Transferring Japanese kaizen activities to overseas plants in China

The Authors

Katsuki Aoki, Faculty of Economics, Kanto Gakuin University, Japan Cardiff Business School, Cardiff University, Cardiff, UK

Abstract

Purpose – The purpose of this paper is to examine management practices in transferring Japanese kaizen activities to overseas plants. The aim is to provide a greater understanding of organisational capabilities that facilitate an incremental organisation-wide innovation.

Design/methodology/approach – Case studies of nine medium- and large-sized Japanese auto-parts overseas plants in China were conducted. The data were analysed by comparing the successful cases of management practices concerning the transfer of Japanese kaizen activities to overseas plants with unsuccessful ones. Based on the assumption that learning is situated in the context, the concept of communities of practice promoted by Wenger is used as the analytical framework of this study.

Findings – The results of the comparative analysis show some common characteristics of management practices in successful cases: they actively used team-based rather than individual-based suggestion schemes; they had human resource practices that emphasised the importance of having workers who could do more than one job and long-term employment; and the managers of successful cases conducted shop floor visits to check the work processes every day.

Originality/value – This study suggests the necessity for the researchers and practitioners of kaizen to consider a healthy balance between three types of organisational capabilities that encourage workers' self-initiative, facilitate cross-functional communication, and discipline workers.

Article Type:

Research paper

Keyword(s):

Japan; China; Continuous improvement; Manufacturing systems; Automotive components industry.

Journal:

International Journal of Operations & Production Management

Volume:

Number:

Year:

2008

pp:

518-539

Copyright ©

Emerald Group Publishing Limited

ISSN:

0144-3577

Introduction

The concept of kaizen (or continuous improvement) has received much attention as a key to Japan's competitive success (Imai, 1986). Over the past few decades, a considerable number of studies, which have focused on Japanese manufacturing techniques (Schonberger, 1982), the Toyota production system (TPS) (Liker, 2004; Ohno, 1988), or lean production (Womack et al., 1990; Womack and Jones, 1996), have illustrated the importance of kaizen. Furthermore, studies of kaizen activities in countries outside Japan, such as Australia (Chapman et al., 1997), Sweden (Lindberg and Berger, 1997) and the UK (Oliver and Wilkinson, 1992), suggest that the concept of kaizen have become routinely accepted throughout the world.

In contrast to the worldwide diffusion of the concept of kaizen, many researchers have illustrated the difficulties for many companies outside Japan to have kaizen activities take root in organisations (Bateman and David, 2002; Bessant et al., 1994). In fact, as the comparison of key performance indicators between Japanese, UK and US auto-parts manufacturers by Oliver et al. (2002) shows, there is still a large gap in terms of the effects of kaizen activities between Japanese and western companies. This highlights the necessity to understand not only the details of the implementation of kaizen activities in countries outside Japan, but also the nature of Japanese kaizen activities itself in more depth.

Studies on transferring Japanese kaizen activities to overseas plants would be helpful for this understanding, because these allow us to approach management practices by examining the well-suited context of Japanese kaizen activities in countries outside Japan. The transfer of Japanese kaizen activities to overseas plants has been researched as part of the studies on transferring Japanese management practices to overseas plants in the USA (Abo, 1994; Kenney and Florida, 1993; Liker et al., 1999), the UK (Elger and Smith, 2005; Oliver and Wilkinson, 1992; Saka, 2004), and China (Hong et al., 2006a, b; Taylor, 1999). These studies suggest that the implementation of Japanese kaizen activities in overseas plants is situated in the cultural and social contexts. For instance, Hong et al. (2006b) illustrate that it is difficult to get active participation from frontline workers in kaizen activities in China, and suggest that great management efforts are needed to create well-suited contexts for Japanese kaizen activities, such as introducing an open plan plant and office layout, and importing daily communal rituals from Japan.

