1. Introduction
The implementation of I4.0T in the textile production system has stimulated micro-level circular economy (CE) practices by considering cleaner production (CP) as an essential tool [
1,
2,
3,
4]. This phenomenon occurs mainly due to the textile industry being environmentally unfriendly, such as high consumption of energy, water, and chemicals [
5], as well as generating a lot of waste at the end of the useful life of textile items [
6]. Thus, CECPs are complementary to the adoption of I4.0T [
7,
8]. It is emphasized, based on UNEP, that CP, since its existence, already sought to close the cycle of the production system by not generating waste and promoting recycling to save raw materials, water, and energy [
9]. Thus, CE originates from CP [
10,
11]. However, CE emerged in a radical way to eliminate linear flows and deployment of cyclical flows of productive resources aimed at the regeneration and use of materials from renewable sources [
12].
As mentioned, the adoption of I4.0T in manufacturing stimulates CECP. For example, big data provided material regeneration through end-of-life clothing repair [
13]; autonomous robots generated technological improvements in wastewater treatment and waste collection [
14]; simulation favored post-consumer waste collection [
15]; IoT promoted a circular business model of online apparel commerce [
13,
16]; Cloud Computing promoted renewable energy and dematerialization of products [
13]. Additive manufacturing stimulated the use of by-products as raw materials and waste reduction [
14]; augmented reality reduced waste by incorporating digital tools for selling and manufacturing apparel [
17]. Cyberphysical Systems enabled energy and water optimization [
18]; Cybersecurity Systems provided confidence to managers to share information between companies, promoting industrial symbiosis [
6,
13]; artificial intelligence played an important role in resource optimization [
19].
It is emphasized that successful CECP implementation contributes to sustainability [
20]. The definition of Strong Sustainability recognizes that the economy is an integrated subsystem within the system that includes society and the environment, which means that economic prosperity must be achieved without compromising social well-being and the health of the planet Oliveira Neto et al. [
21]. Specifically, CECP can contribute to SS in terms of increasing resource consumption efficiency, limiting the consumption of renewable resources to their regeneration rate, reducing greenhouse gas emission, reusing waste as inputs in other processes, replacing toxic inputs with organic materials and non-replenishable energy resources with renewable ones, increasing access to commodities and sustainable manufacturing [
21].
Thus, the textile industry can promote SS by installing more efficient engines and machines [
18], producing dye pigments from by-products of various local, rapidly regenerating crops such as onion husks, artichoke leaves, walnut husks, and chestnut shells [
22]; minimize material disposed of in landfills, which reduces methane gas and CO
2 emissions [
23]; using textile waste as inputs in other sectors such as thermal and acoustic insulation in cars, in construction, coverings in agriculture and paper industry, filling of toys and seats of chairs and armchairs [
6]; commercialize 100% biodegradable and non-toxic sewing threads [
2]; explore the use of renewable energy [
24]; donate clothing in good condition to charities [
6]; produce bio-based materials that refer to sustainable manufacturing, lower environmental impact and better livelihoods of farmers, which causes positive socioeconomic impact [
19].
It was identified only exploratory studies on the relationship between I4.0T and CECP in the textile industry. Evidence of SS was highlighted in the content analysis of the articles. Silva and Morais [
15] (2022) performed a simulation to classify circular strategies in textile waste management outsourcing in Brazil. They considered the use of recycled raw materials. Kumar et al. [
18] analyzed smart environmental management practices and CECP based on simulation technologies, augmented reality, and Cyberphysical Systems aimed at minimizing pollution in the supply chain of Pakistan. Agarwal and Singh [
25] conducted a simulation that evaluated the performance of textile effluent treatment techniques in India. It showed improvement in water circularity and elimination of hazardous products. Shayganmehr et al. [
13] evaluated I4.0T for the implementation of CECP practices and the readiness of the Iranian industry. Bai et al. [
19] analyzed I4.0T related to sustainable performance in China, the main features of which were optimizing resources and increasing efficiency. Tsai [
24] used a combination of simulation and sensing to perform CECP based on green production planning and control in Taiwan. Tsai considered industrial symbiosis of by-product exchange. Bloomfield and Borstrock [
26] investigated the challenges of additive manufacturing to promote CECP in the UK and resource efficiency.
