission reduction is an important element for firms to track in aligning SCE practices. 30, also pointed out that the SCE relies on offering better welfare to people to change behavior toward sustainability. However, a study that includes these points to conduct an extensive SCE assessment is lacking.

Research has indicated that SCE practices in textile firms have become a common concern, especially in developed countries, and have begun to become a severe concern in several developing countries, such as India, as shown in the literature. In Indonesia, many start-ups have emerged that focus on an SCE as their mission, such as Sukkhacitta. However, there are still limited studies on the SCE in Indonesian fashion firms. Thus, Indonesia’s fashion firms are selected as a case study to address the research gaps.

2.2. Proposed methods

Few studies use valid and reliable statistical methods to create a multilevel framework and translate qualitative information (Tseng et al., 2019). This study uses EFA to define the validity of measurements, find the main criteria, and group a high quantity of criteria into a small number of aspects by considering the correlation between criteria [15,16]. 17, employed EFA to determine the relevant influences using the proposed criteria. 31, applied EFA to examine whether the constructs are unidimensional and, consequently, to identify the factors that represent a thing in common, i.e. the diversity of factors used to explain one concept in light of the COVID-19 model. There is still a lack of qualitative information provided in those studies, even though the validity and reliability of the arranged criteria have been determined.

The FSM is applied to construct a critical importance level for each group of criteria. Using fuzzy analysis to measure and convert uncertain data, the FSM is a branch of fuzzy set theory that allows the general assessment of associated criteria. Moreover, this method can facilitate the objective evaluation of subjective opinions and perceptions by professionals [32]. 33, used the FSM to demonstrate that the judgment for flame combustion systems is coherent and compelling. 16, utilized the FSM to address hierarchical structure and decision-making. The literature shows that several studies have used the FSM to examine multicriteria decision-making. This method is considered suitable to ascertain the most important criteria in SCE practice in Indonesian fashion firms.

The DEMATEL is needed in this study to assess both the direct and indirect relationships between aspects and to identify the cause-effect relationship between criteria [18]. 18, used the DEMATEL to establish the cause-effect relationship among the fundamental challenges in establishing an SCE in the fashion industry. 34, determined the causes and effects of barriers to textile trade by using a DEMATEL approach involving a well-defined and comprehensive process that includes textile sector experts. These studies also suggested an additional method for future research to better comprehend the hierarchical interrelationships among SCE implementation obstacles. Based on these studies, a hybrid model was proposed to complement research on the SCE in the textile industry.

2.3. Proposed measures

According to 35, innovating by offering sustainable mini collection clothes (C5) and the customization of eco-conscious wedding attire (C9) as a part of eco-innovation can help minimize environmental costs. Previous studies have also argued that a firm’s ethical standards should be established (C3) to enhance customer sustainability. Furthermore, using lower production costs upfront to gain additional benefits from products can capture the value of the CE. Additionally, appreciating consumers’ concern for sustainability and environmental protection (C7) means that consumers are an essential part of firms that must be treated well based on the principles of the CE [36]. To help consumer behavior change toward sustainability, educating people on sustainability (C2), keeping information transparent (C4), providing access to change sustainable consumer patterns (C14), and organizing social events about sustainability (C10) can be adopted [37, Khor & Hazen, 2016; 38, 39].

In addition, consumer trust can be established by generating social impact by gaining media coverage (C11), enhancing credibility by obtaining awards for sustainability (C1), and optimizing social responsibility to meet community needs for a suitable environment (C16) [40–42]. Corporate social responsibility, such as involving villagers in the whole supply chain (C12) and reinvesting profits in villages to empower villagers (C15), is a concern that is equally as urgent as the sustainability of the environment to protect humanity [43,44]. Moreover, consumer barriers to purchasing sustainable products can be reduced by maintaining service quality (C8) and realizing its importance [45]. Apart from paying attention to consumers, collaborating with various stakeholders for mutual success is an effort to develop a strategic plan to ensure long-term growth and profitability [46]. As a result of a large amount of textile waste, realizing circular fashion by recycling unused clothes (C18) can be a suitable solution for reducing solid waste in landfills (Shirvanimoghaddama et al., 2020). Finding the best natural dyes by continually researching plants (C17) should also be in line with this concept [47].

Focusing on the SCE through longer product lifespans and reusing materials prevents the overproduction of waste and creates better value from each product (Shirvanimoghaddama et al., 2020). This includes redesigning manufacturing processes for enhancing natural dying adoption (C19), adopting zero-waste cutting during the assembly process (C27), recycling pearl waste into clothes buttons (C30), and using certified traceable 100% recycled material (31) [48], Shirvanimoghaddama et al., 2020; Ayçin & Kayapinar, 2021]. The textile industry is cost driven, with the whole industry competing to reduce prices and attempting to lessen the social and environmental costs of the business [49]. Any adoption of reducing the cost of purchasing raw materials by applying the principle of “cleaner production” (C20), using cheap and environmentally friendly fuels available in the community (C23), reducing the cost of material procurement by cultivating natural resources (C29), and diverting the cost of purchasing dyes to artisans (C21) might help firms maximize profits [31,38,49,50]. Activities within the fabricating field, for example, providing training courses to reinforce the skills of artisans (C22), involving local farmers in natural raw materials provision (C24), ensuring that the materials used are taken from nature (C28) and considering increasing the income of artisans (C25), are supposed to improve sustainability performance [32,38,51]. [52, mentioned that traditionally rooted knowledge and skills within the local community preserving the unique culture of artisans (C26) should be recognized as a particular form of cultural inheritance for contemporary culture.

In terms of design, a sustainable society would benefit by applying eco-design packaging (C33) (Shirvanimoghaddama et al., 2020). Improving the existing production system in the community (C34) is a vital step toward an SCE [53]. This includes repurposing small pieces of clothes (C35), using full inspection to ensure the quality of remanufactured products (C38), and striving for people who are involved in production activities to earn a decent wage (C36) [51, Shirvanimoghaddama et al., 2020; 53]. Recording artisanal handicrafts to protect cultural heritage (C37) and increasing the benefits of unique products that are continuously improved in quality (C32) should be part of social construction as a continual process of cultural sustainability [52].

The use of energy from renewable resources (C43), maintaining the uniqueness and authenticity of products by implementing handmade production (C41) and eliminating the cost of purchasing production machines (C39) reflect the third principle of the CE in terms of renewable energy. Furthermore, the price adjustment process for a product is vital for firms, refusing to follow the sales system usually applied by other fashion brands (C42). Additionally, eliminating the cost of purchasing production machines (C39) could reduce the production cost, allowing firms to set prices for consumers based on the level of perceived acceptable price [18,47,50].

The critical steps in producing a fully sustainable clothing item in terms of reducing carbon emissions are exporting products using the most sustainable shipping route (C44) and selecting a more economical and more environmentally friendly means of transportation (C45) [49]. Furthermore, the community is increasingly seen as key to realizing the CE. Therefore, one measure is to better empower female manufacturing workers (C47) [43]. Regenerating the agricultural system for material provision (C46) is also part of forming a successful SCE [Shirvanimoghaddama et al., 2020).

3. Method

This section introduces how to apply the proposed hybrid method to obtain the analytical results to guide their application by practitioners.

3.1. EFA and reliability testing

EFA enables the discovery of potential relationships by adopting the principal component method. 19,emphasized that the Kaiser‒Meyer‒Olkin test is used to confirm the validity of a potential relationship within EFA. Therefore, the value of the Kaiser‒Meyer‒Olkin test result needs to be greater than 0.7, with the $p-value$ needing to be smaller than 0.05 to reach the standard. Moreover, the factor loadings must be greater than 0.4 to confirm validity by checking the rotated component matrix. Once the potential relationships have been discovered, the reliability test must be utilized to confirm internal consistency. Internal consistency relies on Cronbach’s $\alpha$ for confirmation, and it must be greater than 0.6 (17].