This study examines management practices in transferring Japanese kaizen activities to overseas plants through the case studies of nine Japanese overseas plants in China. Based on the assumption that learning is situated in the context (Lave and Wenger, 1991; Nicolini et al., 2003; Suchman, 1987), the concept of community of practice (COP) promoted by Wenger (1998) is used as the analytical framework of this study. The usefulness of this perspective in analyzing kaizen activities is highlighted by the studies that illustrate the divergence between kaizen methods and the actual practice on the shop floor (Baxter and Hirschhauser, 2004; Jørgensen et al., 2003).

Japanese kaizen activities

Japanese kaizen* activities and continuous improvement studies*

This study begins by positioning the subject of this study, Japanese kaizen activities, in the field of continuous improvement studies. Bateman and Rich (2003) classify continuous improvement into two approaches according to the length of time over which the improvement activity is focused; the longer one is continuous improvement; the shorter one is process improvement. In the case of process improvement, short-term (one or two week) programmes that consist of break-through kaizen events are implemented in a focused area (Bateman and David, 2002). However, although many companies have achieved considerable improvement through the implementation of process improvement programmes, the initial improvement is easily eroded back to the pre-improvement level (Bateman and David, 2002).

In terms of the time frame of activity, Japanese kaizen is longer than process improvement. First of all, Ohno (1988, p. 42), the Founder of TPS, states, “improvement is both eternal and infinite.” This suggests that the concept of kaizen in Toyota is not a programme with a limited time frame, but a process of activities that are implemented continuously. In Japan, there is a trend that companies in various industries try to implement kaizen activities in line with the Toyota way ( Nikkei Sangyo Shimbun, 2007). Therefore, it is possible to consider kaizen activities in Toyota as representative of Japanese kaizen activities.

“Continuous improvement” is one of the two pillars of Toyota's basic philosophy and means not only creating a lean system that contributes to cost reduction, but also learning from mistakes and pursuing innovation (Kajiwara, 2002). In fact, Toyota has made a great profit in recent years (e.g. the net income was almost $14 billion in 2007). However, boosting the profit, or “exploitation of old certainties” (March, 1991), is not the only aim of kaizen in Toyota. Learning from mistakes and pursuing innovation include the aspect of “exploration of new possibilities” (March, 1991). Therefore, creating a system of continuous organisational learning aiming towards innovation is also the aim of kaizen in Toyota.

Liker (2004) notes that the concept of kaizen in Toyota is a kind of corporate culture that supports continual organisational learning. This study shares a similar perspective and is also in line with studies that consider continuous improvement as organisational capabilities (Grant, 1991, Nelson and Winter, 1982) or dynamic capabilities (Teece et al., 1997) that facilitate incremental, organisation-wide innovation (Bessant and Caffyn, 1997; Bessant and Francis, 1999). The dynamic capabilities are defined as “the firm's ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments” (Teece et al., 1997, p. 516). In this perspective, the aim of kaizen is to achieve sustainable competitive advantage through the creation of innovative organisational capabilities that are difficult to be imitated by others (Lewis, 2000). It is difficult for the TPS to be imitated by others, because it consists of a lot of interdependent systems, such as just-in-time and total quality control (TQC), and involves all levels of the organisation (Teece et al., 1997).

Organisational capabilities in Japanese companies

Before explaining the relationship between kaizen and organisational capabilities, this section focuses on the three types of organisational capabilities that play certain roles in Japanese kaizen activities. These are organisational capabilities that encourage workers' self-initiative (this study refers to them as type A capabilities), organisational capabilities that facilitate cross-functional communication (type B capabilities), and organisational capabilities that discipline workers (type C capabilities).

Researchers who recognise the effectiveness of Japanese work practices state that Japanese companies have developed capabilities that make their workers or work teams learn and improve their work processes independently (Kenney and Florida, 1993; Koike, 1994). Quality control (QC) circles or small group activities are considered to facilitate such capabilities through giving every employee an opportunity for learning (Cole, 1994). In this perspective, to put it simply, organisational capabilities that encourage workers' self-initiative play an important role in Japanese kaizen activities (type A).

