A Study of the Drivers of Decarbonization in the Plastics Supply Chain in the Post-COVID-19 Era

Abstract

Plastics are an important basic material for national economic development. In the post-COVID-19 stage, green supply chain management has attracted widespread attention. In order to achieve carbon neutrality in the plastics industry, we explored the drivers of supply chain decarbonization in the plastics industry from a microlevel corporate supply chain perspective. Four primary factors and 21 subfactors were identified from the existing literature, and after validation by 12 experts, the causal relationships between the factors were analyzed using the Gray-DEMATEL method. The Gray-DEMATEL method was applied to analyze the causal relationships between the factors. The findings show that joint promotion by stakeholders is the most significant cause driver and market impact is the most prominent driver in the first-level indicator, both of which have a significant impact on low-carbon production. “Process optimization”, “Top-management support”, “Government regulations and support”, and “Information disclosure” are the most significant cause secondary drivers under the corresponding Tier 1 indicator factors, respectively, to provide realistic guidance for companies engaged in the plastics industry to continue to develop a low-carbon circular economy to achieve net-zero emissions under the challenges of COVID-19. Therefore, companies need to focus on the drivers of most importance in this work and understand the interplay between factors.

1. Introduction

The novel coronavirus disease (COVID-19) swept the world in 2019, causing supply chain disruptions, and a number of scholars have studied the impact of COVID-19 on the global supply chain but have focused on the outbreak and epidemic phases [1]. The outbreak has exposed the varying degrees of vulnerability in the supply chains of different industries only at various stages of production, procurement, transportation, and sales. Notably, the recovery phase after the pandemic period has passed; the post-COVID-19 phase of the global supply chain will also continue to suffer from new challenges. Environmental awareness, for example, is growing rapidly with the popularity of information and social media [2], and He and Harris [3] found that during COVID-19, consumers tend to choose more responsible and prosocial companies. While the COVID-19 pandemic disrupted global supply chains, it also provided an opportunity to bring the concepts of sustainability and green practices back into focus [4]. The question of whether supply chain sustainability will change as a result of the COVID-19 crisis has been gaining attention in recent studies [5]. Some scholars have viewed the impact of the COVID-19 pandemic from the perspective of supply chain environmental sustainability, arguing that sustainability strategies and practices can help improve supply chain resilience [6]. Especially in the post-COVID-19 era, updating green supply chain management is important to provide supply chain continuity [7]. Through empirical testing, it was concluded that companies that implemented green supply chain management were better able to respond to the crisis and create more value in the post-Newcastle pneumonia outbreak economy [8].

Since the COVID-19 outbreak, plastic medical products have played an irreplaceable role in the global fight against the pandemic. As the global pandemic continues to grow, the demand for plastic products has increased dramatically. The proliferation of plastics has even created a unique stratum on the Earth’s surface, with significant long-term effects on the global ecosystem. Although plastic as a basic material has brought great convenience to human life, it has also been a major source of environmental pollution and the greenhouse effect. The extraction and processing of fossil fuels from plastic feedstock is carbon-intensive [9]. At present, more than 90% of plastic products are derived from fossil fuels, and their production generates approximately 400 million tons of GHG emissions globally each year [10]. Plastics production is projected to contribute 15% of global carbon emissions by 2050 [11]. These data show that the plastics industry has become a significant and rapidly growing source of industrial GHG emissions, and that addressing the impact of postproduction and postuse processing on the carbon footprint of the plastics industry in the post-COVID-19 era is critical.

In the post-COVID-19 era, with the global GHG reduction imperative, scholars have begun to pay attention to the net-zero emissions of the plastics industry and have proposed measures and ways to mitigate the carbon footprint of the plastics industry. Among them, renewable plastics are widely recognized for their ability to reduce the environmental impact of their carbon footprint [12]. The use of bio-based plastics as long-life materials to replace building materials and improve the production route of vinyl can also reduce the carbon footprint on the environment [13]. Meys et al. [14] present a bottom-up model of plastics production and waste plastics disposal that predicts five pathways to life-cycle GHG emissions from plastics in 2050, suggesting that a circular carbon economy (a combination of recycling, biomass utilization, and carbon capture and utilization (CCU) technologies) could achieve net-zero GHG emissions from plastics. Zheng and Suh [11] propose a quantitative assessment of strategies to mitigate plastic GHG emissions (renewable plastic substitution, use of a proportion of renewable energy, recycling, and total demand reduction) on a global scale and showed that a combination of the four strategies works best.

In summary, current scholars’ research related to plastic pollution is distributed to the damage to the land, marine, and atmospheric environments from the production and waste management processes of plastics [15,16]. The issue of carbon reduction in the plastics industry has received academic and governmental attention in terms of macropolicies and specific measures to reduce carbon emissions in plastics companies. However, what drives the reduction of the carbon footprint of the plastics industry at the supply chain level has not attracted much research attention.

This study aims to narrow the knowledge gap by addressing the following research questions:

• At the supply chain level, what factors drive the decarbonization of plastics supply chain members?
• Given the interdependence of the factors, what measures can be taken to drive carbon reductions in the plastics supply chain?

