An integrated approach to explore the barriers to lean manufacturing in the context of the COVID-19 pandemic: implications for sustainability

2.1 Lean manufacturing practices in apparel sector during COVID-19

Lean manufacturing is a methodology consisting of a collection of management techniques and the acquisition of manufacturing operations that may be utilized to eliminate waste and minimize the uncertainty of suppliers, customers and internal resources and processes (Bashar et al., 2021; Hasan et al., 2022). However, the pandemic has had a negative effect on economic growth since most countries have spent less on capital projects to properly equip their health sector, reducing investments in other sectors (Béland et al., 2020). The outbreak led to a modification in work practices intended at protecting the health of employees. Moreover, researchers are paying more attention to management techniques that promote continuous development of companies, such as LM in this pandemic time (Tortorella et al., 2021). Investigations on LM have found benefits across a wide range of areas, including cost savings, better quality, shorter delivery times and increased employee performance (Piercy and Rich, 2009; Laureani and Antony, 2010; Carlborg et al., 2013; Hadid and Mansouri, 2014; Ojasalo and Ojasalo, 2018; Cavdur et al., 2019). In addition, lean concepts have made textile and clothing businesses more sustainable and competitive (George et al., 2022). As a result, the anticipation of gaining these advantages, along with the need for firms to be more competitive, has encouraged industrial organizations all over the world to widely implement LM (Buer et al., 2021).

Bangladesh's apparel industry is essential to the country's economic and social progress. After China, Bangladesh is currently the world's second-largest exporter of clothing (Khan and Ullah, 2017). After beginning its development in the 1970s, Bangladesh's apparel industry witnessed tremendous expansion, becoming the multibillion-dollar, export-focused business it is today. The total earning from the export of apparel goods in 1984–85 was 31.57mn USD (Bashar and Hasin, 2019). Although it may seem smooth sailing, the recent pandemic has hindered this growth severely. According to BGMEA, Bangladesh's apparel export in the calendar year of 2020 is 27.471bn USD, significantly lower than 33.073bn USD in the calendar year of 2019 (Swazan and Das, 2022). Also, this has been the lowest level of apparel export since 2015 (Swazan and Das, 2022). Therefore, lean practices in Bangladesh have been suffering already. In the middle of this pandemonium, the desire to implement lean has waned, creating additional barriers on top of those already in place (Islam, 2020).

2.2 Barriers affecting the lean implementation in COVID-19 pandemic

A list of barriers was proposed based on the literature review of Kabir et al. (2021), Parmar and Desai (2020), Basher et al. (2022), Mishra et al. (2021), Abu et al. (2019), Singh and Rathi (2021), Bashar and Hasin (2019), which consists of several barriers and challenges to lean implementation. Kabir et al. (2021) claims that LM may suffer from a lack of internal and external financing. Moreover, lack of intensive communication is also mentioned as a barrier by the authors. According to Singh and Rathi (2021), weak technological infrastructure has proven to be a major challenge to support any lean activities in the apparel sector. Parmar and Desai (2020), stated that lack of strategic reward and performance management prevent employees from participating in lean and continuous improvement activities. Bhadu et al. (2022) and Maware and Parsley (2022) found that organizations are often monolithic in nature and less willing to change management structure according to changing times. Mishra et al. (2021) identified insufficient time as a drawback to leanness, especially in the context of COVID-19. The authors also mentioned inadequate skills of worker and negative attitude toward lean hinders lean implementation. Bashar and Hasin (2019) maintained that employees often lack the core understanding of lean concept and how to implement lean tools. Thus, lack of proper understanding of lean concept is identified as a barrier. On the other hand, Abu et al. (2019) and Hasan et al. (2022) both mentioned improper selection and customization of lean tools as a significant cause preventing lean implementation. A total of 17 critical lean implementation barriers were chosen based on existing research and expert opinion that have been shown in Table 1.