Evidence of a relationship between I4.0T, CECP, and SS was also verified by Jin et al. [
16], who investigated circular business models based on product–service systems that use IoT in apparel companies targeting sustainable manufacturing. Chen et al. [
27] proposed integrated strategies based on big data and IoT to promote green chemistry in CECP and eliminate toxic products. Shirvanimoghaddam et al. [
6] discussed the use of IoT, additive manufacturing, Cyberphysical and Security Systems, and artificial intelligence for innovation and circular strategies in textile fashion and promote the exchange of by-products between companies. Jia et al. [
28] mentioned IoT and Cyberphysical Systems in their survey of circular economy barriers and practices in the textile and apparel industry; they highlighted the reduction of pollutant gas emissions. Faria et al. [
29] analyzed mobile applications that use IoT and cloud computing to support online apparel commerce with circular economy features and facilitate access to second-hand clothing. Alcayaga et al. [
30] investigated big data, IoT, and Cyberphysical Systems in smart circular systems at the micro-production level, in which they stimulate sub-product exchange between firms.
Also, Angelis and Feola [
22] addressed big data and Cyberphysical Systems in a case study in Italy on the characteristics of circular business models. They mentioned actions toward the use of non-toxic organic inputs. Franco [
2] conducted a multi-case study in Austria and Italy on circular production based on IoT and Cyberphysical and Security Systems, which use rapidly regenerating renewable resources (Fatimah et al. [
23]). A study applied a focus group to develop a sustainable and smart waste management system in Indonesia that minimizes methane gas emissions. Larsson [
17] conducted action research that evaluated four augmented reality projects in Sweden, focusing on digital sales and manufacturing tools that minimize waste in the textile and apparel value chain. Oliveira Neto et al. [
14] evaluated by survey the degree of sustainable resilience by the size of the textile industry to propose incremental changes from CP to CE, briefly mentioning additive manufacturing in product design and autonomous systems that improve working conditions. In this context, no confirmatory study was identified that has evaluated whether the implementation of I4.0T stimulates CECP actions in large textile industries located in Brazil, as well as whether the stimulation of CECP actions directs SS. Thus, based on this research gap, this study will answer the following question: Does the adoption of I4.0T supports CECP by promoting SS actions in large textile industries located in Brazil?
Thus, the objective of this study is to evaluate whether the adoption of I4.0T promotes CECP aimed at SS actions in large textile industries located in Brazil. This study presents a relevant theoretical contribution because it shows the I4.0T that promotes CECP actions, directing the SS. In addition to elucidating for business managers, the opportunity to invest in I4.0T to generate circularity in the production system is an important aspect of the fulfillment of the agenda for 2030.
4. Results and Discussion
The indicators of the measurement model were obtained (AVE, R2, R2adj, Cronbach’s Alpha, Composite Reliability), but the discriminant validity between the two constructs (SS and CEE) was not verified. For this purpose, the bivariate correlations of Pearson between all variables in the model were calculated. After evaluating the correlation values, the technology variable T_5 (Cybersecurity) was eliminated. The textile industry experts mentioned that the adoption of cloud computing generated the need for investment in Cybersecurity and justified that employees need training on information care in the digital transformation era. It is emphasized that in the correlation, it was found that the adoption of Cybersecurity is of little importance. Thus, large textile industries in Brazil have worked on the implementation of technologies essential for their operation without emphasizing the protection of intellectual property and risks to information systems. A significant increase in the use of information technology and connected systems in the textile industry, which includes process automation, production monitoring, supply chain management, and online sales, may encourage the industry to increase technology in defense against cyber threats such as hacker attacks, data theft, operational disruption, and industrial espionage. This result is at odds with the finding found in the case study conducted in the textile industry in Iran, which showed that I4.0 system security was one of the enablers with the strongest effect on CECP deployment [
13]. In this regard, large textile industries in Brazil are lagging in employing security technologies, such as implementing firewalls, intrusion prevention and detection systems, data encryption, and network monitoring. The mentioned deficit can inhibit business with large corporations in the sector.