This study is dedicated to answering the above two questions and examining the drivers of decarbonization of the plastics supply chain in the post-COVID-19 era from a Chinese perspective. Firstly, we review the extant literature to identify the main drivers of the carbon footprint of plastic supply chain to answer the first question. Secondly, from the internal and external perspectives of the supply chain, we combine the evaluation of the factors by experts and business managers, using a combination of gray theory and DEMATEL to explore the causal relationship between the factors [17], and then the path and countermeasures for carbon reduction in the plastic supply chain are given according to the research findings to answer the second question. We provide recommendations for the plastics industry’s policy makers and regulators to achieve green practices in the industry, thereby increasing the resilience of the plastics supply chain facing the COVID-19 challenge.

The study is divided into 7 sections. Section 2 summarizes the drivers through a literature review. Section 3 introduces the Gray-DEMATEL method. Section 4 analyzes the causality of the drivers, and Section 5 discusses the results. Section 6 draws conclusions and implications. Finally, limitations of this paper and future research directions are described.

2. Literature Review

Under the background of the plastic ban and carbon neutral policy, the plastic industry will face the pressure of low-carbon transformation and green sustainable development. The carbon footprint behavior of the plastics supply chain is related to both internal and external drivers. Internal drivers include low-carbon production and end-of-life (EOL) management, while external drivers include stakeholder behavior and market factors in the plastics supply chain.

2.1. Identify Drivers Internal to the Supply Chain

“Internal to the supply chain” refers to the processes of product production and distribution and the corresponding internal environment that occurs within the enterprise. As the plastic production stage can indirectly affect the carbon cycle due to the different raw materials and energy sources chosen by companies, the EOL management stage can also directly generate carbon emissions due to the different choices made [18].

2.1.1. Low-Carbon Production

Olefins are an important raw material for plastics production [19]. A large amount of energy is consumed from the supply of raw materials to the manufacture of finished plastic products, and low-carbon production is the main strategy to reduce the carbon footprint of this process [12]. Biopolymers are often considered to be environmentally friendly alternatives to petrochemical polymers [20], and many scholars have proposed replacing fossil fuel-based plastics with bio-based plastics to reduce GHG emissions [21,22]. As the energy mix of the plastics supply chain becomes progressively decarbonized, the greater the proportion of renewable energy use (e.g., wind, solar), the greater the reduction in emissions [23]. Optimization of the modification process helps to safeguard the physical properties of the product and facilitates the application of a high percentage of recycled material [24]. Improved blow-molding process control techniques can reduce scrap rate, improve yield and capacity efficiency, and reduce electricity consumption [25]. Most plastic products are either used for packaging before or during transportation. Designers and engineers must work toward the long-term development goal of protecting the environment and resources and need to simultaneously consider new environmentally friendly packaging designs to match new environmentally friendly materials [26]. In this way, the eco-design of products can achieve environmental sustainability [27]. Designing packaging appropriately for industry-specific applications, establishing sound data on the distribution of packaging specifications affecting carbon footprint, and sequentially preoptimizing design solutions can be effective in mitigating carbon footprint [28].

2.1.2. EOL Management

Proper management of waste plastics can reduce carbon emissions in the recycling process [29,30]. There is research that demonstrates that investments in plastic waste management can contribute to achieving the Paris Agreement’s carbon reduction commitments in India and Indonesia [31]. A company’s green development cannot be achieved without the support and investment of top management. Top management’s approach and support often determines the likelihood of the successful adoption of sustainable supply chain management programs from an industrial perspective [32]. In terms of options for EOL management, incineration is a major source of GHG emissions, and landfills can pollute the ocean, thus affecting their ability to sequester carbon [18]. Therefore, the practice of reducing landfills and incineration can be effective in controlling carbon emissions [30,33]. Innovations in recycling technology can increase the recycling rate of plastics [34]. Therefore, designing safe and environmentally friendly recycling technologies to handle plastic waste on the reverse supply chain recycling network is also one of the drivers of carbon footprint mitigation [35,36]. More and more companies are now creating an effective after-use plastics economy. Fundamentally improving the economics, quality, and utilization of recycling; promoting reusable packaging; and expanding the adoption of industrial decomposable plastic packaging for targeted applications are key [37,38]. Climate change is one of the drivers for investing in waste management infrastructure [39]; i.e., improving waste management infrastructure as well as the design of recycling stations will also reduce the proportion of waste plastics to influence carbon footprint emissions [40].

2.2. Identify Drivers External to the Supply Chain

“External to the supply chain” refers to the stakeholders outside the supply chain network and the corresponding external environment. This subsection collects the corresponding drivers of carbon footprint mitigation from two strategic factors, namely, the joint promotion by various stakeholders and market influence.

