The Academic Development Trajectories of the Lean Production Based on Main Path Analysis Method

Abstract

Enterprises looking to be competitive are constantly looking for a continuous increase in productivity, quality, and level of services. With the development of the industry 4.0 concept, manufacturers are more confident about the new advantages of automation and systems integration. Lean management is a well-developed and empirically proven managerial strategy. Combining lean and industry 4.0 practices seems to be a necessary evolutionary step to further raise the level of operational excellence. This study applied the main path analysis method to explore the development trend of lean management in the academic field. First, this study adopted the Scopus database to collect relevant papers, then analyzed their overall development trajectory by using Main Path 437 software, and used the g-index and h-index to identify more influential journals. Next, this study clustered the papers with similar topics into several groups, and then used Wordle software to present the keywords of each group in a word cloud that serves as a reference for naming. Thus, the top five groups obtained are as follows: “Lean production concept and application”, “Lean Six Sigma concept and application”, “Lean system integration and application”, “Lean construction concept and application”, and “Lean healthcare concept and application”. Finally, this study provides explanations and conclusions on each group’s development trajectory, as well as research recommendations in the field of lean production. The findings can serve as guides for industry, government, and academia as they develop future lean production development strategies. This study utilized an integrated analysis approach to successfully and effectively depict the trajectory of lean production development and applications, identify future research and development directions, and generate technological forecasts.

1. Introduction

Industry 4.0 relies on machines’ ability to quickly collect, analyze, and exchange enormous data sets. Modern technologies such as, Cyber-Physical Systems (CPS) or the internet of things (IoT) provide for a faster, more flexible problem-solving and more efficient value production while reducing expenses [1]. In consideration of industry 4.0 and contemporary information and communication technologies, businesses must define procedures in an effective manner before automating industrial processes. industry 4.0’s deployment of horizontal and vertical networking provides a more seamless integration of customers and suppliers in the (value-adding) process. These processes are significantly influenced by vertical and horizontal integration [2]. Therefore, industry 4.0 demands a certain amount of process orientation with well-defined processes, suppliers, customers, tasks, and times. A fundamental feature and objective of lean production systems is the development and implementation of waste-free, standardized, and efficient processes with a strong customer focus. As a result, lean production becomes an essential building block for the industry 4.0 infrastructure [3,4]. Combining lean and industry 4.0 practices seems to be a necessary evolutionary step to further raise the level of operational excellence.

The Toyota Production System (TPS) has been used in Japan since the 1950s, and in the West since the 1980s, under the name lean management. Lean management was proposed by the Massachusetts Institute of Technology. In a research project called the “International Automobile Project”, they found that Japan’s Toyota Motor Corporation’s production organization and management method are the most suitable production model for modern manufacturing through a large number of surveys and comparisons of Japanese companies. The goal of this production model is to reduce production costs by optimizing the process, improving the coordination of the production process, completely eliminating all waste in the enterprise to improve production efficiency, and using the least amount of work to create the highest economic value for the end consumer, which is called lean manufacturing or lean production. The main methods used by lean manufacturing can be summarized as follows.

Numerous reviews of the literature on lean production have been conducted by academics. However, the majority of these reviews have concentrated solely on a single lean production strategy or application. In addition, the limitations of the analysis methods employed by these scholars have resulted in a relatively small number of papers being evaluated. Therefore, the outcomes of the aforementioned studies cannot be generalized to all lean production applications. Despite the fact that the aforementioned researchers have identified a number of development trends, the overall development trajectory of lean production warrants further investigation. In this study, a comprehensive review of lean production and its applications was conducted to identify the overall development trajectory and emerging production research topics on lean production.

2. Literature Review

Numerous academics have published works on lean manufacturing. The authors introduced the idea of lean production, developed a model, and applied it to the manufacturing sector in [5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37]. Among them, Emiliani [4,5,6,7] emphasized the need to simultaneously change business processes and employee personal behaviors in order to successfully implement lean production, with a focus on enhancing employee personal behavior models. The suggested model, which used lean production to reduce individual behavior and waste in the production process, combined behavioral science, philosophy, economics, and industrial engineering. References [28,29,30,31,32] mainly discussed the impact that enterprises will face when implementing lean production. They aimed at small and medium-sized enterprises and large enterprises of “Just In Time (JIT)” and “Waste Minimization “. The three major aspects of “Just In Time (JIT)”, “Waste Minimization” and “Flow Management” were discussed and compared. The research results pointed out that the impact of JIT production on operational performance is more significant in large companies, while waste minimization is more significant in small and medium companies. In addition, flow management is not significant in small, medium, and large companies. The connection between internal awareness and the use of lean thinking was examined by Malik et al. [34] and Amaro et al. [35]. Their study’s findings demonstrated that internal interactions inside an organization can enhance organizational awareness, and that this phenomenon has a big impact on how lean thinking is implemented in employee involvement, internal technology development, and customer management.

In addition to manufacturing, lean production has been widely used in other fields, including construction, medical care, education, and so on. In recent years, it has combined other concepts. Other technologies were integrated by certain researchers to achieve lean production [37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57]. For example, lean six sigma integrates lean production and six standard deviation management, absorbing the advantages of the two management modes, making up for the shortcomings of the single management mode, and achieving better results. The authors of [37,38,39,40,41,42,43,44,45] applied lean six sigma by combining lean production and six sigma to address production difficulties. The potential advantages of the Cyber-Physical System (CPS) for the company as a whole were verified by Lappe et al. [46] using actual data from a gear manufacturer. The lean system was integrated with information platforms to enable customers, enterprises, factories, and suppliers to face dynamic orders on time, shorten the lead time, increase the total efficiency of the systems, and generate the best benefits.