In a new round of the model, after the removal of variable T_5, variables SS_1 and SS_4, increase efficiency in resource consumption and reuse waste as inputs in other processes, respectively, were also extracted from the model for not presenting significant values. SS_1, which aims to increase efficiency in the consumption of resources, was removed because large textile industries located in Brazil have difficulties in optimizing energy, water, and production inputs because the machines are old and the manufacturing park has a low level of automation, which hinders the efficient use of production resources. For example, machines are turned on during idle periods, which reduces the efficiency of energy consumption. This also occurs with water, which is used on a large scale in dyeing. This finding is in line with what was mentioned by Silva and Morais [
15] about the textile industry in developing countries whose operation is based on rudimentary low-efficiency technologies. The textile industry in Brazil faces significant challenges regarding the low efficiency level in the use of electricity, water, and other natural resources because it uses obsolete equipment and systems that consume a lot of energy and water, reuse of textile fibers, and waste of raw materials in production. In this sense, this research contributes to the theory by offering information on low efficiency in the use of resources in the textile industry in Brazil, which can be considered in future studies. In addition, managers can take the information from this study as a reference to develop appropriate strategies to raise the efficiency of operations, requiring investment in production machinery with lower energy consumption, water recirculation and treatment systems, and the adoption of circular economy practices, such as recycling textile waste. Furthermore, the research can make society aware of the importance of sustainability in this sector and its environmental and social impacts.
Also, the SS_4 that mentions the reuse of waste as inputs in other processes is not applied in the textile industry because it still lacks the integration more effectively with sewing workshops to reuse the scraps in the manufacture of products with their seams. The specialist mentioned that this is a very select public that buys textile articles made with scraps, such as bed sheets and bedspreads. Thus, the textile industry prefers to shred the waste and transform it back into cotton and thread for manufacturing in the company itself. This finding is corroborated by the article by Oliveira Neto et al. [
14], which concluded that the reuse of defibrated waste generates relevant economic and environmental gains, besides promoting circularity of more than 95%, denoting the reason why the textile industry prefers the reuse of textile waste internally instead of reusing it in other processes.
After re-spinning, the model discriminant validity was met.
At the end of the software calculation process, it was obtained values of the average variance extracted (AVE) greater than 0.50 (or 50%), indicating that the model meets the criteria of Fornell and Larker [
51]. Also, Cronbach’s alpha coefficients and Composite Reliability were above 0.70. Similarly, R2 and R2_adj (adjusted) show values considered large [
51] (
Table 2).
Finally, the effect size (explained by unexplained part—f2 = R2/(1 − R2)) “total” (from TI4—beginning of the model to all other constructs) is greater than 0.35, indicating relevant, i.e., it has great importance for the overall fit and its causal relationships.
Following the model fit, the square roots of the AVEs were calculated and compared with the correlations between the constructs to assess discriminant validity (DV) using the Fornell–Larcker criterion. This procedure is performed because the AVEs are the coefficient of (mean) Pearson determination or the mean squared correlations. Comparing the average correlation of a construct with the correlation with the other constructs assesses the independence of the construct from the others. Thus, to meet this requirement (“DV”), it is expected that the square roots of the AVEs will be higher than the correlations with the other constructs (
Table 3).
The analysis of
Table 3 shows that the Fornell–Larcker criterion was met in all correlations, only between CEV, which relates virtualization and CEE and exchange (in gray in
Table 3). The relationship gets higher in CEE of 0.003. Since this value is low, the decision was to leave the model without its removals. Also, the Heterotrace–Monotrace matrix was evaluated (
Table 4) [
53,
54]. The non-presence of the value one (1) between the minimum and maximum values of the confidence intervals of the causal relationships also indicates that the model exhibits discriminant validity.
Furthermore, the factor loadings or correlation of Pearson coefficients (loading) (
Table 4) showed high and significant values (
p ≤ 0.05). This fact shows that the model was very well adjusted and that its variables are adequate to evaluate the constructs.
Finally, the analysis of causal relationships indicates that the path coefficients (or angular coefficients of the linear regression lines or betas) have values above 68%.