2.2.1. Joint Stakeholder Promotion

In the face of carbon emissions from plastics, various stakeholders in the supply chain influence low-carbon and low-emission actions to address climate change [41]. The implementation and support of government regulations is important, and regulations restricting plastic production as well as legislation and incentives for plastic pollution can mitigate pollution [42,43]. Increased consumer awareness of environmental protection has put pressure on manufacturers and retailers [44,45], thus urging manufacturers to consider consumer preferences in their production designs. All supply chain partners and actors are engaged with internal and external recyclers to effectively operate and manage plastic waste [46]. Nongovernmental organizations (NGOs) and green promotion organizations force business organizations to critically consider their plans for ecological and social sustainability [47]. The cooperation between enterprises and R&D institutions plays an important role in promoting the rapid development of enterprises and the technological progress of enterprises [48]. Collaboration among supply chain members is one of the key factors in the development of new technologies, processes, and products [49]. Thus, it promotes the green development of the plastics supply chain.

2.2.2. Market Impact

As carbon markets are established and companies seek to go green to gain competitiveness, markets are beginning to influence the carbon emissions of the supply chain [50]. Programs and organizations that participate in carbon trading can trade in the carbon market based on the surplus of corporate allowances, and enhanced disclosure of the completeness and quality of information on carbon footprint measurement and management, climate change strategies, and risk management processes and results are also factors in carbon footprint mitigation [51]. The core of the New Plastics Economy theory is the circular economy of plastics, and the Global Commitment for a New Plastics Economy promotes green and sustainable development of the plastics supply chain [52]. With the development of a low-carbon economy, companies will adopt more low-carbon strategies to maintain their competitiveness in the market [53]. They also will seek to shift their climate-related strategies from a focus on risk management and bottom-line protection to one that emphasizes business opportunities and bottom-line enhancement [54].

In view of this, the framework flow of the next section of the study is shown in Figure 1. In addition to this, based on the above review, a total of 25 drivers for mitigating the carbon footprint of the plastics supply chain were identified. The final list of factors is shown in

Table 1. Key drivers of mitigating the carbon footprint of the plastic supply chain.

Indicator	Code	Subcriteria	Code	Brief Description	Source
Low-carbon production	C1			Solutions to reduce carbon footprint in the supply of materials to the production of plastics.	Zheng and Suh (2019) [11]
		Use of bio-based plastics	C1.1	Bio-based plastics instead of fossil fuel-based plastics.	Hillmyer (2017), Weiss et al. (2012) [21,22]
		Renewable energy use	C1.2	Low-carbon energy use is a strategy to reduce greenhouse gas emissions in the plastics production phase.	Posen et al. (2017) [23]
		Process optimization	C1.3	The optimization of the modification technology helps enhance the physical properties of the product, which improves the utilization rate of raw material.	Mekonnen et al. (2013) [24]
		Blow molding yield improvement	C1.4	Improve blow-molding process control technology to reduce scrap, improve yield rate and capacity efficiency, and reduce electricity consumption.	Pinto et al. (2019) [25]
		Packaging product design	C1.5	Design industry-specific products to reduce lifecycle carbon footprint.	Zheng (2012), Burke et al. (2021), García-Arca and Prado (2008) [26,27,28]
EOL management	C2			A series of management methods for waste-plastics to reduce carbon emissions in the recycling process.	Hopewell et al. (2009), Lazarevic et al. (2010) [29,30]
		Top-management support	C2.1	Top-management support of plastic recycling.	Muduli et al. (2013) [32]
		Landfill and incineration reduction practices	C2.2	Reducing plastic waste in landfills and incineration to control carbon emissions.	Lazarevic et al. (2010), Eriksson and Finnveden (2009) [30,33]
		Recycling technology and green innovation	C2.3	Designing safe and environmentally friendly recycling technologies for plastic waste.	Kuusela et al. (2021), Flygansvær et al. (2019), Huang and Yang (2014) [34,35,36]
		Create an effective after-use plastic economy	C2.4	Fundamentally improve the economics, quality, and utilization of recycling; promote reusable packaging; and expand the adoption of industrial decomposable plastic packaging for targeted applications.	Agenda (2016), Paletta et al. (2019) [37,38]
		Investing in waste management infrastructure	C2.5	Improving waste management infrastructure and the design of recycling stations.	Wilson (2007), Lebreton and Andrady (2019) [39,40]
Joint stakeholders promotion	C3			Various stakeholders in the supply chain influence low-carbon actions to address climate change.	Hu and Wang (2021) [41]
		Government regulations and support	C3.1	The implementation and support of government regulations are important, as are restrictions on plastic production and legislation and incentives for controlling plastic pollution.	Filatov et al. (2018), Garcia et al. (2019) [42,43]
		Consumers’ awareness of environmental protection	C3.2	Consumer awareness of environmental protection puts pressure on manufacturers and retailers.	FuiYeng and Yazdanifard (2015), Walker et al. (2021) [44,45]
		Strategic alliance between supply chain partners and recycling companies	C3.3	The involvement of all supply chain partners and actors with internal and external recyclers to effectively operate and manage plastic waste.	Zhou et al. (2018) [46]
		NGOs and green promotion groups	C3.4	Nongovernmental organizations (NGOs) and green promotion organizations, forcing business organizations to critically consider their ecological footprint and social responsibility.	Walker et al. (2008) [47]
		Cooperation between enterprises and R&D institutions	C3.5	R&D outsourcing plays an important role in promoting the rapid development of enterprises and technological progress of enterprises.	Kim and Lim (2015) [48]
		Cooperation among supply chain members	C3.6	Collaboration among supply chain members is a key factor in the development of new technologies, processes, and products.	Soosay et al. (2008) [49]
Market impact	C4			Market factors as factors influencing supply chain carbon emissions.	Du et al. (2021) [50]
		Carbon emissions trading	C4.1	Participation in carbon emissions trading programs.	Mahapatra et al. (2021) [51]
		Information disclosure	C4.2	Disclosure on carbon footprint measurement and management, climate change strategy, and risk management completeness and quality of process and results information.	Mahapatra et al. (2021) [51]
		The development of the new plastic economy	C4.3	Global Commitment to a New Plastics Economy drives companies toward a circular plastics economy.	Rhein and Sträter (2021) [52]
		Business competition	C4.4	With the development of the low-carbon economy, companies will adopt more low-carbon strategies to maintain market competitiveness.	Hongjuan and Jing (2011) [53]
		Combination of climate change with business strategy	C4.5	Companies seek to shift their climate-related strategies from a focus on risk management and bottom-line protection to one that emphasizes business opportunities and bottom-line enhancement.	Hoffman (2007) [54]