Lean production is also used by several researchers in other disciplines [58,59,60,61,62]. The effects of “lean construction” and “continuous improvement plans” on employee safety were examined by Nahmens and Ikuma [58]. Bajjou et al. [59,60,61] used an interaction matrix to investigate how lean construction tools affect the sustainable growth of the environment, economy, and society. They compared the differences between lean thinking in the construction industry and the manufacturing business. The relationship between lean thinking and the idea of sustainable design was discovered by Carvajal-Arango et al. [62], who also examined the benefits of lean construction processes for the environment, society, and the economy. A structural equation model was developed by Ahmed and Wong [63] to assess the flexibility of lean construction on the job site. Shakoor et al. [64,65] used precise thinking to redesign the four spaces of the emergency room’s, men’s, and women’s consultation rooms, as well as the men’s and women’s observation rooms, in order to reduce emergency patient’s wait times while reconfiguring the number of medical beds that each department should have.

Accordingly, it can be shown that lean production not only lowers production costs and completely eliminates all waste in an organization to enhance production efficiency, but also generates the best economic value for end users with the least amount of work. However, current usage and applications are insufficiently extensive and complex. This study applied main path analysis to conduct a more thorough and in-depth discussion on the related literature on lean production, and then analyzed the evolution and development of the lean production field in order to have a more complete understanding of the development trajectory of lean production and its applicable fields to facilitate in-depth development in the future.

The literature reviews indicated that lean production is frequently used in the manufacturing industry, and they were solely concerned with its application in particular industries. Lean production is especially applicable to a variety of business operations. Lean production is starting to be applied in other industries, such as office and service processes. The reviews mentioned above offer a great deal of insight into fields related to lean production, but they do not indicate the general direction of technological advancement in the field. A review of the aforementioned studies indicates that researchers have not compiled a comprehensive overview of the literature on lean production, identified the development and potential of lean production, or compiled a list of the technologies that have been studied. Determining the overall trajectory of lean production’s development can fill in any gaps in knowledge about it and show where it will go in the future.

3. Proposed Methodology

To understand the historical evolution and development of lean production, it is necessary to collect and read the relevant literature on lean production as the basis for keyword search. Keywords are selected in the Scopus database for literature search. All the data, such as author, title, source, abstract, etc., are exported from the Scopus database as text files to facilitate subsequent analysis of the sample data.

After exporting the data obtained from the Scopus database, the CSVed software organizes the data and checks whether there are duplicate literature data and deletes the data with blank fields, such as authors and titles, to complete the data collection. The collected data are sequentially analyzed as follows:

• Step 1. Main path analysis: Main Path 437 software performs the search path count (SPC) weight calculation method and chooses the global main path and the key-route main path, which are the two main path types. Moreover, Pajek software is used for visualizing the main path of the analyzed data result. The main function of the Main Path 437 software is to analyze the imported documented data, connect the correlation between the data, establish a citation network, and find the main path and its branch context. Pajek software is a software used to perform large-scale network analysis. It can visualize large-scale complex networks and makes it easier for researchers to understand the connection characteristics of a network.
• Step 2. Basic statistical analysis of journals: The top 20 most influential journals in the field of lean production based on the g-index are presented.
• Step 3. Growth curve analysis: Loglet Lab 4 software could predict the growth curve of the academic field of lean production.
• Step 4. Cluster analysis: Group Finder software divides academic documents with similar topics or the same nature into several clusters.
• Step 5. Text exploration: Based on the title content of each cluster, the frequency of each group’s keywords, and rank, the keywords are calculated by their frequency as a reference for naming each cluster and are presented in a word cloud.

Through the above five analysis results, the development trajectory and future development trends of the current lean production academic research field can be found. Some important techniques related to the above information are described in detail as follows:

A. 

Main Path Analysis

The main path analysis can be used to process a large number of cited documents and study the knowledge context of any academic field. The main path starts from the source point (starting point) of the academic field literature to the meeting point (end point) of the academic field literature, and the calculation of each connecting line segment is the weight. Next is to find the main path according to the weight of the connecting line segment. The following introduces the three weight calculation methods currently used by many people, namely SPC (Search Path Count), SPLC (Search Path Link Count), and SPNP (Search Path Node Pair) [66,67,68,69,70,71,72,73,74,75,76,77].

• SPC (Search Path Count)
  The SPC method is to take any link in the network and calculate the number of all possible paths from the source point to the end point of the link line, and also calculate the number of all possible paths between the source point of the link line and the sink point. Finally, one can multiply the two numbers together to get the weight of all the connecting lines.

2.

SPLC (Search Path Link Count)

The SPLC method is to take any link in the network and calculate the number of all possible paths from the source point and nodes to the end point of the link line, and also calculate the number of all possible paths between the end point of the link line and the sink point. Finally, one can multiply the two numbers together to get the weight of all the connecting lines.

The SPNP method is to take any link in the network and calculate the number of all possible paths from the source point and nodes to the end point of the link line, and also calculate the number of all possible paths between the source point of the link line and the sink point, and the number of all possible paths between the source point of the link line and the sink point. Finally, one can multiply the two numbers together to get the weight of all the connecting lines.

In this study, the Main Path 437 software could export the main path by using literature data compiled from the Scopus system. When exporting the main path, the Main Path 437 software also exported a text file, named MainPath, which included the journal name, publication period, and the g-index and h-index of the journal, where the g-index means that academic literature or research results are ranked in descending order of citations, “the g article with the highest ranking has at least the highest g2 citations”, and the h-index means that “h articles have been cited more than or equal to h times in all articles of a certain author.” This study used the g-index as the main indicator and the h-index as a supplement to evaluate the influence of journals in the academic field, and list the top 20 influential journals in the field of lean production research.