3.2. FSM-DEMATEL

3.2.1. Computation of criteria

After confirming validity and reliability, experts are required to assess the interrelationships between the proposed criteria and structured aspects based on their knowledge and experiences. Expert assessments are presented as linguistic preferences that include no influence $\left(N\right)$, rare influence $\left(R\right)$, equal influence $\left(E\right)$, slight influence $\left(S\right)$ and strong influence $\left(H\right)$. These linguistic preferences can be denoted as ${\left(L_{xy}\right)}_{ce}^a$. Here, $a$ represents the number of structured aspects from $c$ number of proposed criteria for which the interrelationships are assessed by $e$ number of experts. Accordingly, these accumulated assessments can be rewritten as the following equation:

(1) ${\left(L_{xy}\right)}_{ce}^a={\left(N_{xy},R_{xy},E_{xy},S_{xy},H_{xy}\right)}_{ce}^a$(1)

However, these accumulated assessments are still qualitative. The following equations need to be used to convert them into crisp values ${\left({\overline{CP}}_{xy}\right)}_c^a$:

(2) ${\left({\bar{L}}_{xy}\right)}_c^a={\left(\frac{N_{xy}}{e},\frac{R_{xy}}{e},\frac{E_{xy}}{e},\frac{S_{xy}}{e},\frac{H_{xy}}{e}\right)}_c^a$(2)

(3) ${\left({\overline{CP}}_{xy}\right)}_c^a={\left(1\times \frac{N_{xy}}{e}+2\times \frac{R_{xy}}{e}+3\times \frac{E_{xy}}{e}+4\times \frac{S_{xy}}{e}+5\times \frac{H_{xy}}{e}\right)}_c^a$(3)

For the obtained crisp values, the following equation needs to be used to arrange them into self-relation matrices ${\left(B_{xy}\right)}_c^a$:

(4) ${\left(B_{xy}\right)}_c^a={\left[\begin{array}{cccc}{\overline{CP}}_{11}& {\overline{CP}}_{12}& \cdots & {\overline{CP}}_{1y}\\ {\overline{CP}}_{21}& {\overline{CP}}_{22}& \cdots & {\overline{CP}}_{2y}\\ ⋮& ⋮& \ddots & ⋮\\ {\overline{CP}}_{x1}& {\overline{CP}}_{x2}& \cdots & {\overline{CP}}_{xy}\end{array}\right]}_{c\times c}^a={\left[b_{xy}\right]}_{c\times c}^a$(4)

The self-relation matrices need to be integrated into a direct relation matrix by adopting the following equation:

(5) $D=\frac{{\sum }_{x=1}^c{\left[b_{xy}\right]}_{c\times c}^a}{a}={\left[d_{xy}\right]}_{c\times c}$(5)

The following equations are utilized to generate the total relation matrix:

(6) $\bar{D}=\frac{d_{xy}}{\underset{1\le x\le c}{max}{d}_x}$(6)

(7) $D{ }^{\mathrm{\prime }}=\bar{D}\times {\left(U-\bar{D}\right)}^{-1}={\left[d_{xy}^{\prime }\right]}_{c\times c}$(7)

where $U$ is the unit matrix.

After obtaining the total relation matrix, the following equations are adopted to calculate the dependence and driving power:

(8) $d^p={\left[\sum _{x=1}^{c}d{\prime }_{xy}\right]}_{c\times 1}={\left[d{\prime }_x\right]}_{c\times 1}$(8)

(9) $d^r={\left[\sum _{y=1}^{c}d{\prime }_{xy}\right]}_{1\times \mathrm{c}}={\left[d{\prime }_y\right]}_{1\times \mathrm{c}}$(9)

Mapping the criteria based on the $\left(d^p+d^r,d^p-d^r\right)$ coordinates draws the cause and effect diagram. $\left(d^p+d^r\right)$ is the $x$ axis, which represents the importance of the criteria. $\left(d^p-d^r\right)$ is the $y$ axis, which categorizes the criteria into cause and effect groups. If $\left(d^p-d^r\right)>0$, it means that the criteria that possess a higher causal influence affect the other criteria; conversely, if $\left(d^p-d^r\right)<0$, it means that the criteria belonging to the effect group are influenced by the causal criteria. Moreover, $\left(d^p+d^r,d^p-d^r\right)$ can divide the cause and effect diagram into four quadrants. Criteria that are located in the first quadrant are called driving factors because these factors have higher importance in possessing the causal feature. The criteria in the second quadrant are voluntary factors, which have lower importance but have the causal feature. The criteria in the third quadrant are independent factors due to their lower importance and possession of the effect feature. The criteria in the last quadrant are called core problems because these criteria have higher importance and the effect feature. In other words, these problems cannot be directly addressed, meaning that these criteria rely on the driving factors to improve their performance.

3.2.2. Computation of aspects

Subsequently, the following equation is used to generate the measurement criticality to present the interrelationships between the proposed criteria and structured aspects:

(10) ${\left(F_{xy}\right)}_c^a=\sqrt[c]{\prod _{x=1}^c{\left({\overline{CP}}_{xy}\right)}_c^a}={\left(f_{xy}\right)}_{1\times c}^a$(10)

To obtain the factor weights, the following equation is adopted:

(11) ${\left({\bar{F}}_{xi}\right)}_{\bar{c}}=\frac{{\left(F_{xy}\right)}_c^a}{{\sum }_{x=1}^{\beta }{\left(F_{xy}\right)}_c^a}={\left(f_{xi}\right)}_{1\times c},i=1,2,\cdots ,c$(11)

where $\beta$ is the criteria belonging to the structured aspect based on EFA.

The membership function of each aspect needs to be aggregated by adopting the following equation:

(12) ${\left({\overline{\overline{L}}}_{ij}\right)}_k={\left[\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(N_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(R_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(E_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(S_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(H_{xy}\right)}_c^a}}{a}\right]}_{c\times k}={\left({\ell }_{ij}\right)}_{c\times k}$(12)

These membership functions need to be associated with factor weights to obtain the total membership function by adopting the equations below:

(13) ${\left(M_{xj}\right)}_{\bar{c}\times k}={\left({\bar{F}}_{xi}\right)}_{\bar{c}}\times {\left({\overline{\stackrel{‾}{L}}}_{ij}\right)}_k={\left(f_{xi}\right)}_{1\times c}\times {\left({\mathrm{\ell }}_{ij}\right)}_{c\times k}={\left(m_{xj}\right)}_{\bar{c}\times k}$(13)

where ${\left(m_{xj}\right)}_{\bar{c}\times k}$ can be redefined as ${\left(m_{xj}^N,m_{xj}^R,m_{xj}^E,m_{xj}^S,m_{xj}^H\right)}_{\bar{c}\times k}$.

The total membership function must be converted into crisp values and be arranged into the direct relation matrix $M{ }^{\mathrm{\prime }}$ of the aspects by using the following equations:

(14) $M{ }^{\mathrm{\prime }}={\left(1\times m_{xj}^N+2\times m_{xj}^R+3\times m_{xj}^E+4\times m_{xj}^S+5\times m_{xj}^H\right)}_{\bar{c}\times k}={\left(m_{op}^{\prime }\right)}_{a\times a}$(14)

Repeating Equations (5)(5) $D=\frac{{\sum }_{x=1}^c{\left[b_{xy}\right]}_{c\times c}^a}{a}={\left[d_{xy}\right]}_{c\times c}$(5) -(9(9) $d^r={\left[\sum _{y=1}^{c}d{\prime }_{xy}\right]}_{1\times \mathrm{c}}={\left[d{\prime }_y\right]}_{1\times \mathrm{c}}$(9) ) generates the dependence and driving power of the aspects. Then, the threshold value is calculated to identify the influence among the aspects via the following equation:

(15) $\theta =\frac{{\sum }_{o=1}^a{m}_{op}^{\prime }}{a^2}$(15)

If $\theta <m_{op}^{\prime }$, there exists a significant interrelationship between two aspects; otherwise, there is no obvious interrelationship.

1. Proposed Analytical Procedures

2. The draft criteria are selected based on reviewing the literature to confirm validity. However, for the selected criteria, consultation with experts is necessary to reflect the reality of the case firm. To reduce the issue of subjectivity, this study conducts a questionnaire to obtain respondents from the Indonesian public by employing random sampling.

3. The collected respondents used five-point Likert scales, for which it is necessary to adopt EFA to structure the potential aspects. For the structured aspects, it is necessary to employ the Kaiser‒Meyer‒Olkin test, with results passing the threshold value of 0.7 and $p-value<0.05$ to ensure validity. In addition, the factor loading from the rotated component matrix must be higher than 0.4. Once the structured aspects are confirmed, the reliability test must be adopted to ensure internal consistency by checking whether Cronbach’s $\alpha >0.6$.