.

3. Methodology

3.1. Application of Gray-DEMATEL

DEMATEL refers to the conversion of interdependencies into cause and effect groups through matrices and the finding of the key factors of complex structural systems with the help of an impact relation diagram [55] and is widely used to analyze the influences in various complex systems [56]. However, the DEMATEL method is susceptible to human judgment inaccuracies and unpredictability of the surrounding environment, and the introduction of gray theory can attenuate such effects [57]. Gray number theory is the use of gray fuzzy concepts to solve uncertain attribute decision problems, and its main advantage is the use of gray number intervals rather than an exact value when making decisions, making the decisions more flexible and thus making the results closer to reality [58].

Currently, the Gray-DEMATEL method has been used by scholars to identify the influencing factors in many fields. For instance, Haleem et al. [59] used the Gray-DEMATEL approach to explore the inter-relationship between factors influencing the implementation of traceability systems in the food supply chain; Garg [60] used the Gray-DEMATEL approach to determine the interdependencies between e-waste mitigation strategies through cause–effect analysis; and Govindan [61] used Gray-DEMATEL to analyze how critical success factors for artificial intelligence can drive sustainable frugal innovation. This paper combines the gray number theory with the DEMATEL method, which helps to solve the problem of variability of different enterprises that are difficult to reflect when experts score, and is closer to the real situation, making the evaluation of the carbon footprint of plastic supply chain mitigation more objective and scientific.

3.2. Methodology Steps

In this study, we used the Gray-DEMATEL approach to explore the causal relationship between the drivers of carbon footprint of the mitigation plastic supply chain, with the following steps.

Step 1: Construct a matrix of carbon footprint drivers for plastic supply chain mitigation. Experts were asked to score the four first-level indicators in the list of drivers with their corresponding subcriteria and analyze the impact of individual factors on other factors. The degree of impact is judged by five indicators: no effect, very low effect, low effect, high effect, and very high effect. Additionally, referring to the Gray method, the semantic variables of expert evaluation are obtained, as shown in

Table 2. Semantic variables of expert evaluation.

Semantic Variables	Respective Gray Scale	Specific Values
No effect	[0, 0]	0
Very low effect	[0, 0.25]	1
Low effect	[0.25, 0.5]	2
High effect	[0.5, 0.75]	3
Very high effect	[0.75, 1]	4

.

Step 2: The corresponding gray matrix ($\otimes {{\displaystyle A}}_{ij}^m$) is calculated according to Equation (1) [57]:

$$\otimes A_{ij}^m=(\underset{¯}{\otimes }{A}_{ij}^m,\overline{\otimes }{A}_{ij}^m)$$

(1)

where 1 ≤ m ≤ n; 1 ≤ i ≤ c; 1 ≤ j ≤ c. n represents the number of scoring experts, i; j represents the rows and columns; c represents the number of factors; and m is the sequence of experts.

Step 3: Establish the average Gray matrix (${\otimes }^v{A}_{ij}$) using Equation (2):

$${\otimes }^v{A}_{ij}=({\displaystyle \sum _m\frac{\underset{¯}{\otimes }{A}_{ij}^m}{n}},{\displaystyle \sum _m\frac{\underset{¯}{\otimes }{A}_{ij}^m}{n}})$$

(2)

Step 4: Convert the average gray matrix into a clear relationship matrix.