B. 

Growth curve

Since the Loglet Lab 4 software can be used to analyze time series data, the accumulated publication data of the academic literature in the field of lean production obtained from the Scopus database was imported into the Loglet Lab 4 software. The vertical axis was the cumulative number of articles in the academic field of lean production, and the horizontal axis was the year. Then, the growth curve inferred the growth and maturity period of the academic field of lean production.

C. 

Cluster analysis

The cluster analysis method used in this study was to utilize the Group Finder software to divide academic documents of the same nature or similar topics into several groups, and to individually name each group according to the content of the article’s keywords. The cluster analysis method of the Group Finder software adopts Edge-betweenness clustering, and the clustering steps are as follows:

• Calculate all the borders in the network (betweenness): Choose any two points from all nodes. In the shortest path between the two points, the total number of the shortest paths through the line segment is the number of boundaries of the line segment.
• Remove the line segment with the maximum number of boundaries.
• After completing the preceding steps, if a cluster is separated from the network, the modularity of the cluster will be calculated; if a new cluster is not separated, steps 1 and 2 need to be repeated until all the line segments are removed. Modularity (modularity) is mainly used to compare the strength of the correlation between the internal and external nodes of the cluster. When a network has a high modularity coefficient, it means that the connections between the clusters are relatively close, while the connections between the clusters are relatively far.
• Finally, the group with the greatest modularization coefficient is selected, which is the best grouping result of the cluster analysis.

D. 

Text exploration

The content of each group title classified by cluster analysis was imported into the Wordle software to calculate the frequency of each word in a large amount of text and present it in a word cloud. Words such as prepositions and definite articles were not included in the calculation, and all group-related keywords were ranked by frequency, and finally, each group keyword was used as a reference for group naming.

4. Result

The Scopus database is the largest database of peer-reviewed abstracts and citations, including scientific journals, books, and conference proceedings. Scopus provides a comprehensive overview of world research results in the fields of science, technology, medicine, social sciences, art, and humanities, and has smart tools to track, analyze, and visualize research. In addition, the range of journals included in Scopus is wider than in other databases of the same type. The various journal indicators used by Scopus can also be widely used to assess the influence of literature, journals, or researchers, as well as the influence of departments, schools, institutions, or countries.

Therefore, this study uses “Lean Production OR Lean Manufacturing” as a keyword to search the Scopus database. The search date was 6 April 2021 and the search results had a total of 8892 data records. After sorting and removing duplicates, and deleting those with unknown author names and blank authors, titles, and years, a total of 8567 pieces of data records were obtained as sample data for this study.

Next, the Excel application plotted the number of papers and the cumulative number of papers, 8567, published in the academic field of lean production from 1991 to 2021, as shown in Figure 1. In the figure, the blue bar is the number of papers, and the orange bar shows the cumulative number of papers. From the figure, it can be seen that the academic field of lean production has been producing papers with a growth trend of 50–60 per year from 1991 to 2002. After 2003, nearly 100 papers were published per year. Among them, the number of papers published in 2019 was the greatest, which indicates that the academic field of lean production was in a period of vigorous development during that period, and it also shows the degree of attention paid to the academic field of lean production that year.

Next, the g-index plays an indicator for finding the top 20 journals that have a greater impact on the academic field of lean production. The relevant information is organized as shown in Figure 2. The following gives a brief introduction to the top 5 journals in order.

Number one is the International Journal of Production Research, which is a leading journal in the fields of manufacturing, industrial engineering, operations research, and management science. It mainly publishes papers related to manufacturing, operations management, and decision-making assistance. The research results in these papers are often used to solve complicated decision-making issues in production processes, such as design, measurement, management, control, etc.

The second to fifth journals are the International Journal of Operations and Production Management, the Journal of Manufacturing Technology Management, the International Journal of Production Economics, and the Journal of Cleaner Production. The International Journal of Operations and Production Management mainly publish supply chain management-related innovation research; the Journal of Manufacturing Technology Management mainly publishes research on topics related to manufacturing technology management, production, marketing, etc.; the International Journal of Production Economics mainly publishes all research related to manufacturing, processing, and general production; and the Journal of Cleaner Production mainly publishes research related to environmental sustainability.

4.1. Analysis of the Growth Curve in the Academic Field of Lean Production

On the basis of 15,029 scholars that have invested in the academic field of lean production and have published a total of 8567 papers from 1991 to 2021, this study used the Logistic Growth model in the Loglet Lab 4 software to map the growth curve analysis chart for the academic field of lean production, as shown in Figure 3.

4.2. Main Path Analysis in the Academic Field of Lean Production

This section uses the SPC weight calculation method in the Main Path 437 software to achieve the main path analysis in the academic field of lean production. First, the global main path is applied to obtain the main path with the largest total weight of the lean production academic field, and then the key-route main path is used to obtain the development structure of the lean production academic field. In other words, two paths could explore the development path of the academic field of lean production.

4.2.1. Global Main Path

Combining the global main path derived from Main Path 437 with the Pajek software can help visualize the global main path and obtain the image of the global main path in the academic field of lean production, as shown in Figure 4, which is the largest total weight of the network. The global main path has 16 nodes in total. The green nodes in the figure represent the source points of the papers; the blue nodes represent the meeting points of the papers. Each node represents a paper, and the nodes are connected by directional arrows, and the direction of the arrows indicates the direction of the flow of knowledge. Moreover, there is a string of text codes next to each node, which means the first English letter of the surname of the main author of the paper plus the surnames of other authors, and the last number is the year of publication of the paper. If the code arranged in the database is duplicated, it will be sorted in lowercase English letters and added at the end of the code to make a distinction. The following will focus on the 16 articles on the global main path.