4. The interrelationships between the proposed criteria and structured aspects still rely on experts to assess. However, these assessments adopt linguistic preferences, requiring Equations (1)(1) ${\left(L_{xy}\right)}_{ce}^a={\left(N_{xy},R_{xy},E_{xy},S_{xy},H_{xy}\right)}_{ce}^a$(1) -(3(3) ${\left({\overline{CP}}_{xy}\right)}_c^a={\left(1\times \frac{N_{xy}}{e}+2\times \frac{R_{xy}}{e}+3\times \frac{E_{xy}}{e}+4\times \frac{S_{xy}}{e}+5\times \frac{H_{xy}}{e}\right)}_c^a$(3) ) to convert them into crisp values for further computation. Equation (4)(4) ${\left(B_{xy}\right)}_c^a={\left[\begin{array}{cccc}{\overline{CP}}_{11}& {\overline{CP}}_{12}& \cdots & {\overline{CP}}_{1y}\\ {\overline{CP}}_{21}& {\overline{CP}}_{22}& \cdots & {\overline{CP}}_{2y}\\ ⋮& ⋮& \ddots & ⋮\\ {\overline{CP}}_{x1}& {\overline{CP}}_{x2}& \cdots & {\overline{CP}}_{xy}\end{array}\right]}_{c\times c}^a={\left[b_{xy}\right]}_{c\times c}^a$(4) arranges the crisp values into the self-relation matrix. Equations (5)(5) $D=\frac{{\sum }_{x=1}^c{\left[b_{xy}\right]}_{c\times c}^a}{a}={\left[d_{xy}\right]}_{c\times c}$(5) -(7(7) $D{ }^{\mathrm{\prime }}=\bar{D}\times {\left(U-\bar{D}\right)}^{-1}={\left[d_{xy}^{\prime }\right]}_{c\times c}$(7) ) are utilized to obtain the total relation matrix. Then, using Equations (8)(8) $d^p={\left[\sum _{x=1}^{c}d{\prime }_{xy}\right]}_{c\times 1}={\left[d{\prime }_x\right]}_{c\times 1}$(8) and (9(9) $d^r={\left[\sum _{y=1}^{c}d{\prime }_{xy}\right]}_{1\times \mathrm{c}}={\left[d{\prime }_y\right]}_{1\times \mathrm{c}}$(9) ), we generate the dependence $\left(d^p\right)$ and driving power $\left(d^r\right)$. Based on the coordinates $\left(d^p+d^r,d^p-d^r\right)$, the criteria can be mapped onto the four quadrants of the cause and effect diagram for visual analysis.

5. Regarding the aspects, the measurement criticalities need to be obtained by Equation (10)(10) ${\left(F_{xy}\right)}_c^a=\sqrt[c]{\prod _{x=1}^c{\left({\overline{CP}}_{xy}\right)}_c^a}={\left(f_{xy}\right)}_{1\times c}^a$(10) . The measurement criticalities need to be transferred to the factor weights by employing Equation (11)(11) ${\left({\bar{F}}_{xi}\right)}_{\bar{c}}=\frac{{\left(F_{xy}\right)}_c^a}{{\sum }_{x=1}^{\beta }{\left(F_{xy}\right)}_c^a}={\left(f_{xi}\right)}_{1\times c},i=1,2,\cdots ,c$(11) . Then, Equations (12)(12) ${\left({\overline{\overline{L}}}_{ij}\right)}_k={\left[\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(N_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(R_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(E_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(S_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(H_{xy}\right)}_c^a}}{a}\right]}_{c\times k}={\left({\ell }_{ij}\right)}_{c\times k}$(12) -(14(14) $M{ }^{\mathrm{\prime }}={\left(1\times m_{xj}^N+2\times m_{xj}^R+3\times m_{xj}^E+4\times m_{xj}^S+5\times m_{xj}^H\right)}_{\bar{c}\times k}={\left(m_{op}^{\prime }\right)}_{a\times a}$(14) ) are adopted to generate the total membership function and acquire the crisp values. These crisp values need to be arranged into the direct relation matrix by repeating Equations (5)(5) $D=\frac{{\sum }_{x=1}^c{\left[b_{xy}\right]}_{c\times c}^a}{a}={\left[d_{xy}\right]}_{c\times c}$(5) -(7(7) $D{ }^{\mathrm{\prime }}=\bar{D}\times {\left(U-\bar{D}\right)}^{-1}={\left[d_{xy}^{\prime }\right]}_{c\times c}$(7) ) to compute the total relation matrix. Then, using Equations (8)(8) $d^p={\left[\sum _{x=1}^{c}d{\prime }_{xy}\right]}_{c\times 1}={\left[d{\prime }_x\right]}_{c\times 1}$(8) and (9(9) $d^r={\left[\sum _{y=1}^{c}d{\prime }_{xy}\right]}_{1\times \mathrm{c}}={\left[d{\prime }_y\right]}_{1\times \mathrm{c}}$(9) ), we obtain the dependence and driving power of the aspects. The aspects also need to be mapped onto the cause and effect diagram based on the coordinates $\left(d^p+d^r,d^p-d^r\right)$. Finally, Equation (15)(15) $\theta =\frac{{\sum }_{o=1}^a{m}_{op}^{\prime }}{a^2}$(15) can obtain the threshold for identifying the interrelationships among the aspects.

4. Empirical results

This section introduces the background of the case firm and presents the detailed analytical computations in explaining the derived results.

4.1. Case information

Focusing on empowerment, sustainable production, and environmental responsibility, Sukkhacitta was founded in 2016 by Denica Flesh, a former development consultant in the social development program at the World Bank. Ms. Flesh returned to Indonesia to create an impact on changing the standards of Indonesia’s fashion industry. She started conducting research by going from village to village, and then realized that these laborers are struggling due to the unfair system. The firm was created to change that system by building four craft schools providing continuous training, from business skills training to environmentally friendly practices and quality management. As a result, approximately 1482 people were impacted, the incomes of the firm’s artisans have increased by 60%, and 36 scholarships were funded. Regarding ecosystem impacts, more than 10,000 poly-plastic bags have been avoided through zero waste pouches, 1.2 Mio liters of toxic wastewater has been prevented, 25 tons of CO2 has been saved, and 20 hectares of soil has been regenerated. Sukkhacitta hopes to compete locally and globally while still providing economic, social, and environmental benefits by constantly improving its business activities in terms of sustainability and the CE. The purpose of the firm can be found in its name, Sukkhacitta, which means happiness in Indonesian, and it embodies the mission of returning pride to those who have been invisible for a long time.

In line with its vision, the firm is striving for and seeking the best way to implement an SCE in its business lines. For example, the firm has continuously researched natural dyes, as clothes dyeing is the most polluting aspect of textile production. The firm utilizes indigo leaves and palms for blue and golden fruits for yellow while simultaneously performing regenerative agriculture to maintain the plant life cycle. In addition, the firm has grown its cotton regeneratively, meaning in ways that restore soil health and help bind CO₂ from the atmosphere without pesticides or fertilizers. Furthermore, the firm recycles leftover fabric from production into innovative products, such as custom wedding clothes and unique clothes for children, with zero plastic packaging and paper tags. In this way, the firm has saved +25 tons of CO2 (and avoided +10,000 poly-plastic bags). Although the firm is moving toward an SCE, it still lacks a model as a guideline to facilitate its efforts to maximize its SCE practice. Unless such a guideline exists, the firm will have difficulty utilizing its resources, resulting in ineffective SCE development. For example, the firm is still having difficulty determining an effective way to become genuinely circular even though several methods have been implemented, such as offering free repair services and using durable materials to extend the life of clothes. To address this issue, this study proposes the FSM-DEMATEL to assist the firm in exploring the decisive factors of SCE practices under limited resources.

4.2. Analytical results

1. This study adopted random sampling to distribute 346 questionnaires to the Indonesian public by using an online survey. For the respondents (as shown in Table 1), it is necessary to utilize EFA and reliability testing to discover the potential relationships between the criteria and the structured aspects, as presented in Table 2. The factor loadings of all criteria are greater than 0.4, and the Kaiser‒Meyer‒Olkin value is 0.961 and significant at the 0.000 level, which confirms the validity of the discovered relationships. There are two criteria, restoring the intricate ecosystem for protecting the environment and materials provision and maintaining customer loyalty by providing after-sales service, which have been deleted because the factor loadings do not reach the 0.4 level.

   Table 1. Sample of the collected respondents.