• Standardized processing:

$$\underset{¯}{\otimes }{\stackrel{•}{A}}_{ij}=(\underset{¯}{\otimes }{\stackrel{\nu }{A}}_{ij}-{}_j{}^{\min }\underset{¯}{\otimes }{\stackrel{\nu }{A}}_{ij})/{\Delta }_{\min }^{\max }$$

(3a)

$$\overline{\otimes }\stackrel{•}{A_{ij}}=({\overline{\otimes }}^v{A}_{ij}-{}_j{}^{\min }\overline{\otimes }{}^{v}A{}_{ij})/{\Delta }_{\min }^{\max }$$

(3b)

$${\Delta }_{\min }^{\max }={}_j{}^{\max }\overline{\otimes }{\stackrel{\nu }{A}}_{ij}-{}_j{}^{\min }\underset{¯}{\otimes }{\stackrel{\nu }{A}}_{ij}$$

(3c)

• Compute total normalized crisp value:

$$B_{ij}=(\frac{(\underset{¯}{\otimes }\stackrel{•}{A_{ij}}(1-\underset{¯}{\otimes }{\stackrel{•}{A}}_{ij})+\overline{\otimes }\stackrel{•}{A_{ij}}\times \overline{\otimes }\stackrel{•}{A_{ij}})}{(1-\underset{¯}{\otimes }{\stackrel{•}{A}}_{ij}+\overline{\otimes }\stackrel{•}{A_{ij}})})$$

(3d)

• Compute final crisp values:

$$B_{ij}^{*}=(\min \underset{¯}{\otimes }\stackrel{•}{A_{ij}}+(B_{ij}\times {\Delta }_{\min }^{\max }\left)\right)$$

(3e)

Step 5: Directly affect the matrix regularization to obtain “N”. Sum the rows of the matrix and set the maximum value of the rows and Max, that is:

$$N=B_{ij}^{*}/\mathrm{Max}$$

(4)

Step 6: Calculate the integrated matrix “T”:

$$T=N{(I-N)}^{-1}$$

(5)

where, I is the identity matrix.

Step 7: Obtain the causal parameters. The element $t_{ij}$in matrix T indicates the degree of direct and indirect influence of factor i on factor j, or the degree of combined influence of factor j from factor i. According to Equation (6a,b), calculate the degree of influence (D), the degree of being influenced (C), the degree of centrality (D + C), and the degree of cause (D − C). Where D indicates the comprehensive influence value of a row element on other factors; C indicates the comprehensive influence value of a column element by other factors; D + C indicates the position of the factor in the index system and the size of its role; D − C indicates the influence of a factor on other factors, and if it is positive, it indicates a strong influence on other factors and is called the cause factor, but if it is negative, it indicates a strong influence by other factors and is called the effect factor.

$$D={[{\displaystyle \sum _{j=1}^c{{\displaystyle t}}_{ij}}]}_{\mathrm{c}\times 1}$$

(6a)

$$C={[{\displaystyle \sum _{j=1}^c{{\displaystyle t}}_{ij}}]}_{1\times c}$$

(6b)

Step 8: Export cause–effect diagrams. Construct the coordinate system by taking the obtained cause degree as the horizontal coordinate and the center degree as the vertical coordinate. In addition, the number of each influencing factor is indicated in the figure. Then, the mean and standard deviation of the combined impact matrix are calculated; the sum of the two is expressed by θ as the threshold value, compared with the element t_ij within the combined influence matrix, and if an element $t_{ij}$is greater than θ, it indicates that factor i has a significant influence on factor j, marked with a directed arrow in the scatter plot [62].

4. Data Collection and Analysis

4.1. Data Collection

Many of the current studies on the combination of supply chain and drivers or barriers use questionnaires and interviews to obtain primary data. For example, Farooque et al. collected data from 105 questionnaires from Chinese food supply chain stakeholders to examine barriers to circular food supply chains [63]. Zhang et al. obtained data from interviews with experienced practitioners to examine barriers to smart waste management in China [64].

Therefore, we invited project managers and executives with experience in plastics supply chain industry management and academics in related fields as the target group of our research. The data collection methods we used are similar to those of other research topics [65]. Since the scope of the study is in China, we looked for Chinese plastics industry employees and academics whose research interests are in supply chain sustainability, taking into account that there is no language barrier. The identified factors were used to design a rating scale according to the criteria in the first step above [59], and twelve experts were invited to score them, including four academics working in supply chain sustainability research fields and eight practitioners working in the plastics industry. Their average years of work were 12.8 years and 10.6 years, respectively. The interviewees for this study were entirely volunteers, and initial conversations were conducted to ensure that all experts on the team had achieved a level of knowledge of plastics production and supply chain. The data were obtained through a paper-based questionnaire, and the interviewees were also trained on the background knowledge of low-carbon development in the plastics supply chain and introduced to the research context of this study before formally completing the questionnaires. Finally, 12 questionnaires were effectively collected with a 100% response rate.

Table 3. Expert team’s information and profiles.

NO.	Academic/ Practitioner	Department	Position	Year of Work
1	Academic	Business School	Professor	24
2	Academic	Business School	Senior Lecturer	5
3	Academic	School of Management	Professor	20
4	Academic	School of Economics and Management	Lecturer	2
5	Practitioner	Operations	Sales Operations	1
6	Practitioner	3D Printing Management	Company Shareholders	5
7	Practitioner	Government Agencies	Deputy Director of Cooperative Economic Guidance Division	15
8	Practitioner	Communication Industry	Beijing District Assistant	1
9	Practitioner	Investment Management	Project Manager	4
10	Practitioner	Business Management and Operations	Energy Trainee	1
11	Practitioner	Trading	General Manager	33
12	Practitioner	Finance	Vice President	25

presents the information and overview of the expert panel.