Table 5 shows the path coefficients to evaluate the hypotheses. With this, H1 was validated, considering that the adoption of I4.0T by large textile industries moderately (0.68) promotes CE at the micro-production level. This finding is justified because despite loadings above 0.70 of I4.0T in the model fit, three of them (additive manufacturing (T_6), Cybersecurity System (T_9), and artificial intelligence (T_10) showed correlation with no variable (
Table 6), which directly reflects the lack of use of 3D technology for product development and redesign for resource efficiency (dematerialization) (CEV_1:0.54), which also showed no correlation. Additive manufacturing is not yet planned in textile production but could be applied in the apparel chain to promote personalization and personal style, even if with a higher added value, corroborated by Shirvanimoghaddam et al. [
6] and Shayganmehr et al. [
13]. While the adoption of the Cibersecuty System and T_5—cloud computing has a high cost, as mentioned earlier, and artificial intelligence consists of a later stage of implementation because, currently, the large textile industry is in the process of analyzing big data and applying machine learning techniques and have not yet applied autonomous systems based on artificial intelligence algorithms. This finding corroborates with Bai et al. [
19] and Shayganmehr et al. [
13], who identified the most relevant I4.0T for CECP that include Big Data and machine learning, with Cybersecurity and additive manufacturing in less significant positions due to acquisition cost. This result provides the basis for future studies to expand scientific knowledge in this field and enhance existing theoretical models because the incorporation of big data and machine learning in the Brazilian textile industry brings significant advantages. Managers of Brazilian textile companies can benefit from the information in this study when implementing additive manufacturing, Cybersecurity, and artificial intelligence to ensure a comprehensive approach and maximize the transformation potential and competitiveness of the Brazilian textile industry. In addition, the use of these technologies can generate positive impacts on society, such as reducing the consumption of natural resources, improving working conditions, producing more sustainable products, and promoting economic growth.
And other three technologies showed correlation with at most two variables, which are Simulation (T_3:0.73), IoT (T_4:0.69), and Cyberphysical System (T_8:0.86), denoting early stage of deployment in the textile industry because it is still little deployed Cyberphysical System to promote CE, which includes sensors and actuators, which are simpler technologies used in automation processes. In addition to low deployment with a focus on IoT CE, a primary aspect for managers is to use embedded technology for data-driven decision-making. Also, simulation is still little used to stimulate CE in production, denoting the low number of correlations; it only stimulated emission level reduction/elimination (CEO_5:0.90) but limited renewable resource consumption to its regeneration rate (SS_2:0.50) and reduced greenhouse gas emissions (SS_3:0.51). However, this finding is different from that found in the literature, denoting that there is potential for textile companies to leverage the use of simulation, IoT, and Cyberphysical Systems targeting economic and environmental gains provided by CECP practices, as was identified in the research of Shayganmehr et al. [
13], Alcayaga et al. [
30], and Bai et al. [
19]. Thus, for future studies, it is important to study the application of simulation, IoT, and Cyberphysical Systems in the context of Brazilian textile industries. The practical contribution is to stimulate managers to explore the potential of IoT, simulation, and Cyberphysical Systems aimed at CECP practices, resulting in SS actions. Furthermore, the implementation of these technologies can generate qualified employment opportunities and boost the technological and economic development of the country.
Summarizing only three (30% of I4.0T) showed correlation with several variables, considering autonomous robots (T_2:0.88), augmented reality (T_7:0.80), and big data analytics (T_1:0.79). The adoption of autonomous robots positively influences the ability to operate independently by performing specific tasks, such as material handling, waste collection, sorting, and recycling, in a precise manner. This finding corroborates Alcayaga et al. [
30], who indicated the use of robots in textile remanufacturing processes. Also, augmented reality has been used in the textile industry in Brazil in the design review of production processes and product prototypes and in employee training and capacity building. Larsson [
17] corroborates the use of augmented reality to attract customers to redesign clothes instead of buying new ones. However, Bai et al. [
19] mention that the application of augmented reality in the textile industry is understated. In this sense, this controversy indicates the opportunity for confirmatory research on this topic. In addition, big data analytics has been increased in the textile industry located in Brazil to collect, analyze and interpret data for more assertive decision-making in intelligent strategies. Chen et al. [
27] indicated the use of big data analytics to define green chemistry strategies in the textile industry. Importantly, the number of textile industries located in Brazil that use autonomous robots, augmented reality, and big data is growing, denoting contribution to the theory. In practical terms, this study can encourage managers to consider the economic gains obtained through CECP practices in feasibility studies of the acquisition of I4.0T, despite the investment cost, the need to restructure production processes and personnel training. For society, this study shows technological advances that reduce environmental impact and risk to human health.
