Impact of Smart, Green, Resilient, and Lean Manufacturing System on SMEs’Performance: A Data Envelopment Analysis (DEA) Approach

Abstract

In the present era of the fourth industrial revolution, small and medium enterprises (SMEs) are adopting smart, green, resilient, and lean (SGRL) practices to enhance their performance and achieve sustainability. For SMEs to perform well in their supply chains and satisfy customers, the impact of the combined effects of SGRL manufacturing on SMEs’ performance needs to be studied. Although SGRL manufacturing has been studied independently in order to understand its impact on SMEs’ performance, there is still a need for significant research on its combined effect. The objective of the present work is to evaluate the performance of SMEs and to understand the combined effect of SGRL manufacturing on SMEs’ performance. This research applied the data envelopment analysis (DEA) methodology to evaluate 30 SMEs identified in the northern region of India. A DEA model was developed that considers environmental, operational, and social performances as output criteria while considering SGRL practices as input criteria. Sixteen decision-making units (DMUs) were identified as inefficient using the DEA approach and one of them was considered for a case study for comparison with efficient SMEs. The case study employed a Strength, Weakness, Opportunity, and Threat (SWOT) analysis to provide remedial action to one of the selected underperforming SMEs, i.e.,SME11. The strengths, weaknesses, opportunities, and threats of SME11 were identified and strategies were suggested by benchmarking SME11 with one of the efficient SMEs, i.e., SME23. The findings of this research work will help policymakers, owners, and managers of SMEs take necessary actions and enhance their performance by adopting the proposed DEA model using SGRL manufacturing practices.

1. Introduction

Around 95% of firms worldwide are small- and medium-sized enterprises (SMEs), which employ between 60% and 70% of the global workforce [1]. India’s manufacturing SMEs significantly contribute to the country’s gross domestic product (GDP) and employ nearly 40% of India’s workforce [2]. Although it is acknowledged that SMEs contribute significantly to the economic growth of any nation, they also place a great deal of strain on the environment, both individually and collectively [3].

Currently, SMEs are aware of the need to act on sustainable dimensions and find new competitive manufacturing methods due to the rising expectations concerning sustainable performance and process transparency. Smart, Green, Resilient, and Lean (SGRL) manufacturing is proving to be a successful solution in this situation. Moreover, manufacturers are using smart, green, resilient, and lean strategies either independently or in combination to enhance their supplychain-related and organizational performance. Several research studies have shown that implementing these paradigms alone would not produce the desired goals, but that combining smart, green, resilient, and lean techniques can produce superior outcomes [4,5,6].

An integrated theoretical framework for the integration of smart, green, resilient, and lean methods in the context of manufacturing was proposed by [7]. With the aid of Industry 4.0 (14.0) technologies that are part of smart manufacturing, it is feasible to integrate various tactics into one intelligent and sustainable platform [8]. The three sustainability-related features of SGRL’s smart model are positively correlated; consequently, the other three green, resilient, and lean paradigms are strengthened [9,10]. Green manufacturing creates an atmosphere that is eco-friendly while enhancing worker health and safety [11]. The production system’s resilience makes it possible to deal with disturbances and bounce back, thereby maintaining manufacturing operations [12,13]. Finally, lean manufacturing reduceswaste and enhances employees’ self-esteem [14].

Although lean manufacturing, green manufacturing, resilient manufacturing, and smart manufacturing have all been studied independently with respect to their effects on SMEs’ performance [15,16,17], there is still a significant knowledge gap regarding their combined effects on performance, as the literature on an integrated SGRL manufacturing framework to assess the performance of SMEs is at nascent stage. Furthermore, various techniques have also been used to assess the performance of SMEs; nevertheless, there is still a need to bridge the gap regarding the way in which toeffectively evaluate their performance and offer solutions to SMEs to improve their performance. The data envelopment analysis (DEA) framework proposed in this study helps evaluate SMEs based on their implementation of SGRL practices as well as their supply chain performance, and it also aids the segregation of SMEs into efficient and inefficient classes. DEA is a linear programming technique that provides dynamic, collective, comparative results for evaluating the productivity of organizations based on multiple inputs and outputs [18]. Furthermore, a case study was conducted using Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis and benchmarking such that SMEs can compare themselves to relevant peer SMEs and, after benchmarking with the efficient SMEs, develop their own SWOT-derived strategies, which will enhance their performance.

The main goal of this paper is to help SMEs maximize their performance, but it also aims to show how the combined effects of smart, green, resilient, and lean manufacturing affect SMEs’ performance. In this study, 30 SMEs were evaluated, of which 14 were found to be efficient and 16 were found to be inefficient SMEs. Moreover, while evaluating these SMEs through the results of the conducted DEA, it was observed that SMEs that had higher levels of SGRL implementation constituted the better-performing SMEs.

The objectives of this research study are as follows:

• To identify SGRL practices (input variables) and performance dimensions (output variables) in Indian manufacturing firms;
• To develop a DEA model and evaluate the efficiencies of SMEs;
• To segregate SMEs based on efficient and inefficient units;
• To provide solutions to inefficient SMEs using benchmarking and SWOT analysis.

The rest of thia research is organized as follows. The next section presents a comprehensive literature review that identifies SGRL practices and performance dimensions as input and output variables for the DEA model proposed in the study, followed by a detailed description of the research methodology. In the next section, the results and analysis are explained. The discussion is presented in the subsequent section, and the conclusions, limitations, and future research directions are discussed at the end.

2. Literature Review

Industry 4.0 technologies are combined with green, resilient, and lean manufacturing processes to form SGRL production. Thisconstitutes an intelligent system that uses Industry 4.0 to target the preservation and protection of natural resources, the reduction and absorption of disturbances, the elimination of waste, and the production of value. The SGRL manufacturing processes, their relevance, and their importance with respect to boosting the performance of Indian SMEs are all comprehensively addressed in this section.

2.1. Smart Manufacturing

To improve manufacturing performance and maximize the amount of energy and labor needed throughout the production process, a form of manufacturing known as “smart manufacturing” (SM) is employed [10,19]. Additionally, it applies the concepts of interoperability, real-time control, and servitization to digitize every production element, employing modern tools and networked machinery [20]. Organizations are currently dealing with the fourth industrial revolution, or “Industry 4.0.”This revolution has produced a new reality and method of industrial organization based on digitization and the extensive use of information. In the modern world, industry 4.0 technologies are employed extensively to create smart manufacturing solutions such that the two concepts are frequently considered equivalent [21]. Augmented reality (AR) and virtual reality (VR) can be usedto teach personnel who lack manufacturing experience or are ill-equipped to handle the usage of cutting-edge technology [10]. Since they make it easier to see the finished product in a virtual setting, these technologies can help lower the cost of design and prototypes [22]. The manufacturing process is efficiently monitored via big data analytics (BDA) predictive capabilities; as a result, production problems and quality issues can be identified in advance due to the recapitulation of data patterns in time series [19]. BDA may be used to evaluate data from customer orders, sales, inventory, and production resources to help manufacturers understand their customers’ purchasing patterns and make smart decisions in a proactive manufacturing environment [23,24]. Moreover, additive manufacturing (AM) promotes reverse engineering and quick production reconfigurations by providing flexibility, time savings, and cost reductions [25]. Artificial intelligence (AI) improves manufacturing such that it becomes autonomous, self-deciding, and self-optimizing [13,26].