2.3 Existing methods and rationale behind the proposed method

In previous research studies, a number of lean implementation methodologies were proposed. Researchers and academics have proposed a number of models to help management and professionals understand how to implement an effective improvement strategy. The vast majority of these models are founded on various theories of multi-criteria decision-making including the analytical hierarchal process (AHP), analytical network process (ANP) (Wong et al., 2014), Interpretive Structural Modeling (ISM) (Jadhav et al., 2014; Bhadu et al., 2022), Decision Making Trial And Evaluation Laboratory (DEMATEL) approach (Singh and Rathi, 2021), etc. On the other hand, some literature have focused and performed their studies based on total interpretive structural modeling (TISM). Such as, Ramiya and Suresh (2021), Chaple et al. (2021) and Vinodh (2020) have performed some studies based on modeling the barriers of lean implementation utilizing the TISM method. In addition, Jain et al. (2023) have devised a TISM methodology to analyze the factors that would make it easier for Indian manufacturing facilities to undertake mass customization. In this research, a novel framework integrating Delphi method and fuzzy TISM has been proposed for examining and modeling the barriers of lean implementation. The purpose of the Delphi approach is to improve the reliability of subjective evaluations by gathering qualitative data and opinions from experts via a standardized questionnaire and a series of face-to-face sessions (Hasson and Keeney, 2011). The justification for employing the Delphi approach in the present study is that lean implementation is a complicated challenge requiring systematic analysis of expert opinions that are prone to subjectivity (Goodman, 1987). Although the Delphi approach yields a better expert opinion, one of its drawbacks is that it cannot be used to identify the causal relationships between the variables. The suggested decision support framework overcomes this shortcoming by combining the Delphi technique with fuzzy TISM. To our knowledge, no prior study has, however, suggested using the fuzzy TISM model based on the Delphi method to analyze the barriers to LM in the context of the COVID-19 pandemic. Such integration is a unique strategy that also strengthens the process by obtaining more precise feedback from the experts, increasing its robustness.

2.4 Research gaps and contributions

Numerous studies have explored how lean can be implemented in the past. For example, Salonitis and Tsinopoulos (2016) have made an effort to identify the barriers and drivers of lean implementation in the Greek manufacturing industry. Shah and Hussain (2016) and Mohan Prasad et al. (2020) also performed similar studies on the textile industry in Pakistan and India, respectively. Hasan et al. (2022) put their effort into finding out the current situation in lean implementation in the Bangladesh apparel industry. Despite the fact that a number of studies have been done analyzing the impact of COVID-19 on Bangladesh's ready-made garment industry, they were mostly concerned about its impact on garments workers. Sen and Antara N (2020) and Kabir et al. (2021) tried to evaluate the impact of COVID-19 on Bangladeshi garments workers. But there were no studies conducted regarding the new challenges the novel coronavirus has brought in lean implementation. After an extensive literature review, it was found that there was very little light shed on the barriers to implementing lean in Bangladesh in the wake of a new virus, COVID-19. Lean implementation was already a challenge in the apparel industry of Bangladesh. But the novel coronavirus has brought new challenges to this already complex problem. In this study, the writers categorized and ranked the barriers of lean implementation in the Bangladesh apparel industry.

3.1 The Delphi technique

The first phase of data collection consisted of Delphi method where the relevant barriers of lean implementation were finalized. The Delphi method needs between 12 and 15 respondents to ensure consensus (Linstone and Turoff, 1975; Vogel et al., 2019). So, an expert panel including 12 professionals and five academicians was selected through purposive sampling method (Linstone and Turoff, 1975). The inclusion criteria for experts were years of experience, expertise in the relevant field and employment (Linstone and Turoff, 1975). The resulting panel of experts comprised academics and business people with more than 10 years of expertise in LM. The backgrounds of these eminent individuals are shown in Table 2 below. In this Delphi phase, three rounds of meeting were conducted. In the first round, the list of initially identified barriers of lean implementation from the literature was analyzed in relation to RMG industry of growing economy during the COVID-19 pandemic. In the second round, the experts were allowed to add new barriers and remove any irrelevant barriers from the list. In the final round, 11 barriers were selected for further analysis. The list of barriers that have been completed and are relevant to the setting of the rising economy is shown in Table 3.