   Criteria Respondent	C1	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	C12	C13	C14	C15	C16	C17	C18	C19	…	C47	C48	C49
   R1	5	5	5	5	4	5	4	5	5	5	5	5	4	4	5	5	4	3	3	…	5	3	3
   R2	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	3	…	5	5	2
   R3	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	1	2	…	2	2	2
   R4	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	…	5	5	5
   R5	5	5	4	5	5	5	5	5	5	5	5	5	5	5	5	5	5	3	4	…	5	5	4
   R6	5	5	5	5	5	5	5	5	5	5	5	5	5	5	4	5	5	3	4	…	5	5	5
   R7	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	3	…	5	5	5
   R8	4	4	4	4	5	4	4	4	4	5	4	4	5	4	4	4	4	4	5	…	4	5	4
   R9	5	5	5	5	5	5	4	5	5	4	5	5	5	4	5	5	5	4	5	…	4	4	4
   R10	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	…	5	5	5
   R11	5	4	5	5	5	5	5	4	3	4	5	5	5	5	5	5	3	3	3	…	5	3	2
   R12	5	5	5	5	1	5	1	5	5	5	5	5	5	5	5	5	1	1	1	…	5	5	3
   R13	4	5	4	5	4	5	5	4	5	5	4	4	4	5	3	4	3	4	4	…	5	4	5
   R14	5	5	4	4	4	4	3	4	3	4	4	4	3	3	4	5	4	3	3	…	4	3	4
   R15	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	3	…	5	5	5
   R16	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	3	3	…	5	5	5
   R17	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	…	5	5	5
   R18	5	5	4	4	4	4	4	4	5	4	5	5	4	4	5	4	4	5	3	…	4	5	4
   …	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…
   R343	3	3	5	4	4	3	4	4	4	5	4	4	5	4	5	5	4	3	3	…	5	5	5
   R344	5	5	4	5	4	5	5	5	5	5	5	5	5	5	4	5	5	3	3	…	4	4	5
   R345	4	4	5	4	5	4	4	4	4	4	4	4	4	4	4	4	4	4	4	…	4	4	4
   R346	5	4	3	4	4	3	3	4	3	5	4	3	4	4	3	5	3	3	3	…	4	4	3

   Table 2. EFA and reliability testing results.

   Aspects	Cronbach’s $\alpha$	Criteria	Factor loading
   Stakeholder awareness (A1)	0.949	C1	0.663
   		C2	0.661
   		C3	0.653
   		C4	0.649
   		C5	0.647
   		C6	0.625
   		C7	0.607
   		C8	0.607
   		C9	0.590
   		C10	0.587
   		C11	0.580
   		C12	0.575
   		C13	0.568
   		C14	0.546
   		C15	0.520
   		C16	0.520
   		C17	0.514
   		C18	0.449
   Sustainable engagement (A2)	0.923	C19	0.704
   		C20	0.655
   		C21	0.654
   		C22	0.648
   		C23	0.627
   		C24	0.626
   		C25	0.601
   		C26	0.581
   		C27	0.553
   		C28	0.545
   		C29	0.511
   		C30	0.474
   		C31	0.423
   Sustainable production and consumption (A3)	0.897	C32	0.651
   		C33	0.547
   		C34	0.541
   		C35	0.533
   		C36	0.520
   		C37	0.468
   		C38	0.447
   Energy savings and efficiency (A4)	0.851	C39	0.842
   		C40	0.826
   		C41	0.639
   		C42	0.592
   		C43	0.585
   Carbon emission reduction (A5)	0.854	C44	0.704
   		C45	0.643
   Welfare improvement (A6)	0.614	C46	0.549
   		C47	0.532

2. The structured aspects include stockholder awareness (A1), sustainable engagement (A2), sustainable production and consumption (A3), energy savings and efficiency (A4), carbon emission reduction (A5) and welfare improvement (A6), satisfying the Cronbach’s $\alpha >0.6$ standard. This shows that the discovered relationships are highly significant and possess validity and reliability.

3. Experts (information on whom is shown in Table 3) are required to assess their interrelationships by adopting linguistic preferences, as presented in Table 4. However, the linguistic preferences are qualitative, and Equations (1)(1) ${\left(L_{xy}\right)}_{ce}^a={\left(N_{xy},R_{xy},E_{xy},S_{xy},H_{xy}\right)}_{ce}^a$(1) -(3(3) ${\left({\overline{CP}}_{xy}\right)}_c^a={\left(1\times \frac{N_{xy}}{e}+2\times \frac{R_{xy}}{e}+3\times \frac{E_{xy}}{e}+4\times \frac{S_{xy}}{e}+5\times \frac{H_{xy}}{e}\right)}_c^a$(3) ) must be used to obtain crisp values, which are shown in Table 5. Table 6 provides the total relation matrix for the criteria by employing Equations (4)(4) ${\left(B_{xy}\right)}_c^a={\left[\begin{array}{cccc}{\overline{CP}}_{11}& {\overline{CP}}_{12}& \cdots & {\overline{CP}}_{1y}\\ {\overline{CP}}_{21}& {\overline{CP}}_{22}& \cdots & {\overline{CP}}_{2y}\\ ⋮& ⋮& \ddots & ⋮\\ {\overline{CP}}_{x1}& {\overline{CP}}_{x2}& \cdots & {\overline{CP}}_{xy}\end{array}\right]}_{c\times c}^a={\left[b_{xy}\right]}_{c\times c}^a$(4) -(7(7) $D{ }^{\mathrm{\prime }}=\bar{D}\times {\left(U-\bar{D}\right)}^{-1}={\left[d_{xy}^{\prime }\right]}_{c\times c}$(7) ). Equations (8)(8) $d^p={\left[\sum _{x=1}^{c}d{\prime }_{xy}\right]}_{c\times 1}={\left[d{\prime }_x\right]}_{c\times 1}$(8) and (9(9) $d^r={\left[\sum _{y=1}^{c}d{\prime }_{xy}\right]}_{1\times \mathrm{c}}={\left[d{\prime }_y\right]}_{1\times \mathrm{c}}$(9) ) can be used to generate the dependence and driving power, as shown in Table 7. The computation is $d_{C1}^p=0.604+0.595+0.591+0.596+0.598+0.594+0.598+0.593+0.600+0.592+0.595+0.597+0.593+0.592+0.598+0.594+\cdots +0.589+0.581=27.890$ and

   $\begin{array}{rl}& d_{C1}^r=0.604+0.638+0.604+0.624+0.605+0.609\\ & +0.627+0.615+0.615+0.622+0.624+0.629\\ & +0.630+0.618+0.623+0.628+0.608+0.622n\\ & +0.634+0.617+0.612+0.643=29.197\end{array}$
   . Drawing the cause and effect diagram in accordance with the coordinates $\left(d^p+d^r,d^p-d^r\right)$ maps the criteria, as shown in Figure 1.

   Table 3. Information on the experts.

   	Name	Position	Work Experience	Educational Level
   EX1	Dr. Ir. Suwarso, M.Si	Operational director	34 years	Doctoral degree
   EX2	Ir. Purwadi, MM	Operational director	28 years	Master’s degree
   EX3	Prof. Dr. Ulfah Utami, M.Si	Lecturer	25 years	Doctoral degree
   EX4	Dr. Sukarsono, M.Si	Lecturer	30 years	Doctoral degree
   EX5	Drs. Muh Saiful, M.Si	Health and safety manager	12 years	Master’s degree
   EX6	Tsanil Falah S.ST	Health, safety and environment division	5 years	Bachelor’s degree
   EX7	Kevin Harsono ST MT	Principal architect	5 years	Master’s degree
   EX8	Rizal Fawzi S.ST	Marketing	5 years	Bachelor’s degree
   EX9	Dr. Ir, Rachmi Widiriani, M.Si	Operational director	21 years	Doctoral degree

   Table 4. Sample of expert assessments for C1 under A1.

   	EX1	EX2	EX3	EX4	EX5	EX6	EX7	EX8	EX9
   C1	E	E	E	E	E	E	E	E	E
   C2	H	S	S	H	H	S	E	S	H
   C3	E	E	E	E	E	S	S	S	E
   C4	S	S	S	H	S	H	S	S	S
   C5	H	S	H	H	H	S	E	S	N
   C6	H	S	H	E	H	H	E	E	R
   C7	S	S	S	H	S	S	E	E	R
   C8	S	S	S	H	S	S	S	E	S
   C9	H	S	H	H	H	E	E	R	E
   C10	E	E	E	E	E	E	S	H	S
   C11	E	E	E	E	E	S	E	S	N
   C12	S	S	S	H	R	H	R	H	E
   C13	E	E	E	E	E	S	E	H	E
   C14	S	S	S	H	S	H	E	R	S
   C15	E	E	E	E	E	S	S	H	E
   C16	H	S	S	H	H	S	S	E	H
   C17	H	S	H	E	H	S	E	S	R
   C18	H	S	H	E	H	R	E	N	R
   C19	H	S	H	E	H	S	E	E	E
   C20	H	S	H	H	H	H	E	R	S
   C21	H	S	S	H	H	H	E	H	E
   C22	H	S	S	H	H	S	E	H	S
   C23	H	S	S	H	H	S	E	S	N
   C24	S	S	S	H	R	H	S	E	S
   C25	H	S	S	H	H	H	E	N	E
   C26	H	S	S	H	H	S	E	S	E
   C27	H	S	H	H	H	H	E	E	R
   C28	H	S	H	E	H	H	E	N	E
   C29	S	S	S	H	R	H	E	R	E
   C30	H	S	H	H	H	R	E	R	E
   C31	H	S	H	H	H	S	R	E	E
   C32	H	S	H	H	H	E	E	H	R
   C33	H	S	H	H	H	S	R	R	E
   C34	H	S	H	H	H	S	S	R	S
   C35	H	S	H	H	H	S	E	H	S
   C36	H	S	S	H	H	H	E	E	E
   C37	H	S	S	H	H	R	E	N	R
   C38	H	S	H	H	H	S	E	H	N
   C39	H	S	H	H	H	S	E	R	N
   C40	H	S	S	H	H	S	E	R	S
   C41	H	S	H	H	H	S	E	S	H
   C42	H	S	S	H	H	R	E	H	H
   C43	H	S	S	H	H	S	E	H	H
   C44	H	S	S	H	H	S	E	H	E
   C45	H	S	S	H	H	S	E	E	H
   C46	S	S	S	H	R	H	E	H	E
   C47	H	S	S	H	H	H	E	R	S

   Table 5. Sample of the membership functions and crisp values for C1 under A1.