Emiliani [5] pointed out that if lean production is to be effectively implemented, it is necessary to make the employees’ personal behaviors and business processes change simultaneously. The author proposed a model that is based on world-class manufacturing methods, and emphasized the improvement of the personal behaviors of employees. This model can simplify the arduous task of personal development, and at the same time ensure consistency between business processes and employees’ personal behaviors.

Emiliani [6] succeeds Emiliani [5]. The model proposed includes behavioral science, philosophy, economics, and industrial engineering, which was to reduce personal behavior and waste in the production process through lean production, so that individuals can continue to create value and develop a healthier working environment to promote economic growth.

Emiliani [7] compared the processes of online auctions and traditional procurement between companies, and discussed the reactions of stakeholders and the problems that arise when online auctions are introduced into traditional procurement organizations. The author suggested that online auctions should adopt modern supply chain management methods and lean production techniques that meet the needs of buyers and sellers in order to truly eliminate waste and reduce costs.

Christopher and Towill [6] explored the combination of lean production and quick response systems to create a cost-effective supply chain, and proposed manufacturing and logistics models to implement the necessary infrastructure.

Hines et al. [9] pointed out that early people did not fully understand the concept of lean production, resulting in poor management effectiveness. Therefore, the authors explored the evolution of the concept of lean production and its application in organizations to provide more in-depth research in the future.

Holweg [10] focused on studying the influence of the Massachusetts Institute of Technology’s International Automobile Project to promote lean production, and through interviews with key personnel participating in the project, he recorded the important development process of the project. Finally, it is confirmed that this project does have a certain degree of influence in the field of lean production.

Browning and Heath [11] conducted research and discussion on the F-22 production system case at Lockheed Martin. The authors integrated existing relevant information and data, revised their original research framework, and redefined the impact of lean production on production costs. Research has found that reducing work tasks does not guarantee a reduction in production costs. Combining lean production with agile manufacturing can bring greater added value.

Yang and Lu [12] discussed how to solve the problem of positioning the heart rhythm regulator, and pointed out that this problem can be solved through “Multiple Attribute Decision-making Method” and “Value Stream Mapping (VSM)” by meeting the requirements of high service levels and low inventory costs. In addition, a case study of the heart rhythm regulator was used to prove that the inventory cost was reduced by more than 57% after the aforementioned method and the implementation of a lean production strategy.

Bhamu et al. [13] used a case study on the implementation of a value flow map (VSM) in the Indian automotive industry. The results of the study indicated that the implementation of a value flow chart can indeed improve a company’s productivity and quality.

Bhamu et al. [14] applied lean production to automate production lines in India to achieve production balance and meet customer needs. This research also provided managers with methods to increase productivity in automated production lines.

Bhamu et al. [15] reviewed the literature and research methods related to lean production in the past, and found that the field of lean production mostly focused on experiments to verify theories, and was applied by many types of industries. However, at the same time, it was also discovered that although lean production had become a system that required integration of multiple elements, it still lacked a standard implementation process framework.

Gupta et al. [16] studied the manufacturing process for radial tires and related waste, then, proposed a novel dynamic system model to evaluate the waste, and applied the model to the radial tire manufacturing plant and verified its effectiveness. In addition to evaluating the overall performance of the tire manufacturing plant, the model also clarified the level of environmental protection that the organization can achieve through the concept of lean production by bringing substantial benefits to the practice of lean production and a reduction in waste.

Dieste et al. [17] investigated whether organizations that adopted lean production practices have improved environmental measures, and explored the positive impact of lean production practices on environmental green indicators.

Abu et al. [18] studied the current deficiencies in the implementation of lean production in Malaysia’s wood furniture industry from four aspects: motivation, obstacles, challenges, and applications. The results of the study showed that companies that adopt lean production believe that implementation barriers are mostly related to employees, including factors such as lack of labor, lack of implementation knowledge, and employee resistance to change.

After analyzing the aviation industry environment, Amrani and Ducq [19] established a set of standardized lean system models. The models introduced technologies, such as visual management, standardization, and tact time to meet the market demand for product diversification and product complexity in the aviation industry, and have obtained considerable benefits.

Schulze et al. [20] pointed out that most manufacturers that focus on order production need to meet customer needs and the production costs are limited by customers. The traditional production method can solve the aforementioned problems through lean production, but if it is applied to order production, there will be obstacles to implementation. The obstacles are roughly divided into: organization, management, finance, customers, etc.

Through the global main path, the development trajectory of mainstream research in the academic field of lean production is shown. The preliminary literature from the source mainly proposes the concept of lean production and discusses related technologies. In the midterm, it began to explore the actual cases of lean production. Recently, it has turned to literature reviews of lean production and related case studies.

4.2.2. Key-Route Main Path

It is difficult to obtain a complete development structure only by relying on the main path analysis. Therefore, the key-route main path draws the extended critical main path diagram to ensure that influential literature will not be missed. As shown in Figure 5, by observing the interrelationship of multiple paths, the development path of the academic research field of lean production is presented at different periods.

There are 23 papers on the key-route main path, as shown in Figure 5. Compared with the global main path in Figure 4, all 16 documents can be found on the key-route main path, indicating that the number of documents appearing in the global main path is large. Most of them have a certain influence on the academic field of lean production. In addition, on the key-route main path, seven new papers did not appear on the global main path. The following is a brief description of these seven papers.