3.2 Fuzzy TISM method

4.1 Results of the study

This section represents the findings of the application of the proposed fuzzy-TISM framework to explain the relationships among the barriers of lean implementation in Bangladesh's RMG business in relation to COVID-19. In order to determine the link between the factors connected to the study work's purpose, an expert committee made up of academic and industrial specialists was formed (Step 1). From the previously published studies and expert recommendations as stated in Section 3.1, 11 barriers have been identified (Step 2). In the view of the professionals, the fuzzy structural self-integrated matrix (SSIM) of the barriers of lean implementation was developed for each expert (Step 3). Fuzzy linguistic scale from Table S1 was used to define the influential relationship between two barriers. The aggregated fuzzy SSIM as represented in Table S2 was constructed by taking the mean value of the responses from the experts. The interpretive matrix was developed from the aggregated fuzzy SSIM as shown in Table S3 (Step 4). The following stage included replacing the linguistic variables with the relevant triangular fuzzy number to create the final fuzzy reachability matrix from the aggregated fuzzy SSIM (Step 4). Table S4 represents the final fuzzy reachability matrix of the barriers of lean implementation. The final reachability matrix was used to compute each barrier's driving and dependency power (Step 5). Table 4 shows the value of driving power and dependence power of each barrier. According to the driving power, the top five barriers of lean implementation are B4, B2, B9, B1 and B7. Again, based on the dependence power values, the top five barriers are B5, B10, B11, B7 and B8. Based on these data, Fuzzy MICMAC analysis has also been conducted (Step 5). Figure 2 shows the result of the fuzzy MICMAC analysis. Fuzzy MICMAC has divided the barriers into two categories: linkage and dependent. B1, B2, B4 and B9 have been grouped into linkage category. On the other hand, dependent category includes B3, B5, B6, B10 and B11.

Hierarchical relationship diagraph is one of the major outcomes of fuzzy TISM. To construct the diagraph, the defuzzified initial reachability matrix was created using the combined fuzzy SSIM according to Jain and Soni (2019). In Step 6, linguistic expressions “Very high” and “High” were expressed as “1” and “Very low”, “Low” and “No” were expressed as “0”. Table S5 represents the defuzzified initial reachability matrix. Then the final reachability matrix was built after checking and defuzzing the transitivity ties between the barriers as shown in Table S6 (Step 6). Level partitioning has been performed from the defuzzified final reachability matrix to explain the relationships among the barriers in a hierarchical form. Table S7 represents the level partitioning of barriers of lean implementation.

From Table S7 it can be seen that B3 and B5 are at Level 1. B10 and B11 are found at Level 2. Level 3 consists of B6 and B8. Level 4 and Level 5 include B7 and B1, respectively. Finally, B2, B4 and B9 are found at Level 6. Figure 3 shows the fuzzy TISM digraph which is the graphical representation of barriers of LM in level. The influential relationships among the barriers were represents through links and different levels in this fuzzy TISM diagraph (Step 7).

4.2 Validation of results

In this part of the study, the ranking of the barriers, fuzzy MICMAC matrix and fuzzy TISM diagraph were further validated with the help of industry practitioner. A series of focus group discussions were conducted in three phases, which involved 12 industrial managers to validate the research findings. The selection criteria for the participants included their years of working experience, current organizational affiliation and area of expertise. Specifically, participants were chosen based on having at least 15 years of working experience, currently working in a manufacturing organization and possessing knowledge of LM principles. In the first phase, the ranking of lean implementation barriers in the context of post COVID-19 was given to the participants for validation. After the discussion, the participants reached a consensus on the ranking of barriers. During the second phase, the participants were asked to validate the barriers within each class of fuzzy MICMAC matrix. This was done to ensure that the barriers identified in the first phase were accurately categorized and accounted for within the classes. Finally, the fuzzy TISM diagram was presented to the participants along with a hierarchical relationship among the different barriers. The participants were asked for their opinions on the direct and indirect relationships among the barriers, and whether they were appropriate in the context of post COVID-19. The participants expressed that they found the direct and indirect relationships among the barriers to be appropriate and relevant in the current context. Overall, this phase helped to further validate and refine the identified barriers, and provided a deeper understanding of the relationships among them.