   	$N_{C1y}$	$R_{C1y}$	$E_{C1y}$	$S_{C1y}$	$H_{C1y}$	${\left({\stackrel{‾}{}}_{C1y}\right)}_c^{A1}$
   C1	0.000	0.000	1.000	0.000	0.000	3.000
   C2	0.000	0.000	0.100	0.400	0.500	4.400
   C3	0.000	0.000	0.700	0.300	0.000	3.300
   C4	0.000	0.000	0.000	0.700	0.300	4.300
   C5	0.100	0.000	0.100	0.300	0.500	4.100
   C6	0.000	0.100	0.300	0.100	0.500	4.000
   C7	0.000	0.100	0.200	0.500	0.200	3.800
   C8	0.000	0.000	0.100	0.700	0.200	4.100
   C9	0.000	0.100	0.300	0.100	0.500	4.000
   C10	0.000	0.000	0.700	0.200	0.100	3.400
   C11	0.100	0.000	0.700	0.200	0.000	3.000
   C12	0.000	0.200	0.100	0.300	0.400	3.900
   C13	0.000	0.000	0.800	0.100	0.100	3.300
   C14	0.000	0.100	0.100	0.500	0.300	4.000
   C15	0.000	0.000	0.700	0.200	0.100	3.400
   C16	0.000	0.000	0.100	0.400	0.500	4.400
   C17	0.000	0.100	0.200	0.300	0.400	4.000
   C18	0.100	0.200	0.200	0.100	0.400	3.500
   C19	0.000	0.000	0.400	0.200	0.400	4.000
   C20	0.000	0.100	0.100	0.200	0.600	4.300
   C21	0.000	0.000	0.200	0.200	0.600	4.400
   C22	0.000	0.000	0.100	0.400	0.500	4.400
   C23	0.100	0.000	0.100	0.400	0.400	4.000
   C24	0.000	0.100	0.100	0.500	0.300	4.000
   C25	0.100	0.000	0.200	0.200	0.500	4.000
   C26	0.000	0.000	0.200	0.400	0.400	4.200
   C27	0.000	0.100	0.200	0.100	0.600	4.200
   C28	0.100	0.000	0.300	0.100	0.500	3.900
   C29	0.000	0.200	0.200	0.300	0.300	3.700
   C30	0.000	0.200	0.200	0.100	0.500	3.900
   C31	0.000	0.100	0.200	0.200	0.500	4.100
   C32	0.000	0.100	0.200	0.100	0.600	4.200
   C33	0.000	0.200	0.100	0.200	0.500	4.000
   C34	0.000	0.100	0.000	0.400	0.500	4.300
   C35	0.000	0.000	0.100	0.300	0.600	4.500
   C36	0.000	0.000	0.300	0.200	0.500	4.200
   C37	0.100	0.200	0.100	0.200	0.400	3.600
   C38	0.100	0.000	0.100	0.200	0.600	4.200
   C39	0.100	0.100	0.100	0.200	0.500	3.900
   C40	0.000	0.100	0.100	0.400	0.400	4.100
   C41	0.000	0.000	0.100	0.300	0.600	4.500
   C42	0.000	0.100	0.100	0.200	0.600	4.300
   C43	0.000	0.000	0.100	0.300	0.600	4.500
   C44	0.000	0.000	0.200	0.300	0.500	4.300
   C45	0.000	0.000	0.200	0.300	0.500	4.300
   C46	0.000	0.100	0.200	0.300	0.400	4.000
   C47	0.000	0.100	0.100	0.300	0.500	4.200

   Table 6. Total relational matrix of the criteria.

   	C1	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	C12	C13	C14	C15	C16	…	C46	C47
   C1	0.604	0.595	0.591	0.596	0.598	0.594	0.598	0.593	0.600	0.592	0.595	0.597	0.593	0.592	0.598	0.594	…	0.589	0.581
   C2	0.638	0.621	0.623	0.624	0.629	0.625	0.629	0.622	0.630	0.621	0.627	0.626	0.626	0.623	0.627	0.624	…	0.619	0.609
   C3	0.604	0.592	0.585	0.594	0.595	0.591	0.597	0.590	0.596	0.587	0.594	0.594	0.591	0.590	0.595	0.592	…	0.586	0.579
   C4	0.624	0.612	0.607	0.608	0.614	0.611	0.614	0.608	0.617	0.608	0.614	0.611	0.611	0.610	0.615	0.610	…	0.605	0.597
   C5	0.605	0.595	0.590	0.592	0.592	0.591	0.597	0.590	0.597	0.588	0.595	0.593	0.594	0.592	0.595	0.594	…	0.588	0.579
   C6	0.609	0.597	0.593	0.597	0.600	0.593	0.599	0.595	0.602	0.592	0.600	0.598	0.598	0.595	0.600	0.597	…	0.592	0.584
   C7	0.627	0.615	0.609	0.614	0.616	0.613	0.615	0.611	0.619	0.610	0.617	0.616	0.616	0.611	0.620	0.614	…	0.609	0.600
   C8	0.615	0.602	0.599	0.600	0.604	0.599	0.606	0.596	0.605	0.597	0.603	0.603	0.602	0.598	0.607	0.601	…	0.596	0.589
   C9	0.615	0.604	0.598	0.602	0.602	0.600	0.605	0.599	0.603	0.600	0.604	0.603	0.603	0.597	0.606	0.602	…	0.597	0.588
   C10	0.622	0.611	0.606	0.611	0.611	0.610	0.614	0.607	0.614	0.602	0.612	0.610	0.609	0.608	0.613	0.609	…	0.605	0.597
   C11	0.624	0.611	0.604	0.612	0.613	0.609	0.615	0.608	0.616	0.607	0.610	0.612	0.611	0.610	0.614	0.609	…	0.605	0.599
   C12	0.629	0.616	0.612	0.617	0.619	0.614	0.622	0.615	0.620	0.613	0.619	0.613	0.616	0.614	0.621	0.615	…	0.610	0.602
   C13	0.630	0.619	0.614	0.618	0.620	0.616	0.623	0.615	0.623	0.612	0.619	0.618	0.614	0.616	0.620	0.617	…	0.612	0.604
   C14	0.618	0.606	0.602	0.605	0.608	0.605	0.607	0.601	0.611	0.600	0.607	0.605	0.605	0.600	0.609	0.603	…	0.600	0.592
   C15	0.623	0.611	0.605	0.610	0.612	0.610	0.613	0.609	0.617	0.606	0.612	0.611	0.610	0.607	0.610	0.609	…	0.605	0.597
   C16	0.628	0.616	0.612	0.616	0.619	0.616	0.620	0.614	0.620	0.612	0.620	0.617	0.617	0.615	0.619	0.612	…	0.611	0.601
   C17	0.608	0.597	0.589	0.596	0.598	0.595	0.598	0.593	0.599	0.591	0.597	0.597	0.596	0.592	0.599	0.594	…	0.591	0.582
   C18	0.622	0.610	0.605	0.609	0.612	0.607	0.613	0.608	0.613	0.606	0.612	0.611	0.609	0.608	0.614	0.610	…	0.604	0.596
   …	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…
   C44	0.634	0.620	0.616	0.620	0.623	0.619	0.624	0.618	0.626	0.616	0.622	0.620	0.621	0.617	0.624	0.621	…	0.613	0.606
   C45	0.617	0.605	0.601	0.606	0.606	0.603	0.610	0.604	0.609	0.601	0.606	0.606	0.606	0.603	0.609	0.604	…	0.600	0.591
   C46	0.612	0.598	0.594	0.597	0.600	0.597	0.602	0.596	0.602	0.593	0.601	0.599	0.599	0.596	0.601	0.598	…	0.589	0.584
   C47	0.643	0.630	0.626	0.632	0.631	0.629	0.634	0.627	0.635	0.625	0.632	0.630	0.630	0.627	0.633	0.627	…	0.622	0.611

   Table 7. Dependence and driving power of the criteria.