Nonaka and Takeuchi (1995) suggest that Japanese companies which emphasise the importance of tacit knowledge have developed organisational capabilities that facilitate knowledge creation. In this perspective, the key agent of knowledge creation is a group of people who have various experiences and knowledge (Lam, 2000; Nonaka and Takeuchi, 1995). On-the-job training (OJT) plays a critical role in creating such capabilities. Employees in Japanese companies experience various kinds of jobs through the OJT, which helps to reduce social distance between different categories of the workforce (Lam, 2000). In this perspective, it is organizational capabilities which facilitate communication among diverse people that allow Japanese companies to implement incremental organization-wide innovation (type B).

Ihara's (2003) study, which is based on personal experience as a temporary worker in a Toyota plant, suggests that Japanese kaizen activities are not always based on workers' self-initiative, and that the management practices of disciplining workers have a significant effect on the improvement of work processes. Studies on total quality management in countries outside Japan also illustrate the positive effects of disciplining workers on quality (Edwards et al., 1998; Sewell and Wilkinson, 1992). According to Ihara's (2003) observation, every worker in the plant is expected to contribute a suggestion concerning the work processes once a month. However, most of the workers seem to force themselves to make a suggestion just before the deadline (Ihara, 2003, pp. 65-6).

In addition, Ihara (2003) describes shop floor conditions in Toyota as the following: the prefabricated building is enclosed by glass walls on all sides, the inside of the building and toilets are always kept clean, and even the placement of boxes is done with extreme care. He states that, in this situation, every worker always feels that he/she is under someone's supervision, and thus it is mentally easier for the worker to work as he/she is instructed (Ihara, 2003, p. 165). This situation is very similar to Foucault's (1977) “Panopticon,” and makes it possible to discipline workers to conform to the company's rules and ways (Ihara, 2003; Sewell and Wilkinson, 1992). This suggests that organisational capabilities that discipline workers to conform to the company's ways also play a certain role in Japanese kaizen activities (type C).

The process of Japanese kaizen* activities*

This section examines the relationship between these three types of capabilities and kaizen activities based on the original idea of kaizen in Toyota. Muda or waste elimination is the very basic idea of kaizen, because kaizen activities are implemented through the identification and elimination of muda (Imai, 1986, 1997; Ohno, 1988). According to Ohno (1988), muda elimination means enhancing the ratio of “value-added work” (changing the shape or character of a product or changing the assembly). Improving the operating efficiency through the elimination of the seven types of waste, namely waste of overproduction, waste of time on hand, waste in transportation, waste of processing itself, waste of stock on hand, waste of movement and waste of making defective products, is the denominator of this ratio (Ohno, 1988). It is operators that assume “value-added work.” As Shimokawa and Fujimoto (2001) suggest, Ohno had a philosophy that kaizen ideas should be conceived from the operators' standpoint, and took a stance of placing importance on hearing the opinion of operators.

In terms of the identification of muda, there are two different processes, i.e. one is the identification by operators; the other is the identification by others, such as team leaders and supervisors. In the former process, how operators suggest problems or muda concerning their work processes is of major significance. QC circles and suggestion schemes are well known means to facilitate operators making a suggestion. According to Masao Nemoto, who was a key person for TQC in Toyota (Shimokawa and Fujimoto, 2001), it is how every participant says something in a positive manner, not how large improvements are achieved that has a great significance in QC circles (Nemoto, 1992, pp. 56-7). This suggests that the self-initiative of workers to participate in kaizen activities through voicing matters concerning their work processes plays an important role in kaizen activities.

Nemoto (1992) also insists that it is supervisors (Shokucho) or group leaders (Kumicho) who are expected to achieve major improvements. Therefore, it is of major significance how actually noticeable problems and muda are to them in the work process. For instance, work standards in Toyota are used as a means for making muda noticeable by being posted at each work station so that supervisors can determine whether the operator is following the standards (Liker, 2004). A 5S or good housekeeping (Imai, 1997) also creates the condition in which muda is easily noticeable. When machines and tools are clean, one can easily find problems and determine the causes. Ihara's (2003) description, which was mentioned earlier illustrates how thoroughly 5S is being performed in Toyota. This suggests that disciplining workers to conform to work standards or 5S practices plays a critical role in kaizen activities.