5.1 Discussion of the findings

The findings from the application of the proposed methodology provide some insights about lean implementation challenges in apparel business of Bangladesh in the context of COVID-19. The findings reveal that “lack of synchronization of lean planning with the strategic planning”, “lack of understanding of lean concept” and “lack of priority from the top management” are the three top most important barriers of LM implementation in apparel industry of Bangladesh during the pandemic of COVID-19. A decision-support framework to analyze and model the influential relationships among the barriers has been presented in Figures 2 and 3. These findings will help the apparel sector to formulate strategy for implementing the LM practices successfully.

In accordance with the data shown in Table 4, B4 is the most significant barrier according to the driving power value to implement LM in RMG industry. Most of the RMG companies have to change their strategic planning to mitigate the disruptions resulted from the recent COVID-19 pandemic. However, most of the RMG companies of emerging economy like Bangladesh are failing to synchronize the LM strategy with their company strategy. Improper alignment of LM strategy with the company strategy cannot bring the intended benefits for the companies (da Silva et al., 2021). The second most significant barrier has been identified as B2. The traditional LM must be rethought by the RMG companies as the industry context has been changed significantly after the pandemic. Companies of emerging economy are trying to initiate LM tools to tackle these business transformations in the traditional way. However, lack of knowledge about the LM concept makes them unable to adopt new ways to implement the lean successfully in the context of COVID-19 (Liao and Wang, 2021). Moreover, due to lack of sufficient understanding and knowledge, companies' top management usually fails to give enough priority to adopt LM concept. This study has also found the B9 as the third most significant barrier to implement LM concept. Successful implementation of LM requires high investment priority and close attention from the top management (Vanichchinchai, 2021). However, after the COVID-19 outbreak, the focus of most RMG firms in developing economies has switched away from operational strategy and toward finance and market strategy. From Figure 2, it can be seen that all these three barriers are in the linkage cluster as they have close relationship with the other barriers. Moreover, Figure 3 implies that these barriers have influential relationship among themselves as well as with other barriers.

During the COVID-19 pandemic, the successful implementation of LM concept requires substantial amount of internal and external funding as most companies are facing financial crisis in emerging economy. However, lack of proper strategy and knowledge as well as low priority from the companies can lead to B1. Figure 3 shows that B2, B4 and B9 directly influence B1. Moreover, this barrier has been found as the fourth significant barriers according to the driving power value. Leite et al. (2022) has also found that lack of sufficient financial support is one of the major causes of lean implementation failure. Again, successful implementation of LM concept requires active participation of employees and workers of all the levels of the company. Therefore, proper motivation is needed through incentives and rewards to engage them for implementing the LM. However, B7 will fail to keep them motivated and eventually, the implementation of LM will be in vain. Vinodh (2020) also identified that lack of proper reward system is one of the major barriers to implement LM successfully. Figure 2 shows that B7 is also grouped into linkage cluster as this barrier has significant relationships with other barriers. From Figure 3 it can also be seen that B7 directly influences the B8. Skilled employees of managerial levels play significant roles in successful lean implementation as they coordinate the overall implementation process and work as a bridge between the top management and lower level employees in the company (Singh and Rathi, 2021).

Recently, many RMG companies in the emerging economy are trying to adopt modern technologies to improve their operational performances during the COVID-19 pandemic. However, most of them have failed due to lack of sufficient funding and resources. This study also revealed that B1 influences the B6 as shown in Figure 3. According to Abu et al. (2021), weak technological capabilities of the company can adversely affect the successful implementation of LM. During the lean implementation process, it is required to continuously track the progress and analyze the data. Figure 3 depicts that B3 is resulted from the B6. Figure 3 also shows that B6 and B8 lead to B11. Moreover, B10 is resulted from B7. Effective leadership reconciles the successful lean implementation strategies with the company strategy (Singh and Rathi, 2021). Again, there is lack of experts and effective trainers in emerging economy for implementing the LM successfully (Gaikwad et al., 2020). However, these two barriers are often resulted from the previously discussed barriers. This study has also found these two barriers as the second most dependent barriers as shown in Table 4. Finally, lean concepts and tools need to be customized according to the context of emerging economy. However, B10 and B11 can lead to B5 as shown in Figure 3. This has been found as the most dependent barriers while implementing LM concept in RMG industry of Bangladesh. According to the fuzzy MICMAC as shown in Figure 2, B3, B5, B6, B10 and B11 fall into dependent category. While analyzing the fuzzy TISM diagraph, it has been found that all these five barriers are somewhat influenced by the other barriers.