   	$d^p$	$d^r$	$\left(d^p+d^r\right)$	$\left(d^p-d^r\right)$
   C1	27.890	29.197	57.088	(1.307)
   C2	29.310	28.613	57.923	0.697
   C3	27.783	28.388	56.171	(0.605)
   C4	28.669	28.598	57.267	0.071
   C5	27.794	28.682	56.476	(0.887)
   C6	27.994	28.528	56.521	(0.534)
   C7	28.823	28.773	57.596	0.050
   C8	28.211	28.458	56.669	(0.247)
   C9	28.228	28.805	57.033	(0.577)
   C10	28.600	28.392	56.992	0.208
   C11	28.651	28.700	57.350	(0.049)
   C12	28.904	28.625	57.529	0.279
   C13	28.966	28.603	57.569	0.362
   C14	28.376	28.475	56.852	(0.099)
   C15	28.614	28.758	57.372	(0.144)
   C16	28.899	28.546	57.445	0.353
   C17	27.908	28.784	56.692	(0.876)
   C18	28.586	28.752	57.337	(0.166)
   C19	28.237	28.568	56.805	(0.331)
   C20	28.746	28.407	57.153	0.339
   C21	28.862	28.796	57.658	0.067
   C22	29.141	28.195	57.336	0.946
   C23	28.846	28.831	57.676	0.015
   C24	28.758	28.686	57.444	0.072
   C25	28.624	28.540	57.164	0.084
   C26	27.758	28.344	56.102	(0.586)
   C27	28.451	28.262	56.713	0.188
   C28	28.342	28.595	56.937	(0.253)
   C29	28.009	28.548	56.557	(0.539)
   C30	28.053	28.613	56.665	(0.560)
   C31	28.338	28.331	56.669	0.006
   C32	27.862	28.502	56.364	(0.640)
   C33	28.129	28.726	56.854	(0.597)
   C34	28.756	28.265	57.021	0.491
   C35	28.765	28.373	57.138	0.392
   C36	29.390	28.186	57.576	1.204
   C37	28.258	28.406	56.664	(0.149)
   C38	28.231	28.257	56.488	(0.026)
   C39	28.848	28.359	57.207	0.488
   C40	29.358	28.653	58.010	0.705
   C41	28.554	28.622	57.177	(0.068)
   C42	28.543	28.341	56.884	0.202
   C43	28.552	28.385	56.936	0.167
   C44	29.073	28.609	57.682	0.464
   C45	28.353	28.275	56.628	0.078
   C46	28.050	28.317	56.367	(0.268)
   C47	29.514	27.935	57.449	1.579

Figure 1. Cause and effect diagram of the criteria.

1. To calculate the aspects, Equation (10)(10) ${\left(F_{xy}\right)}_c^a=\sqrt[c]{\prod _{x=1}^c{\left({\overline{CP}}_{xy}\right)}_c^a}={\left(f_{xy}\right)}_{1\times c}^a$(10) must be adopted to obtain the measurement criticality as

   (1) ${\left(F_{C1C1}\right)}^{A1}=\sqrt[47]{\begin{array}{c}3.000\times 4.000\times 3.111\times 3.889\times 4.111\times 4.000\times 4.333\times 3.889\times 4.222\times \\ 3.222\times 3.000\times 4.111\times 2.889\times 3.778\times 3.333\times 4.000\times 3.889\times 4.000\times \\ 4.000\times 3.778\times 3.778\times 3.556\times 3.667\times 4.222\times 3.778\times 3.778\times 3.556\times \\ 3.778\times 3.889\times 4.222\times 3.889\times 3.444\times 3.778\times 3.778\times 3.778\times 3.556\times \\ 4.111\times 3.333\times 3.778\times 4.222\times 4.222\times 4.222\times 4.000\times 4.222\times 4.000\times \\ 4.000\times 3.667\end{array}}$(1)

$=3.786$. Once the measurement criticalities are obtained, Equation (11)(11) ${\left({\bar{F}}_{xi}\right)}_{\bar{c}}=\frac{{\left(F_{xy}\right)}_c^a}{{\sum }_{x=1}^{\beta }{\left(F_{xy}\right)}_c^a}={\left(f_{xi}\right)}_{1\times c},i=1,2,\cdots ,c$(11) is used to generate the factor weights, as shown in Table 8. Table 9 displays the total membership function by applying Equations (12)(12) ${\left({\overline{\overline{L}}}_{ij}\right)}_k={\left[\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(N_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(R_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(E_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(S_{xy}\right)}_c^a}}{a},\frac{{\displaystyle {\sum }_{a=1}^{\beta }{\left(H_{xy}\right)}_c^a}}{a}\right]}_{c\times k}={\left({\ell }_{ij}\right)}_{c\times k}$(12) and (13(13) ${\left(M_{xj}\right)}_{\bar{c}\times k}={\left({\bar{F}}_{xi}\right)}_{\bar{c}}\times {\left({\overline{\stackrel{‾}{L}}}_{ij}\right)}_k={\left(f_{xi}\right)}_{1\times c}\times {\left({\mathrm{\ell }}_{ij}\right)}_{c\times k}={\left(m_{xj}\right)}_{\bar{c}\times k}$(13) ). Then, Equation (14)(14) $M{ }^{\mathrm{\prime }}={\left(1\times m_{xj}^N+2\times m_{xj}^R+3\times m_{xj}^E+4\times m_{xj}^S+5\times m_{xj}^H\right)}_{\bar{c}\times k}={\left(m_{op}^{\prime }\right)}_{a\times a}$(14) can be used to generate the crisp values for the aspects, as presented in Table 10. Repeating Equations (5)(5) $D=\frac{{\sum }_{x=1}^c{\left[b_{xy}\right]}_{c\times c}^a}{a}={\left[d_{xy}\right]}_{c\times c}$(5) -(9(9) $d^r={\left[\sum _{y=1}^{c}d{\prime }_{xy}\right]}_{1\times \mathrm{c}}={\left[d{\prime }_y\right]}_{1\times \mathrm{c}}$(9) ) computes the total relation matrix (as shown in Table 11). Implementing Equation (15)(15) $\theta =\frac{{\sum }_{o=1}^a{m}_{op}^{\prime }}{a^2}$(15) identifies the interrelationships among the aspects. Subsequently, the influential diagram (as illustrated in Figure 2) is drawn for the aspects in accordance with Table 12.

Figure 2. Influential diagram of the aspects.

Table 8. Factor weights of the aspects.

		C1	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	C12	C13	C14	C15	…	C45	C46	C47
A1	${\left(F_{xy}\right)}_c^{A1}$	3.786	3.989	3.836	3.851	3.689	3.877	3.873	3.836	3.873	3.815	3.865	3.632	4.006	3.634	3.869	…	3.826	3.595	3.946
	${\left({\bar{F}}_{xi}\right)}_{\bar{c}}$	0.055	0.058	0.056	0.056	0.053	0.056	0.056	0.056	0.056	0.055	0.056	0.053	0.058	0.053	0.056	…	0.495	0.477	0.523
A2	${\left(F_{xy}\right)}_c^{A2}$	3.698	3.761	3.661	3.702	3.561	3.566	3.789	3.608	3.695	3.737	3.770	3.852	3.807	3.879	3.750	…	3.493	3.657	3.800
	${\left({\bar{F}}_{xi}\right)}_{\bar{c}}$	0.055	0.056	0.055	0.055	0.053	0.053	0.057	0.054	0.055	0.056	0.057	0.058	0.057	0.058	0.056	…	0.499	0.490	0.510
A3	${\left(F_{xy}\right)}_c^{A3}$	3.687	3.844	3.616	3.832	3.747	3.539	3.869	3.780	3.629	3.821	3.876	3.718	3.692	3.623	3.564	…	3.585	3.667	3.787
	${\left({\bar{F}}_{xi}\right)}_{\bar{c}}$	0.055	0.057	0.054	0.057	0.056	0.053	0.058	0.057	0.054	0.057	0.058	0.056	0.055	0.054	0.053	…	0.496	0.492	0.508
A4	${\left(F_{xy}\right)}_c^{A4}$	3.498	3.554	3.425	3.507	3.486	3.667	3.589	3.531	3.488	3.622	3.638	3.696	3.584	3.599	3.679	…	3.801	3.736	3.648
	${\left({\bar{F}}_{xi}\right)}_{\bar{c}}$	0.054	0.055	0.053	0.054	0.054	0.057	0.056	0.055	0.054	0.056	0.057	0.057	0.056	0.056	0.057	…	0.504	0.506	0.494
A5	${\left(F_{xy}\right)}_c^{A5}$	3.387	3.501	3.392	3.354	3.432	3.410	3.295	3.464	3.401	3.341	3.304	3.427	3.421	3.492	3.360	…	3.665	3.411	3.550
	${\left({\stackrel{‾}{}}_{\hspace{0.17em}}\right)}_{\stackrel{‾}{}}$	0.055	0.057	0.055	0.055	0.056	0.056	0.054	0.056	0.055	0.054	0.054	0.056	0.056	0.057	0.055	…	0.486	0.490	0.510
A6	${\left(F_{xy}\right)}_c^{A6}$	3.147	3.615	3.173	3.530	3.192	3.188	3.491	3.207	3.355	3.408	3.303	3.657	3.492	3.336	3.520	…	3.171	3.256	3.696
	${\left({\bar{F}}_{xi}\right)}_{\bar{c}}$	0.052	0.060	0.052	0.058	0.053	0.052	0.057	0.053	0.055	0.056	0.054	0.060	0.057	0.055	0.058	…	0.481	0.468	0.532