Booth [21] stated that many companies think that the so-called lean production is to continuously reduce costs, but ignore other aspects of the problem, such as the reduction in technical staff, insufficient design, development capabilities, etc. The author believes that the company is implementing lean production. At the same time, the overall internal agility and flexibility must also be taken into consideration to effectively adapt to the changing market demands.

Mason-Jones and Towill [22] believe that if market information can flow in the supply chain, then it can be used as an auxiliary tool to implement “Decision Support Systems (DSS)”, and at the same time overcome problems related to information sharing. It can bring great benefits to the organization as a whole.

Ben et al. [23] believe that “Agile Manufacturing” can meet fluctuating production requirements, and that lean production can create schedules. If these two points can be integrated, they can bring higher levels to the entire process. Efficiency and effectiveness.

Narasimhan et al. [24] believe that the existing literature does not have sufficient accuracy to explain the difference between “agility” and “leanness”, so they tried to distinguish the difference by testing the manufacturing principles and performance of the two. The research results showed that, in terms of the difference in manufacturing principles and performance, the degree of difference is similar to the originally expected difference, but no obvious distinction was obtained.

Shah [25] proposed a more comprehensive framework for the concept of lean production, and further explained the impact of the overall space and other related factors on lean production by drawing the operating space that matches the lean production.

Sawhney et al. [26] stated that various enterprise organizations are currently making great efforts in equipment maintenance to ensure the maintenance and improvement of production capacity, but no one has proposed ways to streamline fault maintenance. Therefore, the article uses a value flow chart (VSM) to develop a set of operation modes that can be used to streamline fault maintenance.

Mohanraj et al. [27] combined quality function deployment (QFD) and value flow diagram (VSM) to improve the parameters in the process and drew FSM (Future State Map) to implement the improved process.

4.3. Cluster Analysis in the Academic Field of Lean Production

After the main path analysis in the previous section, in order to understand the key research topics in the academic field of lean production, this study used the Main Path 437 software and Group Finder to use the edge-betweenness cluster analysis method to group the academic fields of lean production. A total of 20 groups were obtained from the analysis results. The top five groups were found, and the keywords of the literature titles of these five groups were obtained with the Wordle software, and finally the research topics of the five groups were named as “Lean Production Concept and Application”, “Lean Six Sigma Concept and Application”, “Lean System Integration and Application”, “Lean Construction Concept and Application”, and “Lean Medical Concept and Application”.

As shown in

Table 1. The relevant information of the five groups.

Research Topic	1th Group (1103 Papers) Lean Production Concept and Application	2th Group (127 Papers) Lean Six Sigma Concept and Application	3th Group (122 Papers) Lean System Integration and Application	4th Group (104 Papers) Lean Construction Concept and Application	5th Group (66 Papers) Lean Medical Concept and Application
1	Lean (0.3)	Lean (0.23)	Lean (0.38)	Construction (0.37)	Lean (0.25)
2	Manufacturing (0.2)	Six (0.22)	Manufacturing (0.21)	Lean (0.27)	Healthcare (0.15)
3	Production (0.1)	Sigma (0.22)	Systems (0.08)	Projects (0.05)	Health (0.15)
4	Implementation (0.08)	Manufacturing (0.07)	Integration (0.05)	Industry (0.05)	Management (0.08)
5	Performance (0.07)	Study (0.05)	Management (0.05)	Production (0.05)	Study (0.08)
6	Study (0.07)	Implementation (0.05)	Stream (0.05)	Management (0.05)	Case (0.07)
7	Practices (0.06)	Case (0.05)	Case (0.05)	Study (0.04)	Manufacturing (0.06)
8	Industry (0.05)	Framework (0.04)	Digital (0.05)	Review (0.04)	Emergency (0.06)
9	Case (0.05)	Management (0.03)	Value (0.05)	Implementation (0.04)	Hospitals (0.05)
10	Management (0.04)	Performance (0.03)	Model (0.04)	Safety (0.03)	Production (0.05)
Word cloud

, the relevant information of the five groups is aggregated, including research topics, number of papers, keywords, document growth trend graphs, and word cloud analysis graphs. The keywords are the frequency of occurrence of words in the titles of each cluster. As a sorting order, the numbers in parentheses are the average number of occurrences of keywords in each cluster. For example, the first keyword in the first cluster “Lean (0.3)” means that Lean appeared 0.3 times in the title of the cluster’s literature on average. Keyword sorting can help us quickly understand the research trend of each cluster. From the literature growth trend chart, five major clusters have a recent growth trend.

4.3.1. Lean Production Concept and Application

The first group consists of 1103 papers, the content of which is related to lean production. As shown in Figure 6, its main path has been extended from 1998 to 2020, with a total of 17 papers. The main path of this group is to explore the concept and application of lean production. Especially, references [11,12,13,14,15,20,21] also appeared in the global main path, each focusing on the concept of lean production and actual relevant cases. In addition, references [24,25] also appeared in the key-route main path, discussing related cases of lean production. Accordingly, it can be estimated that these eight papers have a certain degree of influence on the academic field of lean production.

Rahman et al. [28] studied the degree of practice of lean production in Thailand’s manufacturing industry and its impact on the overall operating performance of the company. This research focused on the three major aspects of “Just In Time (JIT)”, “Waste Minimization”, and “Flow Management” of small and medium enterprises and large enterprises. The research results pointed out that just-in-time production (JIT) has a more significant impact on operational performance in large companies, while waste minimization is more significant in small and medium-sized enterprises, and process management is not significant in small and medium-sized and large enterprises. It further proved that lean production has a certain degree of influence on improving operational performance.