2.2. Green Manufacturing

Globalization has altered business practices, especially in India, where the manufacturing industry has achieved steady success because of its low production costs, large population density, and the rising local demand for its products [27]. In the modern world, wherein a lack of energy and natural resources is a major challenge on a worldwide scale, the manufacturing industry uses the word “green”, which means to “make or produce something environmentally friendly” [28]. According to Paul et al. [29], an adequately green manufacturing system lowers operational costs through the efficient and effective use of raw materials, energy, and labor, all of which eventually increase the value of the final product. Deif [28] also claimed that using resources effectively and efficiently leads to better quality and manufacturing processes, which increase a company’s market share and reputation through increased customer satisfaction. According to Cherrafi et al. [30] and Garza-Reyes [15], green practices can help achieve economic and environmental goals. According to Scur and Barbosa [31], leading manufacturers have implemented green practices that improve the working environment and organizations’ public perception. In addition, implementing green practices is cost-effective for the organization, ultimately giving rise to economic and social benefits. This is why small and medium scale firms have started to express interest in applying green and lean methods, in addition to big manufacturers. It is similar to a win–win situation for producers and the environment that results in enhanced product quality and manufacturing methods, decreased production costs, and increased market shares for a given company by pleasing clients looking for green producers and products [28].

2.3. Resilient Manufacturing

According to Gu et al. [32], resilient manufacturing (RM) refers to a manufacturing system that can sustain severe interruptions and recover from an undesirable condition to a desired state. It is thought to have the power to mitigate the negative impacts of disturbances such as machinery failures and troubleshooting and swiftly return to normal circumstances. Manufacturing disruptions can be divided into internal and external disruptions. The external interruptions that mostly affect the SC include market instability and swings, lengthy lead times, the availability of raw materials, and natural disruptions. Internal interruptions are related to internal interruptions and internal production problems such as equipment failures, safety mishaps, and poor product quality [8]. Since a disruptive event may eventually result in the complete or partial loss of production, businesses must evaluate their vulnerabilities and strengthen their resilience [33] (Heinicke, 2014). Manufacturing companies can become more resilient by deploying strategic safety stocks for redundancy and flexible scheduling and workload distribution for flexibility [34]. Despite the value of resilience in production, the redundancy attribute makes it more expensive and necessitates further investments. Although it is unclear how the payback and return on investments can be monetized, a resilient organization will never fully appreciate the benefit of its resilience efforts because it has no means of determining what disruptions it has stopped [35].

2.4. Lean Manufacturing

For the past 20 years, many businesses, particularly those in the manufacturing industry, have implemented cutting-edge technologies and methods to raise product quality while maintaining the stability of the environment. Anand and Kodali [36] explain that lean manufacturing focuses on enhancing company performance in terms of shorter lead times, better material usage, lower costs, higher-quality goods, and quicker deliveries, which allaid the acquisition of the faith and confidence of customers and make organizations more competitive. According to Dennis [37], the primary objective of implementing a lean system is to obtain high-quality goods and services at the lowest possible price and in the shortest amount of time. According to Panizzolo et al. [38], several operational performances were improved by Indian SMEs’ adoption of lean manufacturing. They divided the performance into upstream, downstream, and value-stream categories. They also discovered that although India is a developing nation with a rising economy, it is still hesitant to adopt lean manufacturing processes for a variety of reasons; one such reason is that doing so would require a significant financial outlay for consulting fees. In their study, Mathur et al. [39] discovered that installing a single-minute exchange of die (SMED) tool in Indian make-to-order SMEs led to a notable increase in production. SMED is a lean approach and is utilized to decrease the time spent executing equipment changeovers to a single digit [40]. In a case study of an Indian SME, Singh and Khanduja [41] assessed how the SMED technique affected setup time. Other approaches, such as poka-yoke, 5S, and many others, also demonstrated strong lean outcomes with financial gain [27].

Furthermore, research by Rothenberg et al. [42], Fercoq et al. [16], and Cherrafi et al. [43] discovered that the application of lean methods might have a favorable effect on environmental performance. According to Thanki and Thakker [27], SMEs should employ integrated lean-green strategies and focus on operational performance to protect the environment. They used the analytical hierarchy process (AHP) to assess how these strategies affected organizational performance. By creating an integrated performance-measuring system, Garza-Reyes [15] and Carvalho et al. [44] validated the gap in the literature. They showed how both lean and green approaches aided in boosting performance. According to King and Lenox [45], lean approaches increase environmental performance by lowering the marginal cost of pollution reduction, which boosts afirm’s ability to compete. According to Yang et al. [46], lean techniques enhance a company’s reputation concerning the environmental element. Garza-Reyes et al. [15] have empirically demonstrated that lean manufacturing techniques increase not only the operational performance but also the environmental performance of the logistics industry in Mexico.

2.5. Identification of SGRL Practices (Input) and Performance Dimensions (Output)

The input and output parameters for the DEA model were identified by conducting an extensive literature review. The SGRL practices were chosen as input parameters. The studies performed by Thanki and Thakker [27], Ahmad et al. [40], and Wadhwa [47] showed how the implementation of integrated lean and green practices affects operational performance and environmental performance. They concluded that the implementation of practices such as environmental management systems, green design, kanban, kaizen, value-stream mapping, etc., is extremely important in order to achieve sustainable performance. They also highlighted that implementing lean and green manufacturing together leads to better performance of SMEs. Similarly, the studies conducted by Belhadi et al. [13], Kamble et al. [19], and Zhou et al. [26] showed that the implementation of practices such as big data analytics, additive manufacturing, cloud computing, etc.,greatly influences the operational, social, and environmental performance of industries. Moreover, the studies conducted by Amjad et al. [8] and Chan et al. [48] showed that implementing practices such as multiple sourcing, contingency planning, risk assessment, etc., prepare industries for unfavorable scenarios and prevent them from shutting down due to sudden interruptions. The SGRL practices were taken as inputs for the DEA model, as it can be seen from other studies that they hugely influence the performance of industries.

Similarly, several key success factors were identified and grouped into three performance dimensions: operational, environmental, and social performance. Parameters such as business process effectiveness, health, and performance; long-term relations with customers; and others are taken as outputs in this study, as previous studies have shown that these dimensions are important in order to achieve sustainable manufacturing [17]. Furthermore, these practices were then reduced to 26 through the analysis of expert opinions. The identified SGRL practices along with their references and assigned codes are displayed in

Table 1. Identified SGRL practices along with references.

SGRL Practices		References
Smart manufacturing (SMP)	Big data analytics	[10,13,19,23,24,25,26,49]
	Additive manufacturing
	Robots and sensors
	Internet of things
	Artificial intelligence and Machine learning
	Cloud computing
Resilient manufacturing (RMP)	Information sharing and supply chain collaboration	[8,32,33,34,35,48]
	Contingency planning
	Multiple sourcing
	Flexible manufacturing system
	Risk assessment
	Strategic stock
Green manufacturing (GMP)	Environmental management system	[40,50,51,52,53,54]
	Green design
	Waste treatment
	Reduce, Reuse, Recycle, Recover, Redesign, and Remanufacture (6Rs)
	Environmental emission control
	Reverse logistics
Lean manufacturing (LMP)	Kanban	[27,40,47,55,56,57,58,59]
	Value-stream mapping
	SMED
	Standardized work
	Six sigma
	Total productive maintenance
	Kaizen
	5S

. In addition, three performance dimensions were identified through a literature review. The performance dimensions along with the assigned codes and references are shown in

Table 2. Identified performance dimensions along with references.