Also, it was found that the CECP actions present in large textile industries located in Brazil, resulting from the adoption of I4.0T, generated weak SS (H2:0.43). This result can be explained because two important variables (SS_1—increased efficiency in the consumption of resources and SS_4—reuse of waste as inputs in other processes) were excluded in the model adjustment, mainly because the large textile industries located in Brazil have an old manufacturing park, which makes it very difficult to optimize production resources and interact little with external companies to produce products with the remnants of fabrics, preferring to shred and transform into cotton again for manufacturing. This finding agrees with Larsson [
17], who emphasized the need for textile chains to raise resource use efficiency, and Silva and Morais [
15], who revealed the need to seek sustainable alternatives to cotton in Brazil. In this sense, the opportunity arises for studies to propose the use of I4.0T that optimize resources and promote the reuse of waste as inputs, aiming to increase SS, an important aspect for organizational practice as well, in addition to stimulating the awareness of society, the development of policies and the adoption of sustainable practices in industries.
Although the other variables of SS present loading above 0.74 in the adjusted model it cannot be considered SS because all these variables should have been supported, as guided by Oliveira Neto et al. [
21]. Thus, the results are limited to consuming renewable resources (energy and materials), reducing greenhouse gas emissions, replacing hazardous products with organic ones, and increasing access to basic products to reduce hunger to promote sustainable manufacturing.
Table 6 shows the Pearson correlation that details the correlations of each technology in relation to the actions of CECP and SS, making it possible to explain the results.
It was found that the most relevant I4.0T are autonomous robots (T_2:0.88) that drove 14 CECP actions, augmented reality (T_7:0.86), promoting 8 CECP and Cyberphysical System (T_8:0.86), generating 1 CECP.
With the implementation of autonomous robots in textile production, it began to reduce/eliminate waste generation (CEO_4), as well as recycle and reuse waste from textile articles and packaging (CEL_1, CEL_4), including the adoption of the tracking system of post-consumer textile waste at the end of useful life (CEL_3, CERE_4) with the use of reverse logistics (CEL_5). Thus, the textile industry started using reusable, recyclable, and biodegradable products, improving product quality (CERE_1), limiting resource use, and promoting regeneration (SS_2), contributing to sustainable manufacturing (SS_8). Oliveira Neto et al. [
14] stated that autonomous robot contributes to the resilience of textile companies in the use of water, raw material, and input in sorting material for recycling, as well as improving working conditions. Also, Shayganmehr et al. [
13] indicated that the use of this technology in the textile sector contributes to the achievement of sustainable manufacturing goals.
It was also found that the adoption of autonomous robots in the effluent treatment system is able to extract polluting residues from the water used in the finishing, dyeing, or washing process to promote closed cycle (CERE_3), optimizing water consumption (CEO_3), as well as using natural dye instead of chemical (CERE-5). Through this, it contributed to SS_5 by replacing toxic inputs with organic materials, contributing to sustainable manufacturing (SS_8). Bai et al. [
19] mentioned that autonomous robots contribute to reducing the impact on effluents and waste generation, which reduces environmental impact and improves the livelihoods of farmers, which causes a positive socioeconomic impact.
Another finding is that the autonomous robot optimized energy consumption in the textile manufacturing process (CEO_2), including using renewable energy or cogenerating energy (CERE_2), replaces non-renewable energy resources with renewable alternatives (SS_6), favoring sustainable manufacturing (SS_8). Bai et al. [
19] and Shayganmehr et al. [
13] indicated that autonomous robot improves manufacturing performance through energy recovery. In addition, autonomous robots in textile production reduce emission levels (CEO_5), reduce greenhouse gas emissions (SS_3), and have a positive effect on reducing air pollution in the local community (SS_8). Oliveira Neto et al. [
14] mentioned that autonomous robot also contributes to the high degree of resilience of companies in reducing the emission of pollutant gases.
Thus, this study sheds theoretical light to the knowledge about the relationship between autonomous robots and CECP practices, mainly focused on looping, optimizing, and regenerating the SS actions of replacing non-renewable sources of resources, reducing pollutant gases emission, and promoting sustainable manufacturing. Also, these findings may stimulate managers to adopt autonomous robots, as well as help overcome the acceptance barrier of this technology by machine operators, who see the robot as a threat to their jobs.
With the adoption of augmented reality, the large textile industry has created a department to design new business models for the recovery, reuse, and recycling of textile waste (CES_1), including in the creation process, without the need to develop physical prototypes, using only virtual data. Thus, it uses design for the virtual environment for the development or/and redesign of products and packaging, increasing durability and eliminating the use of toxic substances in the product and production system (CEO_1). It integrates suppliers for developing and selecting renewable and recyclable materials (CEE_3), substituting hazardous materials (CEE_2) and limiting the consumption of renewable resources to their regeneration rate (SS_2), as well as substituting toxic inputs for organic materials in product development (SS_5).