Once muda is identified, it must be reported to the appropriate people to analyse the root causes and to take countermeasures. As Ohno's (1988) “five-why analysis” suggests, preventing a recurrence of muda through the thorough analysis of the root cause is the basic idea of kaizen in Toyota. In this process, communication between different functions has a critical role. If the cause of muda has to do with machines and equipment, communication between the user of the machine, a maintenance person and an equipment (or production) engineer is essential to analyse the root causes. Once the results of such an analysis are reflected in the revision of a work standard, an improved standard is established as the basis for the next round of kaizen. In this process, communication between the people who make work standards and the people who are involved with the analysis is indispensable.

Figure 1 shows this kaizen process and its relationship with the three types of organisational capabilities. This figure illustrates that each of the three types of capabilities plays a critical role in Japanese kaizen activities by supporting each kaizen process. However, management tries to make a trade-off between type A and C capabilities. In the kaizen process on the right side of this figure (concerning type C capabilities), operators strictly follow the methods decided by management, while operators actively make a suggestion to management in the process shown on the left side (concerning type A capabilities). This study considers both kaizen processes to be critical factors in Japanese kaizen activities. It would be difficult to create an organisational climate in which every employee can learn from mistakes and pursue innovation exclusively through the right side kaizen process. It would also be difficult to align the goals of workers who participate in kaizen activities with the company goals only through the left side kaizen process. Japanese kaizen activities are not merely voluntary activities among workers, but organisation-wide activities in order to ensure a continuous implementation of the plan do check action cycle (Imai, 1997).

Analytical framework

Communities of practice and the transfer of Japanese kaizen* activities*

The previous section illustrated that Japanese kaizen activities are supported by the three types of organisational capabilities. However, as many researchers suggest, the reason why organisational capabilities can be a source of competitive advantage is that they are difficult to imitate by others (Barney, 1991; Grant, 1991; Hamel and Prahalad, 1994). The organisational capabilities that have been developed through their own history and consist of complex social relationships are difficult to transfer to other plants.

This study addresses the issue of the transfer of kaizen activities to overseas plants through the concept of COP promoted by Wenger (1998). In this perspective, people learn something through participating in a practice, and also develop their identities through the participation (Lave and Wenger, 1991; Orr, 1996; Wenger, 1998). Here, the concept of practice means, “doing in a historical and social context that gives structure and meaning to what we do” (Wenger, 1998, p. 47). According to Wenger (1998), the COP involves a group of people who mutually engage in a practice through negotiating meaning, jointly pursue their own enterprise, and share resources for negotiating meaning, such as routines, words, tools, ways of doing things and so on.

This perspective suggests that the transfer of Japanese kaizen activities cannot be achieved only by introducing formal procedures in the overseas plants. As Brown and Duguid's (1991) argument of divergence between canonical and noncanonical practices suggests, actual practice cannot be controlled completely through formal procedures. In the strict sense of this perspective, the knowledge concerning Japanese kaizen activities is not transferable to overseas plants, because the meaning of an activity is always situated in the local context (Brown and Duguid, 1991; Suchman, 1987).

Instead, people in overseas plants can produce their own meaning of Japanese kaizen activities by participating in these activities. Management practices are able to allow the people to create their meanings of Japanese kaizen activities by providing them with a context to learn Japanese kaizen activities. It is such management practices that lay the ground for implementing Japanese kaizen activities in overseas plants.

Three types of capabilities in a COP perspective

A COP is able to give people a context in which they learn something and also develop their identities (Lave and Wenger, 1991; Brown and Duguid, 1991; Wenger, 1998). If it was possible to set up COPs for people in overseas plants, they would be able to engage in kaizen activities, negotiate the meanings of kaizen activities, consider kaizen activities as their own enterprise, and share the meanings of concepts and tools used in kaizen activities. In a COP, the participants are able to develop their self-initiative as participants in Japanese kaizen activities by having a meaningful experience of kaizen.