Performance Dimensions		References
Operational Performance (OP)	Long-term relationship with customers	[6,17,60,61,62,63,64,65]
	Customer relationship management (CRM) effectiveness
	Demand uncertainty
	Long-term relationship with suppliers
	SMEs’ effectiveness
	Supply uncertainty
	Business process effectiveness
Environmental Performance (EP)	Effectiveness of environmental system
	Waste reduction
	Reduction in energy consumption and emission
Social Performance (SP)	Corporate social performance (CSR) performance
	Health and performance

.

2.6. Research Gaps

The gaps in previous research studies were identified after a thorough literature analysis, as indicated in Table 3. Several analytical models and indices are available for assessing an organization’s supply chain performance [66]. Typically, vital indicators have been developed to assess an organization’s supply chain performance. The “spider”, “radar”, or “z” chart is the conventional tool for evaluating the effectiveness of a supply chain [67]. The method is graphical, which causes problems with respect to numerous inputs and outputs. A ratio constitutes an alternative approach, but it is challenging to synthesize all the ratios into a single judgment [68].

A model for measuring performance across several elements is necessary to determine a supply chain’s characteristics. The performance of supply chains has been evaluated using multi-criteria-decision-making (MCDM) techniques, such as the Analytic Network Process, Analytic Hierarchy Process, Fuzzy theory, etc. [69]. Still, neither a large number of organizations nor a single organization can be evaluated using this technique. The Supply Chain Operations Reference (SCOR) model and the Balanced Score Card have been used to gauge the effectiveness of supply chains [70]. However, it might be difficult to find precise quantitative statistics regarding the success of SMEs [71]. The supply chain performance methodologies are compared in the Table 3. Regarding the performance measurement of an SME’s supply chain, the current analytical approaches additionally demand that certain constructions and the Critical Success Factor must be quantified. As a result, there is a knowledge gap about regarding the determination of a method with which to assess the performance of SMEs and offer ideas for development.

Table 3. Methods for measurement of performance of SMEs available in literature.

S.No.	Method	Application	Remark	Citation
1	Multi-criteriadecisionmaking	GSCM, Sustainablesupply chain	Only a specific number of case studies can be analyzed using this method.	[69,72,73]
2	Supply Chain Operation Reference (SCOR)	Uses performance metrics for assessment	Cannot be utilized for large samples	[74,75]
3	BalancedScore Card (BSC)	Organizational performance isdetermined via financial aspects	Lack of integration of all supply chain participants and performance dimensions	[76]
4	Life cycle assessment	Carbon footprint performance assessment, environmental supply chain design, assessment of environmental performance	All perspective were not included	[77,78]
5	Fuzzy set approaches	Performance of Supply chain	Cannot be utilized for large samples	[79]
6	DEA	Multi-level DEA used to calculate efficiency of units	Can be utilized for enhancement of performance	[17,66,80]
7	International standardsand composite indicators	When choosing suppliers, ISO 14001 is utilized as evidence of their environmental performance.	Not focused on SMEs	[81]
8	Conceptual framework	Introductory model for evaluating a company’s performance	Case specific	[82]

Table 3. Methods for measurement of performance of SMEs available in literature.

S.No.	Method	Application	Remark	Citation
1	Multi-criteriadecisionmaking	GSCM, Sustainablesupply chain	Only a specific number of case studies can be analyzed using this method.	[69,72,73]
2	Supply Chain Operation Reference (SCOR)	Uses performance metrics for assessment	Cannot be utilized for large samples	[74,75]
3	BalancedScore Card (BSC)	Organizational performance isdetermined via financial aspects	Lack of integration of all supply chain participants and performance dimensions	[76]
4	Life cycle assessment	Carbon footprint performance assessment, environmental supply chain design, assessment of environmental performance	All perspective were not included	[77,78]
5	Fuzzy set approaches	Performance of Supply chain	Cannot be utilized for large samples	[79]
6	DEA	Multi-level DEA used to calculate efficiency of units	Can be utilized for enhancement of performance	[17,66,80]
7	International standardsand composite indicators	When choosing suppliers, ISO 14001 is utilized as evidence of their environmental performance.	Not focused on SMEs	[81]
8	Conceptual framework	Introductory model for evaluating a company’s performance	Case specific	[82]

All the impacts of SGRL manufacturing on the performance of SMEs have been studied separately in studies conducted by Garza-Reyes [15]; Fercoq et al. [16]; De et al. [17]; Ahmad et al. [40]; and Thanki and Thakker [27]. However, there is still a significant knowledge gap regarding the combined effect of SGRL manufacturing on SMEs’ performance, as there havenot been any studies that have considered integrated SGRL manufacturing systems to gauge the performance level of SMEs.

3. Research Methodology

This study necessitated the development of a framework for measuring SMEs’ supply chain performance, taking SGRL practices as input and performance dimensions and, consequently, employing operational environmental and social performance as output criteria. The goal is to observe the impact of SGRL practices on SMEs’ supply chain performance and evaluate the performance of SMEs. The opinions of the managers working at the SMEs are used in this study to quantify the inputs and outputs of the DEA model due to the nature of the criteria. Based on the proper fusion of SGRL practices concerning performance characteristics, the DEA model assists in differentiating between efficient and inefficient SMEs. The ineffective businesses are determined in order to compare their performance to that of their competitors to identify areas that need improvement in SGRL procedures.

Through an extensive review of the literature, a total of 60 SGRL practices and 3 performance factors were initially discovered. Additionally, SGRL practices were reduced to 26 practices based on expert opinions. A questionnaire was developed with the help of a literature review and suggestions given by experts were incorporated into it. One expert from academia and three from the industries were chosen. The experts were selected based on their designation, experience, and knowledge of the latest manufacturing techniques. They rated each practice on the degree of its importance in the manufacturing industries using a five-point Likert scale in which 1 = not important and 5 = highly important. Moreover, an arithmetic means analysis was conducted in which practices with mean scores of less than 3.5 were excluded from further analysis. Finally, 26 practices were found and deemed to be crucial SGRL practices for additional research. Furthermore, another questionnaire was developed for the DEA model, which was discussed in detail in the data collection section. In the following step, a DEA model is developed. The research framework followed is shown in Figure 1.

3.1. Data Envelopment Analysis (DEA)

DEA, developed methodically by Charnes et al. [83], is a nonparametric method for assessing the relative efficiencies of a collection of roughly similar decision-making units (DMUs) by mathematical programming. Using multiple inputs and outputs, the linear programming technique known as DEA produces dynamic, collective, comparative results with which to assess an organization’s productivity [18]. The goal is to create a comprehensive system for measuring supply chain performance that can record overall performance, evaluating the performance of sustainable supply chains using DEA’s capabilities [84]. In DEA, an input–output model must be constructed to prevent the ratio of weights from requiring the precise weight of each criterion. For large businesses, DEA has been used to evaluate the viability of supply chain networks [66]. For large companies, Taticchi et al. [75] developed a supply chain performance evaluation scheme.