Nordin et al. [29] pointed out that many papers pointed out that the main reason for the poor performance of lean production was “there is no way to manage properly in the process of production transformation.” Therefore, they proposed a set of implementation frameworks for lean production, and it serves as the basis for further research.

Bevilacqua et al. [30] discussed the level of lean production methods adopted by the Italian manufacturing industry and their relevance to the company’s operating characteristics and business growth. The results of the study pointed out that regardless of whether high or low level of lean production methods, attention should be paid to the practice of operating characteristics that contribute to business growth because lean production is applied to suppliers, labor, production efficiency, etc. Due to the complexity of operating characteristics and workload, this practice method is more suitable for large enterprises with sufficient resources.

Desanctis et al. [31] studied the influence of factors related to national culture and company characteristics that failed to implement lean production, and explored whether the environment in which the company operates affects the results of a lean production implementation. The research results indicate that national cultural aspects such as gender, egalitarianism, and performance orientation contribute to the success of lean production management and implementation. It also pointed out that maintaining the concept of lean production is more important than the development of lean production technology.

Alkhoraif et al. [32] mainly discussed the difficulties that small and medium-sized enterprises will face when implementing lean production, and conducted a more comprehensive and in-depth analysis of the three aspects of lean production planning, execution, and reporting, and providing relevant information. The research foundation is for future researchers.

Antony et al. [33] found out the current main research directions and novel research processes in the field of lean production, and formulated a conceptual framework for the practice of lean production and the future research agenda to provide relevant research foundations for future researchers.

Malik and Abdallah [34] explored the relationship between organizational internal awareness and the practice of lean thinking. The study pointed out that the interaction between internal members of the organization contributes to the improvement of organizational awareness, and this phenomenon is important for employee participation and internal technology. The realization of lean thinking in development and customer management has a significant impact.

Amaro et al. [35] refer to the emergence of companies that have only a superficial understanding of the definition of lean thinking and the meaning it contains. Therefore, they introduced terms related to lean thinking to further clarify the substantial meaning of lean thinking in order to improve the overall operation of the company by identifying possible value and eliminating waste.

Yuik et al. [36] mainly discussed the “critical success factors (CSFs, Critical Success Factors)” for small and medium-sized enterprises in the electromechanical manufacturing industry to implement lean production. The research found that they are mainly divided into four major factors: management leadership, training and promotion of professional knowledge, participation of employees, and the implementation framework of lean production for small and medium-sized enterprises. The research helped small and medium-sized enterprises in the electromechanical manufacturing industry to prioritize these critical success factors (CSFs) so that the management team can continuously improve their strategies and achieve higher production levels.

Figure 6 shows that this group mainly focused on the concept of lean production and related technologies before 2007. From 2010 to 2017, it changed to discuss the application of lean production in the manufacturing industry. After 2018, it focused on lean production implementation and analysis of thinking.

4.3.2. Lean Six Sigma Concept and Application

The second group has a total of 127 papers, and the content of these papers is related to lean six sigma as shown in Figure 7. Its main path has been extended from 2003 to 2019, with a total of 9 papers. The main path of this group is to explore the concept and application of the lean six sigma.

Smith [37] discussed the benefits of combining six sigma with lean production. According to research findings, the combination of these two factors can improve overall workplace morale, and at the same time, stimulate changes in workplace culture.

Antony et al. [38] pointed out that lean production reduces a lot of unnecessary waste in the production process of the factory, while six standard deviations use statistical methods to solve the problem. If the advantages of these two technologies can be combined, it can bring more benefits.

Kumar et al. [39] proposed the concept of lean six sigma, which combines the two characteristics of lean production to reduce waste and six standard deviations to reduce variation. The authors also proposed an implementation framework for the lean six standard deviations. Through practical application, it is found that the improvement of product quality has not only increased the overall customer loyalty, but also saved a lot of money for the organization.

Pepper and Spedding [40] proposed an implementation framework for the lean six standard deviations, which is a method that can continuously improve the process. However, at the end of the study, it was found that the lean six standard deviations did not have a standardized implementation framework. Instead, the strategy needs to focus on different areas for improvement at different periods in order to effectively optimize the overall system.

Drohomeretski et al. [41] applied three models of “Lean”, “Six Sigma”, and “Lean Six Sigma” to the manufacturing industry in southern Brazil and discussed the differences and complementarity of the three models in production decision-making. The experimental results found that the three models have different degrees of importance in the fields of vertical integration, production planning, and control decision-making.

Thomas et al. [42] pointed out that in the past, the lean six standard deviation method’s effectiveness on the lean six standard deviations was reduced by focusing only on the DMAIC period. As a result, a new implementation framework for lean six standard deviations was constructed to try to strike a balance between lean production and six standard deviations to improve production efficiency and reduce the problems of variation and critical to quality (CTQ). This framework was applied to aviation-related industries, and the results pointed out that the nonvalue added time (Non-Value Added Time) was reduced by 44.5%, and at the same time, more than 2 million GBP of funds were saved.

Ben et al. [43] integrated environmental factors into the lean six standard deviation framework, hoping to reduce the overall nonperforming rate and environmental impact. The framework was applied to the Indian auto parts manufacturing industry, and it was found that not only was the defect rate reduced from 16,000 ppm to 6000 ppm, but also the overall environmental impact was reduced.

Gijo et al. [44] mainly explained how to apply the lean six standard deviation method to an automotive auxiliary enterprise group in India to practice more excellent operations. The application of this method not only effectively reduced the defect rate from 38,000 ppm to 56,00 ppm, but even saved nearly 1.4 million INR in financial items.