However, a COP is also a place that develops the participants' own language and norms that are difficult to share with outsiders, and enterprises that do not always match a company's mission (Wenger, 1998), or what van Maanen and Barley (1985) call a subculture. The engagement in a COP makes it possible both to produce a meaning in a local context and to constrain the original meaning from circulating beyond the local context (Bechky, 2003; Star, 1995; Wenger, 1998). As stated earlier, the three types of organisational capabilities play a critical role in Japanese kaizen activities. Although, type A is matched with the management practices that provide a COP as a context for learning, a COP is able to place limitations not only on the freedom of communication across various communities (type B), but also on disciplining people to conform to the company's ways (type C).

In order to address the issues that go beyond the scope of engagement in a single COP, this study notes people's modes of involvement in the three social systems identified by Wenger (1998): COP, constellation of practices, and designed organisation. As stated earlier, we have a meaningful experience by engaging in a COP. This kind of involvement corresponds to engagement as a mode of belonging to a social system (Wenger, 1998), or active involvement in mutual processes of negotiating meaning.

However, our scope of engagement is always limited to a COP, because a practice is always situated in the local context. Therefore, we cannot engage in a whole company or plant that consists of various COPs, but can imagine the whole picture of a company as a constellation of interrelated practices (Gherardi and Nicolini, 2002; Wenger, 1998). As Wenger (1998, p. 246) states, “No one's purview is the constellation itself, because no one has that scope of engagement.” This kind of involvement corresponds to imagination as a mode of belonging (Wenger, 1998), or creating images of the world and seeing connections through time and space.

A company or plant also has an aspect of functional organisation that is designed to achieve certain goals and missions (Thompson, 1967; Galbraith, 1973), or designed organisation. In a designed organisation, people's efforts are aligned to achieve certain goals according to formal procedures. This type of involvement corresponds to alignment as a mode of belonging (Wenger, 1998), or coordinating people's energy and activities in order to fit within a broader structure and to contribute to broader enterprises.

This study examines management practices that support the three types of organisational capabilities (A-C) through the perspective of COP by relating them to these three modes of belonging (Wenger, 1998): engagement – type A capabilities, imagination – type B capabilities, and alignment – type C capabilities. At the level of COP, workers in overseas plants can develop their self-initiative to participate in Japanese kaizen activities through engaging in actual practice. The main issue in management practices at this level is how to give workers a practice in which they participate and have a real say in improving their work processes, and through which they have meaningful experiences, i.e. how to support type A capabilities. At the level of constellation of practices, people in overseas plants are able to have a broader perspective through imagining connections between different functions or communities. The main issue in management practices at this level is how to give people an opportunity to access other kinds of communities, jobs, knowledge and perspectives, i.e. how to support type B capabilities. At the level of designed organisation, the activity of each worker is aligned in terms of a certain company policy or target. The main issue in management practices at this level is how to ensure workers' coordinated activities through disciplining them to conform to the company's rules and ways, i.e. how to support type C capabilities.

In sum, a COP has both positive and negative effects on the transfer of Japanese kaizen activities to overseas plant. To put it briefly, management practices that support type A capabilities encourage the positive effect of COP, while those that support type B and C capabilities serve to overcome the negative effect of COP.

Methodology

A case study methodology (Eisenhardt, 1989; Yin, 2003) was used to examine management practices concerning the transfer of Japanese kaizen activities in nine medium- and large-sized Japanese auto-parts overseas plants in China (Table I). This aimed to look at generic processes concerning how management practices that support the three types of organisational capabilities work in transferring Japanese kaizen activities to overseas plants in China. Multiple-case sampling was employed to strengthen the confidence of findings (Miles and Huberman, 1994). The main intent was not to conduct random sampling, but to follow replication logic (Yin, 2003). A cross-case comparative analysis was conducted in order to look for within-group similarities as well as inter-group differences (Eisenhardt, 1989).

Sample plants were selected with the aim of investigating the intensified management efforts to transfer Japanese kaizen activities to overseas plants. Automotive-related and China were two key factors in the case selection. First, the parent companies of all these plants were Japanese auto-parts manufacturers that have made an intensified effort to implement kaizen activities with the help of the customers or Japanese automotive manufacturers (MacDuffie and Helper, 1999; Womack et al., 1990). Second, as the growth of automotive production in China, that is estimated to produce more than 10 million cars by 2010 ( Nihon Keizai Shimbun, 2007), suggests, automotive-related companies have recently put a lot of effort into China. All sample plants had Japanese-related automotive manufacturers as customers, and were mainly managed by Japanese managers who came from Japanese parent companies.