In the basic DEA model, it is not necessary to hypothesize a functional form between input and output; instead, one creates a practical input–output set relevant to the DMUs being assessed. For distinct DMUs with a single input and single output, efficient DMUs create a linear piece-wise boundary, which in the DEA’s graphical representation fills a place where other inefficient DMUs’ coordinates (input, output) are located. In other words, the additional inefficient DMU data are contained within this linear piece-wise border, which is why this approach is termed DEA. Both multi-input and multi-output DMUs employ the linear programming approach in the same way. The DEA calculates the greatest efficiency measure for each DMU in comparison to all other DMUs by using this efficient frontier. The piece-wise frontier surrounds all other inefficient DMUs with an efficiency score between 0 and 1, while each efficient DMU is situated with an efficiency value of 1. As a result, DEA creates an effective frontier or boundary that is a piece-wise frontier in line with the facts.

Variable returns to scale (VRS), also known as the Banker, Charnes, and Cooper (BCC) DEA model, is one of many extensions from the literature created from the fundamental Constant Returns to Scale (CRS) DEA model first proposed by Charnes et al. [83]. By relaxing the assumptions of the CRS-DEA model, Banker et al. [85] developed the BCC DEA model. The BCC DEA model is more applicable in a real-world setting. In this research, the “BCC input-oriented model” of DEA proposed by Banker et al. [85] has been taken into account because this model is better suited to real-life problems and since this model is output translationinvariant. In total, there are 30 DMUs, where every DMUj, in which j = 1, 2, …, 30, produces the same 3 outputs (i.e., performance variables), namely, yrj (r = 1, 2, 3), using the same 4 inputs, x_ij (i = 1, 2, 3, 4). The efficiency of a specific DMU can be evaluated by the above BCC, in ‘envelopment form’, as follows:

$${\mathrm{Min}\mathsf{\theta }}_{\mathrm{K}}^{\mathrm{BCC}}$$

(1)

where x are inputs, and y are outputs

$${{\displaystyle \sum }}_{\mathrm{i}=1}^4{{\displaystyle \sum }}_{\mathrm{j}=1}^{30}{\mathsf{\lambda }}_{\mathrm{j}}{\mathrm{x}}_{\mathrm{ij}\le }{\mathrm{Min}\mathsf{\theta }}_{\mathrm{K}}^{\mathrm{BCC}}{\mathrm{x}}_{\mathrm{ik}}$$

(2)

$${{\displaystyle \sum }}_{\mathrm{r}=1}^3{{\displaystyle \sum }}_{\mathrm{j}=1}^{30}{\mathsf{\lambda }}_{\mathrm{j}}{\mathrm{y}}_{\mathrm{rj}\le }{\mathrm{y}}_{\mathrm{rk}}$$

(3)

$${{\displaystyle \sum }}_{\mathrm{j}=1}^{30}{\mathsf{\lambda }}_{\mathrm{j}}=1$$

(4)

$${\mathsf{\lambda }}_{\mathrm{j}}\ge 0 \mathrm{are} \mathrm{all} \mathrm{positive}$$

(5)

where ${\mathsf{\theta }}_{\mathrm{k}}$ is the radial efficiency factor demonstrating the rate of decline with respect to the input levels of firm j;${\mathrm{s}}_{\mathrm{i}}^{-}$ and ${\mathrm{s}}_{\mathrm{r}}^{+}$ are slack variables contributing to extra savings in input i and extra gains in output r. λ_j is the intensity factor depicting the contribution of firm j to the derivation of the efficiency score of firm k. To include all SMEs, the issue mentioned above is repeated 30 times. To determine each SME’s efficiency rating, the corresponding linear programming problem is always solved. The $\mathsf{\theta }$ firms whose solution = 1 are considered relatively efficient or benchmark firms and determine $\mathsf{\theta }$ in terms of the efficient frontier, while firms for which <1 are considered to be inefficient.

The efficiency of each DMU will be calculated using a set of input and output variables by the DEA model depicted in Figure 2. In the DEA model, the performance dimension is regarded as an output variable, and SGRL practices are regarded as an input variable. These parameters were chosen after a review of the literature regarding SMEs’ performance measurement and expert recommendations.

3.2. SWOT Analysis

The SWOT analysis is one of the most used methods by strategic planners in business and industries to ensure that all the elements linked to the projects are recognized and handled. The four elements (strength, weakness, opportunities, and threats), known as strategic factors, determine an organization’s success or failure. The former two factors (strengths and weaknesses) are intrinsic to organizations, whereas the latter two (opportunities and threats) are external factors that cover a wider context, including competition, politics, and the general business environment in which the entity operates [17]. This study uses SWOT analysis to analyze the selected inefficient SMEs and then develop strategies based on the identified strengths, weaknesses, opportunities, and threats. The developed strategies will help the SMEs perform efficiently.

3.3. Data Collection

After reducing SGRL practices from 60 to 26 using expert opinions and by conducting arithmetic mean analysis, as already explained earlier in the research methodology section, the questionnaire (as shown in the Appendix A) for the DEA approach was developed using De et al. [17] as a reference to assess the performance of SMEs. To obtain responses to the questionnaire, 30 manufacturing SMEs registered with the Indian government’s ministry of micro, small, and medium enterprises were identified. The SMEs were selected based on several parameters such as their approachability, type, turnover, and implementation of SGRL practices. The survey was conducted using two respondents from each SME in semi-structured online or offline interviews. Since SMEs have a lower organizational hierarchy than larger organizations, two respondents from each SME were deemed sufficient. Again, the experts were chosen based on their designation, experience, and knowledge of the latest manufacturing practices. The demographic details consisting of respondents’ designation and industry type are shown in Figure 3.

The experts were asked to rate the practices according to the implementation level of these practices in their SMEs using a nine-point Likert scale in which one indicates an extremely low level of implementation while nine indicates an extremely high level of implementation. Moreover, the experts were required to rate the operational, environmental, and social performance of the SMEs using a 1–9 Likert scale in which 1 indicates an extremely low level of performance while 9 indicates an extremely high level of performance. After that, for each objective-type question, the responses from the two respondents were combined, and the mean value was determined.

To run the SMEs performance DEA model, the data were processed in an excel spreadsheet using DEA solver. To distinguish between SMEs that were effective and ineffective, the DEA results were used (using VRS—Variable Return to Scale values). Then, peers were assigned to ineffective SMEs for benchmarking. Following the peer assignment, an analysis of inefficient and peer SME practices was conducted. This analysis is used to develop recommendations for improving inefficient SMEs’ performance by properly balancing the inputs.

4. Results and Analysis

The conceptual DEA model issummarized in Figure 2, wherein the SMEs’ operational, environmental, and social performance are the output variables, and SGRL practices are the input variables. SGRL practices were taken as the input parameters to determine their impact on the operational, environmental, and social performance of the SMEs. A study conducted by De et al. [17] was taken as a reference to construct this model; they used a similar approach to observe the impact of lean and sustainability-oriented practices.

Table 4. Average ratings of input and output parameters for the DEA model.