4.2. Gray-DEMATEL Analysis

The average gray matrix, the final direct relationship matrix, the normalized direct relationship matrix, the synthesis matrix, and the algorithm code of the primary factors and the corresponding subfactors calculated according to steps 1 to 6 in Section 3 are shown in Supplementary Materials. The following section shows only the final calculated indicators of the four dimensions and the derived directed causality diagrams to analyze the relationship between the factors.

4.2.1. Factor Analysis of the Four Level 1 Indicators

The influence degree (D), influenced degree (C), centrality degree (D + C) and cause degree (D − C) calculated according to step 7 are shown in

Table 4. Cause and effect result for level 1 indicators.

Level 1 Indicators	D	C	D + C	D − C	Cause/Effect
C1	3.66	5.63	9.29	−1.97	Effect
C2	4.28	4.26	8.55	0.02	Cause
C3	5.30	3.93	9.23	1.37	Cause
C4	5.04	4.46	9.49	0.58	Cause

(two decimal places are retained), and the directed graph derived from step 8 is shown in Figure 2.

From the results presented in Table 4 and Figure 2 above, it can be visualized that the market impact (C4) has the largest centrality value (D + C) and is the most prominent driver under Tier 1 indicators in mitigating the carbon footprint of the plastics supply chain. The negative value of D − C shows that low-carbon production (C1) belongs to the effect group and is easily influenced by other factors. EOL management (C2), joint promotion by stakeholders (C3), and market influence (C4) belong to the cause group, where C3 has the largest D − C value and is the most significant cause driver in mitigating carbon footprint, followed by C4, both of which have a greater impact on C1.

4.2.2. Subfactor Analysis of C1

The factor index values of the four dimensions were obtained according to the calculation in step 7, as shown in

Table 5. Cause and effect result for C1 subfactor.

C1 Subcriteria	D	C	D + C	D − C	Cause/Effect
C1.1	6.21	6.09	12.30	0.13	Cause
C1.2	5.35	5.49	10.85	−0.14	Effect
C1.3	6.08	5.45	11.54	0.63	Cause
C1.4	5.09	4.80	9.89	0.29	Cause
C1.5	4.27	5.18	9.45	−0.90	Effect

, and the directed relationship diagram derived in step 8 is shown in Figure 3.

According to the magnitude of D + C values, the ranking of drivers is C1.1 > C1.3 > C1.2 > C1.4 > C1.5, i.e., use of bio-based plastics (C1.1) is the most prominent driver, followed by modification process optimization (C1.3). Based on their D − C values, the subfactors were divided into two groups (cause and effect), where C1.1, C1.3, and C1.4 belonged to the cause group, and the process optimization (C1.3) was the most significant cause driver, because they had the largest value (D − C) and the greatest effect on the other factors; C1.2 and C1.5 belong to the effect group and are easily influenced by other factors. In the directed causal diagram in Figure 3, C1.2 is more significantly influenced by C1.1 and C1.3 for values greater than the threshold θ, and C1.1 interacts with C1.2 and C1.1 with C1.3.

4.2.3. Subfactor Analysis of C2

The factor index values of the four dimensions were obtained according to the calculation in step 7, as shown in

Table 6. Cause and effect result for C2 subfactor.

C2 Subcriteria	D	C	D + C	D − C	Cause/Effect
C2.1	4.29	2.52	6.81	1.77	Cause
C2.2	3.15	3.80	6.94	−0.65	Effect
C2.3	3.38	3.59	6.97	−0.21	Effect
C2.4	3.24	3.97	7.20	−0.73	Effect
C2.5	3.40	3.58	6.98	−0.18	Effect

, and the directed relationship diagram derived in step 8 is shown in Figure 4.

Based on the magnitude of D + C values, the drivers are ranked as C2.4 > C2.5 > C2.3 > C2.2 > C2.1, with little difference in the centrality of the four factors, all of which are more prominent, with the creation of an effective after-use plastics economy (C2.4) being the most prominent driver of carbon footprint mitigation in the end-of-life management phase. Based on the positive and negative D − C values, only top-management support (C2.1) is a cause factor, the most significant cause driver for carbon footprint reduction in the EOL phase, and has a significant impact on other factors; the remaining four factors are effect factors. Clearly, if the key driver (C2.1) is strengthened, it will facilitate the remaining three factors to further play an abatement role in the EOL phase.

4.2.4. Subfactor Analysis of C3

The factor index values of the four dimensions were obtained according to the calculation in step 7, as shown in

Table 7. Cause and effect result for C3 subfactor.