Table 9. Aggregated membership functions of the aspects.

	A1					A2					A3					$\cdots$
	$\frac{{\sum }_{a=1}^{\beta }{\left(N_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(R_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(E_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(S_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(H_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(N_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(R_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(E_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(S_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(H_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(N_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(R_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(E_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(S_{xy}\right)}_c^a}{a}$	$\frac{{\sum }_{a=1}^{\beta }{\left(H_{xy}\right)}_c^a}{a}$	$\cdots$
C1	0.021	0.090	0.217	0.407	0.265	0.024	0.095	0.319	0.277	0.286	0.035	0.109	0.293	0.243	0.319	$\cdots$
C2	0.007	0.045	0.270	0.274	0.404	0.012	0.090	0.284	0.338	0.277	0.026	0.038	0.296	0.331	0.310	$\cdots$
C3	0.040	0.057	0.199	0.416	0.288	0.047	0.085	0.293	0.293	0.281	0.071	0.085	0.267	0.293	0.284	$\cdots$
C4	0.012	0.054	0.236	0.449	0.248	0.017	0.113	0.305	0.267	0.298	0.033	0.054	0.260	0.336	0.317	$\cdots$
C5	0.071	0.092	0.215	0.303	0.319	0.057	0.126	0.271	0.277	0.270	0.035	0.061	0.307	0.296	0.300	$\cdots$
C6	0.050	0.095	0.182	0.258	0.416	0.054	0.125	0.288	0.248	0.284	0.045	0.109	0.331	0.277	0.239	$\cdots$
C7	0.000	0.059	0.253	0.430	0.258	0.002	0.095	0.307	0.288	0.307	0.024	0.040	0.305	0.293	0.338	$\cdots$
C8	0.024	0.043	0.241	0.442	0.251	0.035	0.149	0.277	0.234	0.305	0.031	0.071	0.281	0.303	0.314	$\cdots$
C9	0.021	0.083	0.232	0.312	0.352	0.038	0.109	0.277	0.255	0.322	0.031	0.104	0.312	0.298	0.255	$\cdots$
C10	0.021	0.073	0.234	0.395	0.277	0.012	0.095	0.310	0.300	0.284	0.024	0.052	0.293	0.326	0.305	$\cdots$
C11	0.031	0.057	0.191	0.440	0.281	0.000	0.116	0.293	0.281	0.310	0.014	0.061	0.274	0.322	0.329	$\cdots$
C12	0.007	0.158	0.284	0.288	0.262	0.000	0.073	0.293	0.326	0.307	0.031	0.087	0.281	0.319	0.281	$\cdots$
C13	0.007	0.043	0.189	0.442	0.319	0.007	0.080	0.312	0.281	0.319	0.031	0.095	0.324	0.239	0.312	$\cdots$
C14	0.054	0.085	0.241	0.392	0.227	0.005	0.061	0.288	0.329	0.317	0.047	0.106	0.286	0.281	0.279	$\cdots$
C15	0.043	0.052	0.177	0.426	0.303	0.017	0.073	0.317	0.317	0.277	0.066	0.097	0.284	0.303	0.251	$\cdots$
C16	0.014	0.064	0.239	0.270	0.414	0.007	0.095	0.312	0.293	0.293	0.024	0.045	0.329	0.314	0.288	$\cdots$
C17	0.076	0.104	0.177	0.234	0.409	0.038	0.137	0.279	0.270	0.277	0.050	0.116	0.305	0.248	0.281	$\cdots$
C18	0.033	0.104	0.234	0.227	0.402	0.078	0.111	0.267	0.286	0.258	0.028	0.090	0.303	0.298	0.281	$\cdots$
…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	…	$\cdots$
C37	0.033	0.064	0.281	0.227	0.395	0.050	0.109	0.317	0.265	0.260	0.052	0.099	0.277	0.274	0.298	$\cdots$
C38	0.073	0.083	0.246	0.128	0.470	0.035	0.071	0.279	0.312	0.303	0.066	0.118	0.286	0.248	0.281	$\cdots$
C39	0.052	0.087	0.243	0.116	0.501	0.040	0.092	0.286	0.291	0.291	0.066	0.102	0.303	0.310	0.220	$\cdots$
C40	0.014	0.066	0.208	0.317	0.395	0.033	0.106	0.217	0.324	0.319	0.033	0.092	0.300	0.326	0.248	$\cdots$
C41	0.000	0.066	0.246	0.317	0.371	0.040	0.095	0.303	0.260	0.303	0.038	0.095	0.317	0.281	0.270	$\cdots$
C42	0.007	0.073	0.265	0.241	0.414	0.033	0.083	0.303	0.310	0.272	0.033	0.092	0.317	0.300	0.258	$\cdots$
C43	0.031	0.102	0.210	0.291	0.366	0.057	0.116	0.310	0.248	0.270	0.031	0.114	0.318	0.277	0.260	$\cdots$
C44	0.050	0.064	0.187	0.312	0.388	0.064	0.128	0.312	0.217	0.279	0.054	0.090	0.281	0.298	0.277	$\cdots$
C45	0.057	0.071	0.208	0.293	0.371	0.066	0.161	0.246	0.255	0.272	0.052	0.085	0.303	0.329	0.232	$\cdots$
C46	0.028	0.151	0.262	0.305	0.253	0.050	0.090	0.279	0.298	0.284	0.040	0.080	0.300	0.317	0.262	$\cdots$
C47	0.028	0.071	0.203	0.281	0.416	0.009	0.118	0.232	0.331	0.310	0.026	0.052	0.310	0.319	0.293	$\cdots$

Table 10. Sample of the total membership functions and crisp values for A1.

	${\left(m_{xj}^N,m_{xj}^R,m_{xj}^E,m_{xj}^S,m_{xj}^H\right)}_{\bar{c}\times k}$					$m_{op}^{\prime }$
A1	0.029	0.075	0.223	0.356	0.317	3.857
A2	0.045	0.093	0.237	0.230	0.394	3.834
A3	0.048	0.081	0.242	0.202	0.427	3.880
A4	0.021	0.079	0.235	0.257	0.409	3.955
A5	0.053	0.067	0.197	0.303	0.380	3.888
A6	0.028	0.109	0.231	0.293	0.338	3.803

Table 11. Total relational matrix of the aspects.

	A1	A2	A3	A4	A5	A6
A1	6.757	6.388	6.426	6.443	5.417	5.996
A2	6.756	6.382	6.422	6.449	5.419	5.997
A3	6.762	6.389	6.427	6.449	5.421	6.002
A4	6.047	5.713	5.742	5.765	4.732	5.367
A5	6.810	6.427	6.470	6.500	5.474	6.038
A6	6.831	6.461	6.499	6.521	5.481	6.069

Table 12. Relation identification matrix of the aspects.

	A1	A2	A3	A4	A5	A6
A1	6.757	6.388	6.426	6.443	0.000	0.000
A2	6.756	6.382	6.422	6.449	0.000	0.000
A3	6.762	6.389	6.427	6.449	0.000	0.000
A4	0.000	0.000	0.000	0.000	0.000	0.000
A5	6.810	6.427	6.470	6.500	0.000	0.000
A6	6.831	6.461	6.499	6.521	0.000	0.000

5. Implications

This section provides the implications based on the analytical results to enhance understanding and bridge theory and practice.