DMUs	(I)SMP	(I)GMP	(I)RLP	(I)LMP	(O)OP	(O)EP	(O)SP
SME1	3.167	4.167	4.583	5	4.5	4.125	3.5
SME2	4.333	5	4.083	5.5	5	4.25	3
SME3	3.667	4	3.917	3.625	4	3.5	2.75
SME4	2	4	2.5	4.125	3	2.5	1.75
SME5	2.5	4.333	3.833	3.813	3.083	3.125	3
SME6	3.083	3.583	3.333	4.313	4	4.5	3
SME7	3.5	4.417	4.083	4.625	4.5	4.125	3.5
SME8	1.667	2	3.083	2.813	1.833	2	2
SME9	1.5	2	4.5	2.5	1.917	4.125	1.75
SME10	3.5	4.083	4.083	4.875	4.5	5	3.5
SME11	2.167	3.667	2.5	3.625	3.333	3.125	2
SME12	2	3.583	2.333	3.5	3.333	4	2.5
SME13	1.667	3	2	3	2.5	2	2
SME14	3.5	4	4.333	4.313	3.833	4.125	4
SME15	3.5	4.5	3.5	5	4.083	3.25	4.5
SME16	3.833	3.167	4.5	3.125	2.5	3	2
SME19	3.333	4.333	3.5	4.5	4	3	2
SME18	4.083	4	4.917	4.625	4.5	4.25	3.5
SME19	4	4.5	3.583	5	4.083	4.5	3.5
SME20	2.667	3.5	3.5	4.5	3.083	5	3
SME21	2.833	3.833	2.833	4.125	3.833	4.25	3.75
SME22	2	2.833	2.333	3.5	2.5	3	1.75
SME23	3	3	2.917	3.688	3.917	4.25	2.5
SME24	3.5	4	3.083	4.813	4.333	4.5	2.75
SME25	3.083	3.5	3.5	4.5	4	3.75	3
SME26	4.5	4	4.083	5.813	5.083	6	4.5
SME27	4	5	3.833	6.313	5.917	6.5	5
SME28	3	5.5	3.583	4.5	4	3.75	3
SME29	4	4.167	3.667	4.688	4.417	4.25	3.5
SME30	4.167	3.917	4	5	5.167	5	4

displays the information gathered from the 30 SMEs and a questionnaire related to the DEA model. Table 4 shows the average rating given by the experts. The values correspond to the implementation level of the SGRL practices and the degree-of-performance dimension for the SMEs. If we observe the ratings given by the experts, it can be seen that most of the SMEs have a poor or moderate level of implementation of SGRL practices, which results in a moderate or poor level of operational, environmental, and social performance. Next, these ratings are entered in the DEA solver to calculate the performance score of each DMU and to segregate them into efficient and inefficient DMUs.

4.1. Analysis of DEA Result

After data collection, the efficiency of each DMU was determined using the DEA method. The outcome is presented in

Table 5. Performance scores of SMEs.

No.	DMU	Score	Rank
1	SME1	0.9598	18
2	SME2	0.8989	28
3	SME3	1	1
4	SME4	0.9334	23
5	SME5	0.9332	24
6	SME6	0.9455	21
7	SME7	0.9666	16
8	SME8	1	1
9	SME9	1	1
10	SME10	0.9404	22
11	SME11	0.9603	17
12	SME12	1	1
13	SME13	1	1
14	SME14	1	1
15	SME15	1	1
16	SME16	0.9072	26
17	SME19	0.8697	29
18	SME18	0.9464	20
19	SME19	0.8453	30
20	SME20	1	1
21	SME21	1	1
22	SME22	1	1
23	SME23	1	1
24	SME24	0.9679	15
25	SME25	0.9512	19
26	SME26	1	1
27	SME27	1	1
28	SME28	0.9058	27
29	SME29	0.929	25
30	SME30	1	1

, which shows the parameters used to distinguish between the effective and ineffective DMUs. A DMU is inefficient in this situation if its VRS value is less than one, and this rank determines inefficiency in comparison to all other DMUs. The next step in benchmarking is the assignment of peers to each inefficient DMU. The objective is to choose a peer (SME) that shares traits with the ineffective SME to provide recommendations. Then, the peer with the highest lambda value is selected.

Each SME’s score is represented in Table 5 and Figure 4, and the SMEs are ranked by their scores. DEA is a linear programming technique that provides dynamic, collective, comparative results for evaluating the productivity of organizations;therefore, SMEs performing below the level of the top-performing SMEs are indicated by a score of less than 1. Although SMEs with performance scores greater than 0.9 and less than 1 are closer to the group of top-performing SMEs than those with scores less than 0.9, they are still termed inefficient because they can learn from the top-performing SMEs to improve their performance. The SMEs that receive a score of one are ranked first, indicating that they are efficient SMEs who have established an efficient frontier. Other SMEs with scores below one are classified as inefficient SMEs and fall under the efficient frontier envelope. While 16 SMEs were discovered to be inefficient, it was noted that 14 SMEs were found to be efficient. It is worth noting that even though 14 SMEs are efficient there is still space for their improvement, as the degree of implementation of the practices is high in Indian SMEs; however, this study focuseson inefficient SMEs because they are further behind the top group of SMEs in terms of the implementation of these techniques, which can be reflected from theirinferior performance. Additionally, Figure 4 displays the scores that the SMEs attained.

Table 6. Weights of Inputs and Outputs.

No.	DMU	Score	Rank	(I)SMP	(I)GMP	(I)RLP	(I)LMP	(O)OP	(O)EP	(O)SP
1	SME1	0.9598	18	0.16828	0.11208	0.00000	0.00000	0.11075	0.17627	0.00000
2	SME2	0.8989	28	0.00000	0.00000	0.07920	0.12303	0.08983	0.19774	0.00000
3	SME3	1	1	0.00000	0.00000	0.00000	0.27586	0.00000	0.25000	0.00000
4	SME4	0.9334	23	0.50000	0.00000	0.00000	0.00000	−0.33380	0.19988	0.00000
5	SME5	0.9332	24	0.40000	0.00000	0.00000	0.00000	−0.13345	0.00000	0.00000
6	SME6	0.9455	21	0.12801	0.11526	0.01606	0.03219	−0.18736	0.13055	0.00000
7	SME7	0.9666	16	0.08018	0.00000	0.00000	0.15554	0.02076	0.17039	0.00000
8	SME8	1	1	0.18611	0.34488	0.00000	0.00000	−0.03236	0.00000	0.00497
9	SME9	1	1	0.32352	0.25736	0.00000	0.00000	0.00000	0.00000	0.05227
10	SME10	0.9404	22	0.02849	0.00000	0.00000	0.18468	0.06690	0.10181	0.06251
11	SME11	0.9603	17	0.04543	0.00000	0.01210	0.24036	−0.40283	0.16727	0.00000
12	SME12	1	1	0.31413	0.10375	0.00000	0.00000	0.00000	0.30003	0.00000
13	SME13	1	1	0.59975	0.00000	0.00011	0.00000	−0.20099	0.00000	0.00000
14	SME14	1	1	0.02299	0.15656	0.00000	0.06801	0.00000	0.00000	0.00000
15	SME15	1	1	0.00000	0.12192	0.11666	0.00861	0.00000	0.00000	0.00000
16	SME16	0.9072	26	0.00000	0.00000	0.02276	0.28723	−0.53530	0.14876	0.00000
17	SME19	0.8697	29	0.11312	0.00000	0.00000	0.13844	0.08335	0.23826	0.00000
18	SME18	0.9464	20	0.00000	0.04422	0.00000	0.17797	−0.14695	0.10106	0.00000
19	SME19	0.8453	30	0.00000	0.00000	0.00947	0.19321	0.08476	0.07513	0.08253
20	SME20	1	1	0.02280	0.07687	0.10299	0.06882	0.00000	0.00249	0.19846
21	SME21	1	1	0.35298	0.00000	0.00000	0.00000	0.00000	0.00000	0.00000
22	SME22	1	1	0.05247	0.18688	0.15671	0.00000	−0.57951	0.00000	0.14016
23	SME23	1	1	0.00000	0.09044	0.03052	0.17344	0.00000	0.07332	0.09766
24	SME24	0.9679	15	0.00000	0.06508	0.23993	0.00000	−0.20977	0.17496	0.00000
25	SME25	0.9512	19	0.15390	0.13321	0.01694	0.00000	−0.19217	0.14749	0.00000
26	SME26	1	1	0.02729	0.12104	0.09625	0.00000	0.00000	0.00000	0.16667
27	SME27	1	1	0.07579	0.13937	0.00000	0.00000	0.00000	0.00610	0.00000
28	SME28	0.9058	27	0.09292	0.00000	0.00000	0.16027	0.04771	0.21071	0.00000
29	SME29	0.929	25	0.00000	0.00000	0.04073	0.18145	−0.05826	0.14149	0.00000
30	SME30	1	1	0.00000	0.15417	0.09903	0.00000	0.00000	0.18379	0.00000