5.2 Theoretical and managerial implications

To the authors' knowledge, no prior work has combined Delphi technique, Fuzzy set theories, TISM model and LM to suggest an evaluation strategy. As a result, this study contributes to the theory of many disciplines since it unites a wide range of issues into one. From the perspective of the COVID-19 epidemic, it offers a comprehensive picture of lean implementation barriers and challenges faced by Bangladesh RMG industry. Based on apparel sector of Bangladesh, there has only been a little amount of study conducted on these problems (Hasan et al., 2022; Bashar et al., 2021; Kabir et al., 2021; Munim et al., 2022). There was no assessment of the barriers imposed by COVID-19 by the writers of those papers. In order to better understand the relationship between the application of lean approaches and the barriers that stand in the way of doing so, the authors of this study set out to look into the COVID-19 pandemic and its effects on the organizations. Authors have developed a structural model that interprets the transitive and causal connections of the lean implementation barriers in RMG industry of Bangladesh, and also outlines how those elements interrelate with one another by utilizing Fuzzy-TISM. In terms of theoretical contribution, it highlights a new set of lean implementation barriers and numerous cause-effect correlations to assist minimize the pandemic impact of LM implementation. As a final point, the article included a full explanation of the construction process and suggestions for its implementation. Academics and researchers may use this study to learn more about how this method could be used in different contexts.

The current study highlights possible methods that managers in the RMG sector of Bangladesh may take to assist to reduce the COVID-19 pandemic's effects on the RMG sector by adopting LM. Many professionals working in this field will benefit from the results of this research since it shows them how their manufacturing processes may be affected by the COVID-19 pandemic and how they may reduce such effects. Additionally, Most of the RMG companies have to change their strategic planning to mitigate the disruptions resulted from the recent COVID-19 pandemic (Hasan et al., 2022). However, most of the RMG companies of emerging economy like Bangladesh are failing to synchronize the LM strategy with their company strategy (Hasan et al., 2022). Improper alignment of LM strategy with the company strategy cannot bring the intended benefits for the companies (da Silva et al., 2021). Therefore, policy makers should prioritize the implementation of lean in their industries for the overall benefit. As a result, it is suggested that managers prioritize the barriers and challenges and develop suitable methods to overcome them in order to adopt LM successfully. So, the findings of the study may aid companies in making decisions about developing distinctive lean implementations and transforming current systems into competitive ones.

5.3 Implications for sustainable development goals

The model, constructed on the suggested fuzzy TISM framework, will aid in the formulation of successful strategies for the introduction of LM and, by extension, the attainment of sustainability goals (SDGs). The second most significant barrier has been identified as “Weak technological infrastructure (B2), one of the influential barriers of LM. During the COVID-19 pandemic, the successful implementation of LM concept requires substantial amount of internal and external funding as most companies are facing financial crisis (“Lack of internal and external funding -B1”) in emerging economy. As a result, businesses may receive assistance in building LM infrastructures and encouraging innovation during the COVID-19 pandemic, all of which will lead to the growth of long-term, sustainable Industry, Innovation, and Infrastructure (SDG 9). Companies can promote awareness of advantages of LM in their organization by incorporating proper consultants and trainers which (“Lack of awareness of the advantages-B8” and “Lack of efficient consultants and trainers-B6”) will aid in developing resilient LM and lowering product waste, two key components of Responsible Consumption and Production (SDG 12). Lean philosophy encourages maintaining a model of organization focused on elimination of waste to reduce the overall costs, ensuring timely delivery of quality products and materials while respecting people and the environment. So, finding and removing the appropriate barriers of LM implementation can help achieve sustainability in the RMG industry which can facilitate a sustainable transition toward the successful implementation of lean.

As, achieving increasing levels of economic productivity via diversification, technical upgrading and innovation, notably by concentrating on high-value added and labor-intensive industries, is an aim of sustainable development goal (SDG 8) so it can be obtained by proper lean implementation in an organization. Thus, the proposed model is expected to contribute to Responsible Consumption and Production (SDG 12); Decent Work and Economic Growth (SDG 8); Industry, Innovation and Infrastructure (SDG 9) via resilient strategies.