C3 Subcriteria	D	C	D + C	D − C	Cause/Effect
C3.1	8.13	6.48	14.61	1.65	Cause
C3.3	7.59	8.15	15.75	−0.56	Effect
C3.4	6.91	7.09	13.99	−0.18	Effect
C3.5	7.01	6.79	13.80	0.21	Cause
C3.6	6.86	8.25	15.11	−1.38	Effect

, and the directed relationship diagram derived in step 8 is shown in Figure 5.

Based on the D + C values, the ranking of drivers is C3.3 > C3.6 > C3.1 > C3.2 > C3.4 > C3.5, i.e., the strategic alliance between supply chain partners and recycling companies (C3.3) is the most prominent driver under the joint promotion strategy of each stakeholder. Based on the positive and negative D − C values, government regulations and support (C3.1), increased environmental awareness of consumers (C3.2), and cooperation between enterprises and R&D institutions (C3.5) are among the cause factors, with C3.1 having the greatest influence (D − C value max) on the other factors and a significant effect on C3.4, C3.6, and C3.3; NGOs and green promotion groups (C3.4), strategic alliances between supply chain partners and recycling companies (C3.3), and collaboration of supply chain members (C3.6) are among the effect factors. From the direction of the directed arrows in Figure 4, it can be concluded that C3.6 is influenced by C3.5 and C3.2, and C3.3 is influenced by C3.2.

4.2.5. Subfactor Analysis of C4

The factor index values of the four dimensions were obtained according to the calculation in step 7, as shown in

Table 8. Cause and effect result for C4 subfactor.

C4 Subcriteria	D	C	D + C	D − C	Cause/Effect
C4.1	10.11	11.38	21.48	−1.27	Effect
C4.2	11.48	8.46	19.93	3.02	Cause
C4.3	9.69	11.76	21.45	−2.07	Effect
C4.4	10.83	10.52	21.35	0.32	Cause
C4.5	11.38	11.38	22.76	0.00	Cause

, and the directed relationship diagram derived in step 8 is shown in Figure 6.

Based on the D + C values, the drivers were ranked as C4.5 > C4.1 > C4.3 > C4.4 > C4.2, i.e., climate change combined with business strategy (C4.5) was the most prominent driver. Based on the positive and negative D − C values, information disclosure (C4.2) and business competition (C4.4) belong to the cause factors, and information disclosure (C4.2) is the most significant cause driver influencing carbon footprint reduction under this strategy, and has a significant effect on C4.1, C4.3, and C4.5, and C4.5 also has a significant effect on C4.1 and C4.3; C4.1 and C4.3 belong to the effect factors. It is clear that advancing the implementation of key factors C4.2 and C4.5 will further contribute to the mitigation effect of the implementation of drivers C4.1 and C4.3.

5. Discussion

After analyzing the data in Section 4, the results are discussed separately as follows:

(1)

Market impact (with the highest D + C value) is the most prominent driver of carbon footprint mitigation in the plastics supply chain, with low-carbon production being the effect factor leading to a reduced carbon footprint that is vulnerable to influences from both market and nonmarket forces. The joint promotion by various stakeholders is the biggest cause factor for the net effect value (D − C) and has a strong contribution to the reduction of carbon footprint in the plastics industry. Rajeev et al. [66] analyzed stakeholder engagement and influenced the progress of the industry through a case study of the plastics industry, illustrating its importance in achieving sustainable business.

(2)

Under a low-carbon production strategy, the use of recycled plastic resources (with the highest D + C values) is the most prominent driver of carbon-footprint reduction. The bioplastics industry is assessed as an environmentally relevant industry with unlimited market potential and growth potential [67]. The use of renewable energy (C1.2) as an effect factor (negative D − C value), although influenced by the use of recycled plastic resources (C1.1) and the optimization of the modification process (C1.3), also had a significant effect on C1.1. Renewable energy use can achieve greater emission reductions than corn-based plastics at lower cost and with less uncertainty, and if the two are combined, further reductions can be achieved [23].

(3)

Under the EOL management strategy, the highest level of management support (C2.1) is the most significant cause driver of carbon footprint mitigation at this stage and has a significant impact on the other drivers at this stage. Creating an effective economy of after-use plastics (with maximum D + C values) is a prominent driver of this phase. This is in line with the idea put forward by d’Ambrières [68] that the best answer to the problem of waste plastics from an environmental and socioeconomic point of view is to recycle. Therefore, during the EOL phase, stakeholders should be encouraged and coordinated to redesign a more efficient plastics economy.

(4)

Under the strategy of joint promotion by various stakeholders from the external supply chain perspective, government regulations and support (C3.1) is the most significant cause factor, with the largest degree of causality, and has a large influence on other drivers. In recent years, efforts to develop legislative instruments have played a positive role in managing plastic production, consumption, and waste, and taxes and bans have been successful strategies to control plastic pollution at the regional level, not only to improve consumer behavior but also to bring about changes in stakeholder cooperation strategies [69]. Of these, supply chain collaboration is the factor that is most vulnerable to change and is at greater risk.

(5)

Under the market impact strategy, integrating climate change with business strategies (C4.5) is the most prominent driver of carbon footprint mitigation. Information disclosure (C4.2) is the most significant cause driver. Among them, by constructing a supply chain decision model and using the production of PVC pipes in China as an example, some scholars have demonstrated that carbon market fluctuations have a direct impact on the green design and management of the supply chain through various mechanisms [70].