demonstrates the weights of the input and output variables. In Table 6, the level of each SGRL manufacturing practice for each SME can be observed. Moreover, the quality of performance can also be seen from that table. From Table 6, it can be seen that the poor implementation of these performances has resulted in poor performance overall; regarding the SMEs that have relatively better implementation levels, their performance is better in comparison to the other SMEs. From Table 6 it can be observed that lower implementation of these practices directly resulted in the poor performance of the SMEs. Moreover, the SMEs that had better implementation levelsshowed superior performance and were found to be efficient too. Furthermore, it can be observed from Table 6 that different combinations of these practices led to different levels of performance dimensions. Moreover,

Table 7. SMEs and their peer SMES.

DMU	Peers
SME1	SME9	SME12	SME23	SME27
SME2	SME23	SME27	SME30
SME3	SME3
SME4	SME12	SME13
SME5	SME9	SME21
SME6	SME8	SME9	SME12	SME21	SME23	SME27
SME7	SME12	SME21	SME27	SME30
SME8	SME8
SME9	SME9
SME10	SME9	SME21	SME23	SME27	SME30
SME11	SME9	SME12	SME13	SME23
SME12	SME12
SME13	SME13
SME14	SME14
SME15	SME15
SME16	SME3	SME9	SME13
SME19	SME12	SME23	SME27
SME18	SME3	SME21	SME23	SME30
SME19	SME9	SME21	SME23	SME27	SME30
SME20	SME20
SME21	SME21
SME22	SME22
SME23	SME23
SME24	SME12	SME23	SME27
SME25	SME8	SME9	SME12	SME23	SME27
SME26	SME26
SME27	SME27
SME28	SME12	SME23	SME27	SME30
SME29	SME3	SME21	SME23	SME30
SME30	SME30

presents the peers that were assigned to each inefficient DMU so that they can also improve their performance by implementing the right strategies.

Furthermore,

Table 8. Projection of input variables.

	SMP			GMP			RLP			LMP
DMU	Data	Projection	Diff.(%)	Data	Projection	Diff.(%)	Data	Projection	Diff.(%)	Data	Projection	Diff.(%)
SME1	3.167	3.03977	−4.018	4.167	3.99959	−4.018	4.583	3.09722	−32.419	5	4.66861	−6.628
SME2	4.333	3.81452	−11.966	5	3.91526	−21.695	4.083	3.67017	−10.111	5.5	4.94389	−10.111
SME3	3.667	3.667	0	4	4	0	3.917	3.917	0	3.625	3.625	0
SME4	2	1.86687	−6.656	4	3.34993	−16.252	2.5	2.19987	−12.005	4.125	3.30011	−19.997
SME5	2.5	2.33312	−6.675	4.333	3.14562	−27.403	3.833	3.45807	−9.782	3.813	3.51561	−7.799
SME6	3.083	2.9151	−5.446	3.583	3.38787	−5.446	3.333	3.15148	−5.446	4.313	4.07811	−5.446
SME7	3.5	3.38323	−3.336	4.417	3.81073	−13.726	4.083	3.38965	−16.981	4.625	4.47069	−3.336
SME8	1.667	1.667	0	2	2	0	3.083	3.08297	−0.001	2.813	2.81299	0
SME9	1.5	1.5	0	2	2	0	4.5	4.5	0	2.5	2.5	0
SME10	3.5	3.29133	−5.962	4.083	3.72874	−8.677	4.083	3.38917	−16.993	4.875	4.58436	−5.962
SME11	2.167	2.08106	−3.966	3.667	3.47233	−5.309	2.5	2.40086	−3.966	3.625	3.48124	−3.966
SME12	2	2	0	3.583	3.583	0	2.333	2.333	0	3.5	3.5	0
SME13	1.667	1.667	0	3	2.99999	0	2	2	0	3	2.99999	0
SME14	3.5	3.5	0	4	4	0	4.333	4.333	0	4.313	4.313	0
SME15	3.5	3.5	0	4.5	4.5	0	3.5	3.5	0	5	5	0
SME16	3.833	2.05867	−46.291	3.167	2.60769	−17.661	4.5	4.08241	−9.28	3.125	2.83501	−9.28
SME19	3.333	2.89869	−13.031	4.333	3.3184	−23.416	3.5	2.89117	−17.395	4.5	3.91362	−13.031
SME18	4.083	3.63815	−10.895	4	3.78567	−5.358	4.917	3.59569	−26.872	4.625	4.37718	−5.358
SME19	4	3.02506	−24.373	4.5	3.68801	−18.044	3.583	3.02857	−15.474	5	4.2263	−15.474
SME20	2.667	2.667	0	3.5	3.5	0	3.5	3.5	0	4.5	4.5	0
SME21	2.833	2.833	0	3.833	3.833	0	2.833	2.833	0	4.125	4.125	0
SME22	2	1.99999	−0.001	2.833	2.83298	−0.001	2.333	2.33299	−0.001	3.5	3.49996	−0.001
SME23	3	3	0	3	3	0	2.917	2.917	0	3.688	3.688	0
SME24	3.5	2.93164	−16.239	4	3.87151	−3.212	3.083	2.98397	−3.212	4.813	4.45982	−7.338
SME25	3.083	2.93241	−4.884	3.5	3.32904	−4.884	3.5	3.32904	−4.884	4.5	4.13131	−8.193
SME26	4.5	4.5	0	4	4	0	4.083	4.083	0	5.813	5.813	0
SME27	4	4	0	5	5	0	3.833	3.833	0	6.313	6.313	0
SME28	3	2.71737	−9.421	5.5	3.68734	−32.957	3.583	2.84365	−20.635	4.5	4.07606	−9.421
SME29	4	3.47901	−13.025	4.167	3.68553	−11.554	3.667	3.40647	−7.105	4.688	4.35493	−7.105
SME30	4.167	4.167	0	3.917	3.917	0	4	4	0	5	5	0

and

Table 9. Projection of output variables.