The use of augmented reality in manufacturing by the large textile industry has also directed efforts to online support services for machinery and equipment maintenance (CEV_4), as well as for virtualization of textile articles for customers (CEV_2). It is noteworthy that the development of textile articles based on the design for the virtual environment applied in product development and redesign allows the integration of augmented reality for virtual sales catalogs to customers. With this, it contributed to SS because it limits resource consumption by using virtualization, promoting system regeneration (SS_2).
Also, using augmented reality in product/production design replaced inputs (raw materials, water, and or energy) to increase the share of renewable and recyclable resources (CEE_1), generating a reduction in the consumption of non-renewable resources (SS_2), as well as replacing non-renewable energy resources with renewable alternatives (SS_6). Larsson [
17] indicated that augmented reality was applied in a project supporting sustainable development in apparel value chains without extracting virgin natural resources or using toxic products, as well as showed weak evidence of virtualization.
No study was identified that confirms or shows a robust case study on the adoption of augmented reality in the textile industry, especially with a focus on CECP and SS. Thus, this study is the first to confirmatively emphasize that augmented reality is a facilitator of optimization, sharing, virtualization, and exchanges in the textile industry. With this, it is an important result and can shed light on future research. In this context, the implementation of augmented reality generates a high investment, but this study, in view of the gains with sustainability, can create incentives for its adoption. Additionally, promoting knowledge of services and products that improve the well-being of people to society.
The application of big data was important for the design of the environment in the development or/and redesign of the product/production and packaging aiming at recycling, repair, remanufacturing, maintenance, as well as developing new features for durable products and eliminating the use of toxic substances (CEO_1), changing to organic materials, for example, using natural paint extracted from waste from other manufacturing processes (artichoke) instead of using chemical dye (CERE_5, SS_5). The studies by Bai et al. [
19] and Fatimah et al. [
23] stated that big data technology enabled the reduction of toxic materials and less impact on effluents and waste, but without deep exploration, principally with the lens of SS.
Also, with big data analytics, it was possible to promote the recycling of monomers, oligomers, polymers, fibers, and fabrics (CEL_1), in addition to replacing production inputs (raw material, water and/or energy) to increase the share of renewable and recyclable resources (CEE_1) and used by-products from other processes for manufacturing (CERE_5), reducing emissions (CEO_5). In doing so, it limited the consumption of non-renewable resources (SS_2), replaced non-renewable energy resources with renewable alternatives (SS_6), reduced greenhouse gas emissions (SS_3), and replaced toxic inputs with organic ones (SS_5). Angelis and Feola [
22] indicate that big data supported the production of dye pigments from by-products of various local, rapidly regenerating crops such as onion husks, artichoke leaves, walnut, and chestnut shells. It is emphasized that only this qualitative evidence was identified, without a confirmation, mainly not referring to SS. Thus, this study advances the knowledge of the potential of big data analytics to support CECP and SS practices, especially those aimed at reducing emissions, non-renewable resources, and toxic inputs. With digital transformation, textile industries are adopting technologies on machines, generating a large volume of data, and pushing operations managers on the role of the data scientist for this purpose. The role even of the data scientist goes beyond economic analysis; he needs to be concerned about the risks of the use of chemicals in textile production and the excessive volume of textile waste generated at the end of life for society.
The adoption of IoT (T_4) and Cyberphysical System (T_8) promoted the use of digital technician service for system configuration and online support service for machine and equipment maintenance (CEV_4), limiting the consumption of renewable resources to their regeneration rate (SS_2), as well as, reducing greenhouse gas emissions (SS_3). Tsai [
24] presented the automatic monitoring system, based on IoT and Cyberphisical Systems, which records real-time data from the production process which encourages companies to control carbon emissions. Fatimah et al. [
23] concluded that the intelligent waste management system has a regeneration capacity commensurate with the rate of waste generated. Although this research mentions the subject, they do not confirmatively explore the findings. Thus, this study confirms that virtualization promoted by IoT and Cyberphysical Systems reduces greenhouse gas emissions and limits resource consumption, providing a baseline that can be explored by investigations in other sectors to compare the benefits of virtualization and the technologies employed. The information raised also contributes to practice because it can stimulate the shift from physical processes to virtual environments, which reduces environmental impact and costs. In addition, society can benefit by having easier access to services and products through virtualization.