5.1. Theoretical implications

In the influential diagram (Figure 2), carbon emission reduction (A5) and welfare improvement (A6) are the causal aspects of an SCE in the Indonesian case. A5 refers to the pollution of the environment caused by factories, forest degradation, and the burning of fossil fuels such as coal, oil, and natural gas, leading to global climate warming [54]. This causal aspect plays a significant causal role in influencing other effect aspects. These influences cover a strong direct relationship with stakeholder awareness (A1), medium direct relationships with sustainable production and consumption (A3) and energy savings and efficiency (A4) and a weak direct relationship with sustainable engagement (A2). However, A5 has no direct relationship with A6, which means that carbon emission reduction cannot generate an immediate effect on welfare improvement. The influence of A5 should be generated by exporting products using the most sustainable shipping route (C44) to enhance its effect.

Welfare improvement (A6) is considered for eradicating poverty, strengthening social integration, reducing social inequality and improving economic efficiency. Although its causal effect is not as high as that of carbon emission reduction (A5), the importance of A6 is more elevated than that of A5. It has two direct solid relationships with A1 and A4 and two medium direct relationships with sustainable engagement (A2) and sustainable production and consumption (A3). This causal aspect relies on better empowering female manufacturing workers [C47) to generate an impact. This is in line with the finding of 21, that gender gaps in labor practices may prevent an SCE launch from capturing opportunities to share value. These results show that when launching an SCE, welfare improvement can be considered an incentive to encourage labor to participate in SCE practices.

These points mentioned above differ from the findings of 24, namely, that an SCE needs to concentrate on the consumption of natural resources and hazardous and solid waste reuse in the case of China’s Liaoning Province. The obtained aspects can reinforce the theoretical basis of the SCE to enhance the explanation of phenomena in labor-intensive Southeast Asian countries. Moreover, these results reveal that stakeholder awareness (A1], sustainable engagement (A2) and sustainable production and consumption (A3) are the core problems in practicing an SCE. However, these problems are all effect aspects, and they cannot be directly addressed. These results enhance the theoretical basis and bridge theory and practice to guide the case firm to strengthen its SCE efficiently and effectively under resource limitations.

5.2. Managerial implications

Figure 2 reveals that reducing carbon emissions by exporting products using the most sustainable shipping route (C44) and better empowering female manufacturing workers (C47) are the driving criteria for Indonesian fashion firms to pursue an SCE. In other words, Indonesian fashion firms must prioritize these two criteria under limited resources. Doing so may enhance the probability of success for firms to realize an SCE.

Exporting products using the most sustainable shipping route (C44) is a crucial criterion for fashion firms. Sustainable shipment is related to (1) technological measures, (2) operational standards and (3) renewable energy utilization. Despite the drawbacks of using sustainable shipping, which takes longer in delivery, the benefits are more significant, including high flexibility and reliability, meaning that sustainable shipping can transport a large shipment of products across the globe or over a short distance. Sustainable shipping has lower insurance costs and higher safety compared to other shipping modes. Moreover, shipping vessels possess environmentally friendly technology that can reduce energy consumption. The most sustainable shipping route enables a reduction in resource wastes and a transformation of these wastes into resources to provide safe working conditions, ensure the welfare of employees and fulfill the expectations of the public.

Another driving criterion for Indonesian fashion firms is better empowering female manufacturing workers (C47). Sukkhacitta considers its workers to be an essential asset for realizing an SCE while conserving natural resources and the environment. Empowerment is a critical issue in Indonesian fashion firms; thus, the case firm strives to eliminate these issues by paying more attention to human rights and gender equality. For example, the case firm attempts to establish craft schools to provide continuous training for unskilled people, it ensures that wages reach the living standards throughout the supply chain, and it offers scholarships for younger women to learn craft skills. The owner devotes herself to raising awareness via these practices to start a revolution to embed the behaviors of producers, designers, retailers and buyers with 1482 Indonesian residents who are involved, with a 60% increase in their incomes.

However, enhancing credibility by obtaining awards for sustainability (C1), generating social impact by gaining media coverage (C11), reinvesting profits in villages (C15), recycling unused clothes to realize circular fashion (C18), and maintaining the uniqueness and authenticity of products by implementing handmade production (C41) are the core problems of Sukkhacitta. These core problems require the case firm to address them by enhancing the performance of C44 and C47.

6. Conclusion

With the improvement in Indonesian living standards, textile wastes from the fashion industry are increasing. These wastes have become a critical issue requiring the Indonesian government to address them. Therefore, the SCE concept has been proposed by previous studies to reduce consumption, emissions and waste efficiently and effectively. However, previous studies have focused on model development by utilizing qualitative approaches and literature reviews without bridging theory and practice. This study proposes a hybrid method to address the gaps in previous studies by presenting influential diagrams to reinforce the theoretical basis. This proposed hybrid method integrates EFA, the FSM and the DEMATEL to ensure validity and reliability. The FSM-DEMATEL enables us to transform expert opinions into crisp values and provide visual analysis to promote the understanding of the analytical results.

The analytical results highlight that an SCE needs to take into account six aspects: stakeholder awareness (A1), sustainable engagement (A2), sustainable production and consumption (A3), energy savings and efficiency (A4), carbon emission reduction (A5) and welfare improvement (A6). These discovered aspects not only promote the success of implementation but also reinforce the theoretical basis of the SCE. Furthermore, the proposed hybrid method enables us to address the gaps in previous studies by illustrating the complex interrelationships in visual diagrams to enhance understanding by balancing subjective and objective opinions to launch an SCE. The obtained visual diagrams also provide guidelines for related firms to improve their performance under limited resources.

The findings confirm that exporting products using the most sustainable shipping route (C44) and better empowering female manufacturing workers (C47) are the two primary decisive criteria for driving carbon emission reduction (A5) and welfare improvement (A6). These two aspects have important causal effects on the other effect aspects. Moreover, enhancing credibility by obtaining awards for sustainability (C1), generating social impact by gaining media coverage (C11), reinvesting profits in villages (C15), recycling unused clothes to realizing circular fashion (C18), and maintaining the uniqueness and authenticity of products by implementing handmade production (C41) are critical problems for the case firm Sukkhacitta in launching an SCE. Thus, the resources of the firm need to be invested in exporting products using the most sustainable shipping route and better empowering female manufacturing workers to generate the driving power to make improvements.

This study has several limitations. Although the criteria were proposed by reviewing the literature and consulting with experts, 47 criteria are still insufficient to cover the entire SCE concept. Future studies need to include as many criteria as possible to completely reflect the SCE concept. Moreover, the collected criteria are focused on Indonesian fashion firms, which means that the generalizability of this study may be lacking. Further studies can apply the proposed method and generate a comparison with other countries to enhance the generalizability of this study. Furthermore, the invited experts were all from the field of sustainable development and the CE, and no fashion industry managers or textile senior engineers were included. This may cause analytical results that cannot reflect the entire picture of the fashion industry in launching an SCE. Therefore, future studies must balance the field of experts to reflect the real situation of the industry [55].

Acknowledgments

This study was supported by Taiwan NSTC 111-2221-E-451-002-MY3. Thanks to the anonymous reviewers for their comments to improve the quality of this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Science and Technology Council, Taiwan [111-2221-E-451-002-MY3].

Notes on contributors

Chih-Cheng Chen

Dr. Chih-Cheng Chen is an assistant professor at the Department of Business Administration, MingDao University, Taiwan. Dr. Chen finished his Ph.D. program in the Department of Applied Economics at National Chung Hsing University, Taiwan (2002-2007). Currently, he serves as an Associate Editor at the Journal of Asia Pacific Business Innovation and Management (International Journal). His research fields focus on sustainable consumption and production, circular economy, sustainable development, corporate sustainability, and performance analysis. He published 39 papers with 575 citations.

Faza Muhammad Sukarsono

Mr. Faza Muhammad Sukarsono received his master’s degree from the National Taiwan University of Science and Technology in 2022. Currently, he is applying Ph.D. His research interests focus on sustainable circular economy, business management, and multi-criteria decision-making.

Kuo-Jui Wu

Dr. Kuo-Jui Wu (Gary) is a professor at Hainan University. Currently, he is the subject editor of the Journal of Sustainable Production and Consumption, the general secretary of the International Society for Business Innovation and Technology Management (ISBITM), and a reviewer of several SCI/SSCI journals. He acquired a master's degree from De La Salle University, the Philippines in 2011, and a Ph.D. degree in industrial management from the National Taiwan University of Science and Technology in 2016. His research fields focus on multi-criteria decision-making (MCDM), circular economy, sustainable development, circular/sustainable supply chain management, and corporate sustainability. He published 72 SCI/SSCI papers with 3160 citations (ESI: 1, H-index: 31, i10-index: 55).