	OP			EP			SP
DMU	Data	Projection	Diff.(%)	Data	Projection	Diff.(%)	Data	Projection	Diff.(%)
SME1	4.5	4.5	0	4.125	5.06797	22.86	3.5	3.5	0
SME2	5	5	0	4.25	5.13782	20.89	3	3.82224	27.408
SME3	4	4	0	3.5	3.5	0	2.75	2.75	0
SME4	3	3	0	2.5	3.20052	28.021	1.75	2.30012	31.435
SME5	3.083	3.1145	1.022	3.125	4.20308	34.499	3	3	0
SME6	4	4	0	4.5	4.55196	1.155	3	3	0
SME7	4.5	4.5	0	4.125	4.6434	12.567	3.5	3.5	0
SME8	1.833	1.833	0	2	2	0	2	2	0
SME9	1.917	1.917	0	4.125	4.125	0	1.75	1.75	0
SME10	4.5	4.5	0	5	5	0	3.5	3.5	0
SME11	3.333	3.333	0	3.125	3.91948	25.423	2	2.4644	23.22
SME12	3.333	3.333	0	4	4	0	2.5	2.5	0
SME13	2.5	2.50001	0	2	2.00006	0.003	2	2	0
SME14	3.833	3.833	0	4.125	4.125	0	4	4	0
SME15	4.083	4.083	0	3.25	3.25	0	4.5	4.5	0
SME16	2.5	2.5	0	3	3.73776	24.592	2	2.02662	1.331
SME19	4	4	0	3	4.42548	47.516	2	2.75099	37.549
SME18	4.5	4.5	0	4.25	4.4584	4.904	3.5	3.5	0
SME19	4.083	4.083	0	4.5	4.5	0	3.5	3.5	0
SME20	3.083	3.083	0	5	5	0	3	3	0
SME21	3.833	3.833	0	4.25	4.25	0	3.75	3.75	0
SME22	2.5	2.5	0	3	3	0	1.75	1.75	0
SME23	3.917	3.917	0	4.25	4.25	0	2.5	2.5	0
SME24	4.333	4.333	0	4.5	4.87686	8.375	2.75	3.30496	20.18
SME25	4	4	0	3.75	4.62328	23.287	3	3	0
SME26	5.083	5.083	0	6	6	0	4.5	4.5	0
SME27	5.917	5.917	0	6.5	6.5	0	5	5	0
SME28	4	4	0	3.75	4.48661	19.643	3	3	0
SME29	4.417	4.417	0	4.25	4.51326	6.194	3.5	3.5	0
SME30	5.167	5.167	0	5	5	0	4	4	0

depict the projection of the input and output variables with respect to becoming more efficient DMUs; projection involves pushing or pulling the output and input variables of inefficient DMUs as per the efficient frontier. Since they are already on the efficiency frontier, the projections of the efficient DMUs are zero. They can push or pull their variables for inefficient DMUs as per Table 8 and Table 9 below. In other words, inefficient DMUs can increase or decrease the level of certain manufacturing practices to become efficient. It is important to note that DMUs must carefully increase their implementation level of SGRL practices as it is costly and should be implemented optimally [17]. Table 8 indicates that DMUs can utilize their existing resources more optimally. For example, SME16 can save its resources by reducing its implementation level of smart manufacturing practices by 46%, green manufacturing practices by 17.6%, resilient manufacturing practices by 9.28 %, and lean manufacturing practices by 9.28 %. Similarly, Table 9 indicates the degree to which the performance of the inefficient DMUs can be improved if they utilize their inputs efficiently. In other words, it shows the possible increase in performance dimensions if SGRL practices are implemented correctly. For example, if SME16 improves its implementation of SGRL practices, it can improve its environmental performance by 24.592 %. Both Table 8 and Table 9 show that implementing the right combination of SGRL practices can lead to better performance.

The average input and output slacks for the inefficient DMUs are shown in

Table 10. Slack values of inputs and outputs.

	Score	Rank	SMP	GMP	RLP	LMP	OP	EP	SP
Average	0.962	12.4667	0.0925	0.1597	0.1392	0.0357	0.001	0.3198	0.1056
Max	1	30	1.419	1.295	1.302	0.55	0.031	1.425	0.822
Min	0.8453	1	0	0	0	0	0	0	0
Std. Dev.	0.0441	11.4342	0.2758	0.3078	0.3197	0.1091	0.0057	0.4275	0.2443

. Slacks are resources that will not be used, resulting in a lack of output. Since efficient units use all of their resources, the slack values of each input and output for an efficient unit should be zero. Therefore, there is no room for error in the input or output variables of effective units. We are focusing on other parameters, which is why these slacks exist. They represent the difference between the current and projected ratings for any parameter.

4.2. SWOT Analysis

To conduct a SWOT analysis, first, one inefficient DMU and another corresponding peered efficient DMU is selected based on certain parameters for the case study. From the group of inefficient DMUs that were obtained from Table 5, DMU11 was selected based on the approachability, industry type, and cooperativeness of the working staff present in this SME. DMU11 manufactures automobile parts, and the automobile industry is a very important type of industry worldwide; hence, it was rated above the other inefficient DMUs. Furthermore, from Table 7, it is evident that DMU9, DMU12, DMU13, and DMU23 are its peered efficient DMUs. Out of them, DMU23 was selected because it also belonged to the automobile sector. Moreover, other parameters such as approachability, the cooperativeness of working staff, and the implementation level of SGRL practices were also considered when selecting DMU23. The SWOT analysis was performed by determining a potential recommendation for DMU11, an inefficient SME; therefore, a case study was conducted by considering DMU11 as a case of an inefficient SME and comparing it with DMU 23, which is an efficient SME. Below is a discussion regarding the comparative analysis.

SME11 should compare itself to SME23 and work to develop SWOT plans for its growth. Based on learning and benchmarking, improvements have been made by this SME.

The goal of the SWOT analysis: SME 11’s strengths, weaknesses, opportunities, and threats must be analyzed to learn from the benchmark company SME 23 and develop strength–opportunity, strength–threat, weakness–opportunity, and weakness–threat strategies.

Strengths: The company’s long-standing relationships with its clients constitute a strength. These guarantee a steady market and lower demand uncertainty.

Weaknesses: This business has trouble achieving throughput and meeting delivery deadlines. Additionally, there are gaps in management communication. As a result, more products are rejected. To minimize the movement of their workforce and materials, SMEs must reorganize their plant layout. Large investments are needed for theimprovements.

Opportunities: This SME has a healthy market base and good demand. Due to newer technology, SMEs have a chance to draw in more clients. The ability of a small business to produce a variety of products gives it a competitive advantage.

Threats: The labor market poses a threat to SMEs. There are a few unresolved managerial and employee-related issues that cannot be effectively negotiated because of union threats. The market is incredibly cutthroat. The suppliers occasionally bring uncertainty with respect to the supply of material.

Strength–opportunity strategy: It is necessary to manage long-term customer relationships. This will facilitate the formation of strategic alliances with a select group of clients.