As previously mentioned, SS_4 was extracted in the model adjustment, which addresses the reuse of waste as inputs in other processes. In the Pearson correlation analysis (
Table 6), some variables recognized as important but did not generate correlation were found, which explains why SS_4 was excluded from the model, which is CES_2_3: 0.92 related to cooperation of sewing workshops to recover textile waste and other companies to market by-products with the fabric waste. Thus, the textile industry does not yet reuse materials (recycled fabrics, fiber) for new purposes, such as insulation, blanket, cloths, and others (CEL_2:0.88), and with that, it does not establish digital design/stylist service provision for collection creation and material development, cutting and modeling (CEV_3:0.85). With this, the lack of correlation of CES_2_3 denotes the lack of policies to generate jobs, reduce the informality of the textile sector, and not contribute to access to basic products (SS_7:0.87). It is noteworthy that SS_7, although important, did not generate a correlation in the analysis, a worrisome aspect. This finding is in line with what was presented in the studies of Silva and Morais [
15] as well as Oliveira Neto et al. [
14] carried out in the Brazilian textile industry, denoting the absence of industrial symbiosis focused on the exchange of materials and especially not mentioning SS aiming to increase access to basic products. The findings indicate that there is negligence by textile companies towards access to basic products for people, which may be driven by several factors, including excessive focus on profit, pressure for low prices, lack of supply chain transparency, and lack of awareness about corporate social responsibility, denoting an innovative theoretical contribution. Thus, industrial managers can use this information to seek industrial symbiosis networks in which participating agents benefit each other, whether recycling cooperatives or industries in other sectors, to optimize the use of closed-loop materials and avoid the consumption of virgin raw materials. For society, this study contributes by highlighting the importance of public authorities and companies taking measures to facilitate access to basic products for people in need, which can return as a gain in the brand image of the company.
5. Conclusions
This work found that the adoption of I4.0T promotes CECP; however, it neglects SS actions in large textile industries located in Brazil, denoting theoretical originality, in addition to guiding operational managers on I4.0T that promotes CECP, generating SS enables the development of sustainable strategies in operations.
The results highlighted the technologies of big data, augmented reality, and autonomous robots as the most significant enablers for CECP practices. Big data provides processed information for supply chain tracking and management, demand forecasting and inventory management, sustainable product design, recycling and waste management, and product customization and sharing. Data-driven decision support contributes to building a more circular and responsible textile industry. Augmented reality improves the consumer experience, extends product life, reduces waste, and encourages recycling and proper disposal practices. Thus, augmented reality shows itself as a promising technology to drive the circular economy in the textile industry and contribute to a more sustainable sector. Also, autonomous robots perform waste collection and segregation tasks, automated maintenance and repair, efficient transportation, and material recovery.
Another highlight refers to the low impact of I4.0T on SS actions aimed at increasing resource consumption efficiency, the reuse of waste as inputs in other processes, and increasing access to commodities. The lack of these actions can be attributed to several factors, including a lack of awareness, competitive pressures, and lack of adequate regulation. Many managers still do not fully understand the environmental and economic benefits obtained through sustainable technologies and methods, which reflects in the continuity of traditional and not very efficient practices. In addition, competitive pressures may discourage textile companies from investing in waste efficiency and reuse measures, prioritizing cost reduction and profit maximization over the implementation of sustainable practices. These companies may fear that investing in more efficient technologies will generate additional costs and put them at a disadvantage compared to competitors that do not adopt sustainable measures. Another factor is the lack of clear and strict government policies and international standards for SS actions, which allows some companies to neglect their social and environmental responsibilities.
To move in this direction, it is crucial that there is a collective effort of public authorities, companies, and society in general. Governments should implement regulations and policies that encourage sustainable and responsible practices in the textile industry. Companies should take responsibility for adopting sustainable measures, even in a competitive environment, recognizing the long-term benefits for the environment, people, and their own reputation. And society, in general, should demand more sustainable textile products by supporting companies that adopt responsible practices.
This study conducted a survey of large textile companies located in Brazil, which limits the power of generalization of the results. In this sense, future research in other countries and industrial sectors is recommended as a basis for comparison of the findings of correlation between I4.0T, CE practices, and SS actions, which will contribute to the dissemination of knowledge on this topic.