E.VG et al. [45] introduced the application of lean six standard deviations in the system maintenance department of a manufacturing company. This method reduced the complaint resolution time from 12.5 h to 8.5 h. In addition, this research also reduced the turnaround time of core processes in the organization and saved nearly 2.5 million INR in financial expenditure.

Figure 7 shows that this group mainly focused on the concept of a refined six standard deviations before 2010. After 2010, it discussed the application of lean six standard deviations in different industries.

4.3.3. Lean System Integration and Application

The third group has 122 papers, and the content of these papers is related to the integration of lean systems. As shown in Figure 8, its main path has been extended from 2014 to 2021, with a total of 11 papers. The main path of this group aims to explore the integration and application of lean systems.

Lappe et al. [46] believe that the use of CPS can make the information within the organization transparent and improve the flow of information, thereby, improving the efficiency of the logistic process. Therefore, the authors used real data from a gear manufacturer to verify the potential benefits of the virtual–real integration system for the organization as a whole.

Kolberg et al. [47] pointed out that although the use of lean production can bring us great benefits, there are still many restrictions on customized production. The author believes that through CPS and information communication technology, the two fields of “Industry 4.0” and “Lean Automation” can be effectively integrated to meet future market demand and the vision of intelligent production.

Ma et al. [48] believe that the lean automation system integrated through CPS is considered a relatively low-cost tool that can be flexibly adjusted and applied. Therefore, an intelligent and precise automation engine based on CPS is proposed, and the current research results and the existence value of this system are verified through case studies.

Buer et al. [49] studied the current literature to find the correlation between industry 4.0 and lean production, and determined the main research fields and future research directions between the two fields for reference by subsequent researchers.

Rossini et al. [50] studied the impact of European manufacturers’ adoption of industry 4.0 technology and lean production implementation on the overall operational performance of their organizations. The results of the study found that the implementation of industry 4.0 technology and lean production are complementary. When enterprises widely implement lean production, the technical level of industry 4.0 is easier to improve and realize. If you can master this point, managers will be better able to effectively solve related problems when using these two technologies.

Taghavi and Beauregard [51] reviewed 35 documents related to industry 4.0 and lean production, and found the correlation between industry 4.0 and lean production through quantitative and qualitative analysis.

Agostinho et al. [52] believe that lean production can be said to be the basis of industry 4.0, which improves the effectiveness of lean production. However, since it is not clear what skills are required for lean production professionals to face the new environment related to industry 4.0, the author focuses on the impact of industry 4.0 on the skills of lean production professionals. After observation, it can be found that in addition to programming and data science-related technologies, analytical skills and interpersonal communication skills are also some of the important skills.

Santos and Martins [53] used literature analysis tools to discuss papers such as “the impact of the combination of continuous improvement plans and industry 4.0 technologies”. The analysis results show that the most influential journals are related to the fields of “computer science, industrial engineering, management and accounting”. For the continuous improvement projects related to industry 4.0, lean six sigma, lean production, and six sigma account for the largest numbers. The most frequently searched keywords are machine learning, the internet of things, and big data.

Fortuny-Santos et al. [54] used a literature review to assess whether lean production and industry 4.0 can demonstrate effective synergy. The results show that “Real-Time Information” is one of the key factors for the effective generation of collaborative operations. If lean production can provide a stable manufacturing process, industry 4.0 technology can be successfully used to achieve automation and digitization.

Gallo et al. [55] studied which industry 4.0 tools companies currently use, the reasons for their use, and what are the benefits of using the tools? The results point out that the most important industry 4.0 tools are the internet of things and big data technologies. Technology enables companies to increase productivity and flexibility. In addition, human engineering factors are also important factors that enable companies to achieve better performance.

Cagnetti et al. [56] explored how industry 4.0 and lean production technology can be implemented by enterprises in two main directions “strategic management” and “technology implementation”.

Figure 8 shows that this group mainly studied the combined application of virtual and real integration systems and lean production before 2017. Since 2018, it has turned to the combined application of industry 4.0 and lean production.

4.3.4. Lean Construction Concept and Application

There are 104 papers in the fourth group, and the content of these papers is related to lean construction. As shown in Figure 9, the main path was extended from 2005 to 2021. There are eight papers in total. The main path of this group aims to explore the concept and application of lean construction.

Salem et al. [57] found in the research that some lean-built tools have significant improvement effects, while others do not. This may be due to the lack of technical training of personnel, which makes some tools unable to perform their functions effectively.

Nahmens and Ikuma [58] pointed out that the concept of lean construction has recently been introduced into the residential industrialization industry and has achieved remarkable results, but the impact of lean construction on employee safety is little known. Therefore, the author analyzed the impact of “Lean Construction” and “Continuous Improvement Plan” on employee safety and found that the accident rate of builders who implemented the continuous improvement plan was significantly reduced, but regardless of whether the plan was implemented or not, the total work schedule was almost the same.

Bajjou et al. [59] mainly compare the difference between the application of lean thinking in the construction industry and the manufacturing industry. The author compares the three dimensions of “on-site production”, “project category”, and “complexity”, and finally uses tools such as value flow diagrams (VSM) and visual management systems to verify the contribution of lean thinking to the construction and manufacturing industries.

Bajjou et al. [60] used an interaction matrix to analyze how lean construction tools have an impact on the three aspects of “environment”, “economy”, and “society” in sustainable development.