Strength–threat strategy: Information sharing and supply chain collaboration are necessary. These activities will facilitate improved supplier communication and lessen supplier uncertainty. This company needs to conduct risk assessments regularly. This SME can work with rivals to participate ingreater procurement and tendering.

Weakness–opportunity: This company should prioritize enhancing product quality and reducing its rejection rate. The company should place emphasis on flexible manufacturing systems. Correct Six Sigma implementation and work standardization will aid this SME in resolving this issue.

Weaknesses and threats: The workforce must be trained to cope with uncertainty. Big data analytics can be used to further help workers with respect to uncertainty. The projects from the aforementioned SWOT analysis indicate the need for the implementation of process improvement strategies, specifically with respect toprocess improvement that results in the creation of supplier and customer relationships and inventory reduction. It is necessary to improve quality and forge long-term relationships. A skill development program for the workforce through appropriate training must be part of the improvement program. Integrating the process at various supply chain levels requires the implementation of information sharing and supply chain collaboration.

The results of the SWOT analysis highlight how SMEs can learn from their peer SMEs and implement the right SGRL practices to solve their problems and improve their performance.

5. Discussion

Combining SGRL practices has a more significant effect on performance. There is an unmistakable consensus that Industry 4.0 technologies positively impact the three paradigms of green, resilience, and lean practices. Industry 4.0 and the three strategies complement one another. Numerous researchers have argued that SM can aid in overcoming the difficulties that GMsand cleaner production face, including Zhang et al. [23] and Belhadi et al. [13].The current study was able to show how SGRL practices affect the performance of SMEs’ supply chains. This study also highlights the effects of SGRL practices and offers a model for assessing the performance of SMEs. From the results of the DEA, it can be observed that the implementation of these practices at lower levels directly resulted in the poorer performance of the SMEs. Moreover, the SMEs that implemented these practices to a greater extent had better performance and were more efficient. Furthermore, it can be observed that a variation in the combination of these practices affects the final performance of a given SME. The findings of the DEA analysis were found to bein concordance with the study conducted by De et al. [17], which highlighted a similar conclusion for only lean practices and sustainability-oriented innovation.

The SWOT analysis and the case study show that SGRL can improve SMEs’ performance. In addition, studies have concurred that SGRL practices enhance performance [6,7]. The case study makes it abundantly clear that SGRL practices help address the sustainability of SMEs’ supply chains. The comparison of the case studies demonstrates that SME11 can boost its productivity by adopting manufacturing techniques that SM23 has already adopted.

The suggested framework assists SMEs in implementing appropriate strategies and is intended to enhance their performance. Different SGRL practices are considered input criteria, and operational, environmental, and social criteria are considered outputs. Due to the impossibility of obtaining secondary data on these criteria, this study used interviews with key personnel from a group of SMEs in a particular area (the northern part of India) to collect primary data. In this research, DEA is used to analyze the effectiveness of the participating SMEs regarding SGRL practices and sustainability performance. When compared to other contemporary approaches, such as multiple criteria decision-making methods, DEA is the most appropriate method for analyzing the efficiency of SMEs due to the nature of the problem and the characteristics of the criteria for the suggested framework of evaluation (e.g., the AHP, ANP, Fuzzy, etc.). The findings show that SMEs can enhance their performance by implementing SGRL practices.

The suggested DEA-based framework for analyzing the impact of SGRL practices on the performance of supply chains aids SMEs in achieving optimal performance through the adoption of the proper combination of SGRL practices. This framework helps policymakers’ group effective SMEs and recommend improvement measures for ineffective SMEs by providing funding and facilitating their company’s growth via determining how to improve their performance. By benchmarkingthemselvesagainst their peers, each SME can use this framework to determine their unique state of optimal performance and, if necessary, implement improvement measures. This framework has been successful in assisting policymakers to appropriately compare themselves with their peers. Based on the findings and methodology described above, this study assists managers and policymakers in three ways. First, it helps them concentrate on separating SMEs into efficient and inefficient SMEs. Second, after learning from the benchmark SMEs, the SMEs can then compare themselves to relevant peer SMEs and develop their SWOT-derived strategies and projects. Third, when they interact with benchmark SMEs, they learn about common interactions and will be inspired to develop their improvement projects.

6. Conclusions

The combination of the DEA method and SWOT analysis was used in this study to separate SMEs and relate their incorporation of SGRL practicesto their corresponding reported performance.The study demonstrated how SGRL techniques can help SMEs obtain optimal performance and provided suggestions in this regard. This study helps SMEs select which SGRL practices to employ and offers solutions with respect to the SGRL practices taken into consideration for performance improvement.

This study’s implications, given from a managerial perspective, are presented in the following sentences. The framework can be used as a tool that SMEs can use to improve their supply chain and benchmark performance. Peer benchmarking and the aforementioned recommendations can aid SMEs in strengthening their performance. This paper thoroughly explains how to use the framework as a tool. This will give decision-makers and other interested parties a clear evaluative mechanism that can be used to make the necessary adjustments. This study adds to the literature regarding performance measurement for SMEs.

Furthermore, this study provides the theoretical implication that an SME’slevel of implementation of SGRL practices directly impacts its performance, and that different combinations of SGRL practices affect this performance differently. Thus, it is important to implement the right strategy, which can be achieved by learning from better-performing or more efficient SMEs.

Data Envelopment Model: This specific DEA model can assist SMEs in gauging their effectiveness and classifying themselves in accordance with their SGRL procedures. This study provides a framework for analyzing SMEs’ effectiveness while considering SGRL practices. The ability to apply the framework to divide SMEs into efficient and inefficient SMEs will be of interest to policymakers, consortiums, and clusters of SMEs.

This approach provides a condensed summary of SMEs’ characteristics in a region where inefficient SMEs have been discovered, pairing efficient SMEs with inefficient SMEs in this region and improving their performance by attempting to strike an appropriate balance between SGRL practices. DEA supports collaboration and partnering with the right peers. The framework aids in the official adoption, implementation, and exploitation of best practices determined to be suitable and advantageous from benchmarking and anSWOT analysis.

7. Limitations and Future Research Directions

This study puts forth a framework for categorizing SMEs’ performance efficiency according to their SGRL practices. However, the study has limitations that prevent its generalization due to the industry’s nature, size, and sample. The study was carried out in India’s northern region. Due to variations with respect to SGRL implementation across different economies, the SMEs considered are Indian manufacturing SMEs, and their applicability in terms of results and implications to other nations may not be guaranteed. However, these results are applicable in countries where SMEs have implemented SGRL practices at a level comparable to that of India. Research can be carried out for SMEs in different countries using the methodology provided in this study. Moreover, the methodology used assumes that managers working in SMEs are unbiased, possess proper domain knowledge, and have filled in the questionnaires appropriately, which may not always be the case. Also, this study has considered 30 SMEs judgmentally. The parameters mentioned in the above sections were defined for the selection of SMEs; however, the methodology used does not employ any fuzziness or probability distribution, which can be used in future studies. This research has highlighted the implications for potential future research directions. Future studies may compare various regions and economies in terms of evaluating theirSMEs’ performance based on SGRL implementation. The analysis of SGRL practices and their effects on the performance of different sectors of industries will be fascinating; furthermore, research can be conducted for different sectors with respect to how SGRL practices affect the performance of different sectors of industries.