6. Conclusions and Implications

The carbon emission problem brought about by the application of plastic hinders the process of carbon neutralization in this industry. This work aims to distinguish and analyze the key factors that mitigate the carbon footprint of the plastics supply chain from two perspectives, namely, internal and external regarding the supply chain. This is achieved through relevant literature searches and key drivers identified by domain experts. The relationship between the two levels of the indicator system was analyzed separately using the Gray-DEMATEL method, and then the mitigation carbon footprint drivers were divided into two groups (cause or effect). This paper makes an original contribution. Firstly, the carbon neutral theory is integrated with the supply chain; secondly, the framework of the drivers of plastic carbon footprint reduction from the supply chain perspective is identified; and finally, the impact of the drivers at two levels is explored. Therefore, the findings and implications discussed in this paper may be of significant assistance to policy makers, manufacturing and recycling companies, managers, industry experts, and researchers in successfully advancing the interplay between low-carbon development in the plastics industry.

6.1. Managerial Implications

In the context of carbon neutrality, there are opportunities for the plastics industry to achieve net-zero emissions, but the achievement of this goal will require the cooperation of companies upstream and downstream in the plastics supply chain, as well as the efforts of other stakeholders. This work recommends that managers understand the drivers of mitigating the carbon footprint of the plastics supply chain for sustainable industry development. The management implications of this study are as follows:

(1)

This study helps managers understand the interdependence of the drivers that contribute to low-carbon development in plastics. For example, the driver “low-carbon production” was significantly influenced by “joint stakeholders promotion” and “market impact” in the effect group.

(2)

In addition to this, market impact is the most prominent driver. This corresponds to the results of the EOL management driver “Creating an effective after-use plastics economy”. Therefore, managers in the plastics industry should use the concept of circular economy to deploy product eco-design and green supply chain technologies. Among them, integrating the circular economy concept into supply chain management can provide a new perspective for supply chain sustainability [71]. Lack of market pressure and demand is one of the main barriers to the role of smart-enabling technologies in advancing the shift from waste management to a circular economy [64]. This can promote the plastics economy from an open-loop linear model to a closed-loop circular model and use the flexibility of market mechanisms to drive the industry’s low-carbon development.

(3)

Regarding the analysis of the subfactors of the EOL management phase, the driver “top management support” is the most important cause driver due to it having the largest D − C value and a greater influence on other factors. Therefore, recycling company managers should strengthen their plans for sustainable management of the plastics supply chain at this stage.

(4)

Regarding the analysis of the subfactors of “market impact”, the most prominent driver C4.5 (combine climate change with business strategy) and the most important cause driver C4.2 (information disclosure) have a significant effect on C4.1 (carbon emissions trading) and C4.3 (the development of the new plastic economy). Therefore, corporate managers should actively participate in carbon footprint-related disclosure programs and actively address strategies to integrate climate change with business strategies.

6.2. Policy Implications

This work recommends that policy makers understand the dynamics of mitigating the carbon footprint of the plastics supply chain for sustainable industry development. The policy implications of this study are as follows:

(1)

In the production phase of plastics, industry regulators can do more to promote low-carbon production choices among manufacturers by strengthening forces external to the supply chain. Additionally, there is a need to increase the proportion of recycled plastic resources utilized. Therefore, policy makers should rationalize the proportion of biomass plastic applications and promote them. Additionally, they should invest more in modification process technology to improve the stability of recycled plastics.

(2)

Analysis of the subfactors of joint stakeholders promotion shows that government regulations and support are the most important cause drivers. It is important to increase the implementation of the plastic ban policy to cover the production of raw materials, as well as processing, sales, use, recycling, and other aspects of the entire industry chain, to promote the transition of the plastic industry toward plastic that is recyclable, easy-to-recycle, and biodegradable, involving the whole life cycle of green development of the new road.

(3)

C4.5 (combine climate change with business strategy) is the most prominent driver among the subfactors influencing the market. The plastics industry can be encouraged to actively participate in the national carbon emissions trading system and to establish and improve the relevant supporting system.

7. Limitations and Future Research Directions

This paper uses Gray-DEMATEL to analyze the drivers of carbon footprint mitigation in the plastics supply chain, based on the internal and external perspectives of the plastics supply chain. However, due to the limitations of subjective factors, there are some shortcomings in this study; specifically, there are several limitations in the following aspects:

(1)

In this study, the drivers for mitigating the plastics supply chain are only based on current industry developments gathered from the literature and the rapid development of plastics production and recycling technologies and policies, so the literature will need to be updated after a few years.

(2)

The study context is China, and future research should investigate the problem in other countries for comparative analysis. It is worth exploring how to study the factors that mitigate the carbon footprint of the plastic life cycle more deeply and finely.

(3)

Other methods, such as hierarchical cluster analysis and structural equation modeling, can also be integrated in future studies. More quantitative linkages between factors are needed.