Bajjou et al. [61] explored the current level of knowledge of professional architects in Morocco on lean construction, and assessed the potential benefits of lean construction, and at the same time identified key obstacles hindering its successful implementation. The results of the study pointed out that 61% of the interviewees were familiar with the practice of lean construction, which has improved the quality of construction, safety, and environmental standards. However, the key to hindering its success may be the lack of skilled manpower and due to financial resources.

Carvajal-Arango et al. [62] used a literature review to find out the relationship between lean thinking and the concept of sustainable architecture, and further analyzed the contribution that lean construction practices can bring to the economy, society, and the environment.

Ahmed and Wong [63] established a structural equation model to evaluate the adaptability of lean construction on the construction site. According to the research results, it is concluded that this model can improve the effectiveness of the implementation of lean construction on the construction site in the following ways:

(1)

Effective communication and integration between the senior management and the project team.

(2)

Set simple goals.

(3)

Establish the leadership style of the project manager that has adopted lean construction techniques.

(4)

Encourage customers or contractors to invest in lean construction technology.

(5)

Persuade the government to support the development of lean construction technology.

Xing et al. [78] used case investigations to understand whether the Chinese construction industry has fully understood how to use lean construction to maximize project value, shorten schedules, improve construction quality, and reduce waste. The research results pointed out that the interviewees in the case survey indicated that the implementation of lean construction can greatly reduce waiting time and defect rate, and can also effectively improve productivity and production quality. However, they also found that the trust of internal personnel in the organization was lacking and the capabilities of the stakeholders were two major challenges.

Figure 9 shows that this group focused more on the concept and theoretical analysis of lean construction before 2017, and since 2018, it has turned to discussing the case application and effectiveness analysis of lean construction. With the successful application of lean production in the manufacturing industry, the construction industry has also begun to adopt the concept of lean production to achieve “Lean Construction” to reduce waste in the process and increase efficiency.

4.3.5. Lean Healthcare Concept and Application

There are 66 papers in the fifth group, and the content of these papers is related to lean medical treatment. As shown in Figure 10, its main path was expanded from 2008 to 2019, with a total of 6 papers. The main path of this group aims to explore the concept and applications of lean healthcare.

Newbold [79] believes that applying the concept of lean production to the medical industry has a great opportunity to improve the current relatively insufficient medical resources.

De [80] mainly described the application of lean healthcare by reviewing relevant literature on lean healthcare, and evaluating the development trend of lean healthcare over the years. The results showed that although most people have agreed with the potential benefits of lean medicine, how to evaluate its effectiveness more rigorously is still a challenge.

Radnor et al. [81] used case analysis to study the impact of lean concepts in the medical industry. The results of the study found that the development backgrounds of the medical industry and the manufacturing industry are different, which makes the medical industry violate some assumptions in the lean concept. This problem should be corrected as soon as possible, otherwise, the lean concept may severely limit the productivity of the medical system.

Dannapfel et al. [82] proposed the strategy of introducing a lean improvement plan in the medical industry, which is mainly divided into the following steps: (1) establish a clear vision and goal; (2) train the adopters of this plan; and (3) carry out publicity.

Shakoor et al. [64,65] applied precise thinking to the four spaces of the men’s and women’s consultation rooms and the men’s and women’s observation rooms in the emergency room to reduce the waiting time of emergency patients. At the same time, the number of medical beds in each department should have been reconfigured. The results of the study show that the aforementioned four spaces occupy additional beds. It is recommended that these beds be allocated to other relatively crowded departments to achieve an effective use of resources.

Figure 10 shows that this group mainly focused on the concept of lean healthcare before 2014. Since 2017, it has turned to the practical case application of lean healthcare.

5. Conclusions

The cumulative number of papers published in the academic field of lean production so far is 8567. This study used the global main path to find the main research path with the largest total weight, and combined key-route main path to observe the interaction and correlation between the paths. From the results of the two main paths, it can be seen that the research directions of the two are mostly the same. That is to say, the development path of the research field of lean production is from the concept of lean production in the early stages and the discussion of related technologies, the actual case of lean production in the mid-term, and the literature review and related case studies of lean production in the later period. In addition, seven papers that did not appear in the global main path were found in the key-route main path, which made up for the lack of information about the global main path.

This study also further used cluster analysis and text exploration to find the top five groups in the number of articles in the literature, and then analyzed other research fields related to lean production. The five clusters are “Lean Production Concepts and Applications”, “Lean Six Standard Deviations Concept and Application”, “Lean System Integration and Application”, “Lean Construction Concept and Application”, and “Lean Healthcare Concept and Application”. Finally, the global main path of the top five groups finds out the development context and research focus of each group.

By observing the two main paths and the overall context of the five clusters analysis, it can be found that the research field of lean production mainly focused on the discussion of the concept of lean production in the early stages, and gradually turned to studying the cases of lean production applied in different fields in the middle and late stages. In addition, the research topics summarized from the analysis of the five clusters showed that the application fields of lean production were quite diverse. Many industries try to improve unnecessary waste in the production process through lean production and achieve higher operational performance. The concept of lean production is highly valued in different industries.

The main contribution of this paper is to explore the development trend of the academic field of lean production. It is estimated that this field will enter a mature period in 2043. By then, the cumulative number of papers published is expected to reach 14,478, and the overall development will be more complete. Lean production is expected to have a development period of more than 20 years. This study demonstrated the current state of lean production development and potential research. It suggested that future research should continue to focus on this area and conduct a more in-depth and comprehensive discussion. Furthermore, this study used the Scopus database system as the source of the research data, and it is advised that future studies use both the Scopus and WOS databases to gather literature in order to conduct a more thorough analysis and ensure the completeness of the literature.