In order to collect data, the author conducted field research in China from August to September in 2004 and from October to November in 2005. In the field research, semi-structured and unstructured interviews with managers were conducted in the sample plants in China (the details of interviewees are shown in Table I). In each plant, at least two interviews were conducted; one was in a conference room in a semi-structured manner and usually lasted about one to three hours; and the other was on the shop floor in an unstructured manner and usually lasted about one to two hours. All the interviews conducted in a conference room were taped and transcribed, while the interviews on the shop floor were written down in field notes. Follow-up interviews were also conducted in China for case A-D, F and J, and also in Japan for case C, D and F. Most of these interviews were also taped and transcribed. A total of 36 interviews were carried out. Data were also collected through on-site plant observations, and internal company documents, such as standard work sheets and work flow charts.

In every case, the interviewees included a Japanese managing director (MD) or their deputy who was in a position to lead the transfer of Japanese kaizen activities to the overseas plant. This was based on the following intents:

it is possible to hear about the situation of a whole plant because an MD and deputy MD have a responsibility to manage everything in the plant;
it is possible to hear about the experiences of kaizen activities in overseas plants from the standpoint of Japanese managers who might have a detailed knowledge of Japanese methods; and
it is possible to compare the experiences of kaizen activities in overseas plants with Japanese experiences.

The semi-structured interviews focused on the following three topics:

general business conditions;
the implementation of kaizen activities; and
management practices related to the implementation of kaizen activities (the details are shown in the Appendix).

The aim was to investigate not only the implementation of Japanese kaizen activities in China (topic 2), but also the environment surrounding this implementation (topic 1), and management efforts in creating well-suited contexts for Japanese kaizen activities (topic 3). In the framework of COP, it is actual practice that gives workers meaningful experiences, not formal methods. Therefore, the interviewees were asked about details on how kaizen activities and related management practices were implemented, in particular about the opportunities for participation by workers.

In the analysis, the data concerning management practices that support type A, B and C capabilities were selected and classified according to their type. Concerning type A, management practices that offer workers opportunities for participating in kaizen activities through voicing matters concerning their work processes were selected. Concerning type B, management practices that enable plant members to get in touch with people from different functions and communities, and also HRM practices that support cross-functional communication were selected. Concerning type C, management practices that aim to make workers adhere to company rules and work standards were selected. However, difficulties arose because management practices, such as suggestion schemes and QC circles, are implemented in different ways depending on the company, and cannot be classified into only one category. For example, a QC circle can be a place where workers actively participate in kaizen, and can thus be classified as type A. However, in one of the investigated companies, QC circles were conducted among people higher than team leaders. Therefore, when classifying the data, the author paid close attention to the way certain management practices were implemented. When describing the results of the analysis (“basic findings” in this paper), the author always tried to bear in mind the various ways certain management practices were implemented.

Data analysis included multiple readings of the data collected through various channels, such as interview transcripts, field notes, and documents given to the author by the investigated companies. In order to minimize interviewee's bias, data collected from the semi-structured interviews were carefully compared with the data from unstructured interviews on the shop floors and plant observations. In the unstructured interviews, interviewees were asked to explain tangible examples in detail in order to improve the reliability of the data. In the case of tools and equipment, interviewees were asked questions such as which exact part was improved through which kaizen activities, and who was involved in the improvement.

Categorization of sample cases

In order to conduct comparative analysis, the successful cases were identified according to the following two measures. First, the scale of kaizen activities implemented in each plant is used as a measure of success. This means that successful plants should be able to implement kaizen activities on a large-scale. Second, the degree of self-initiative of workers is employed as another measure of success. Here, the self-initiative of workers was based on the perception of Japanese managers. This study considers the perception of Japanese managers as a critical factor for measuring the self-initiative of workers, because they would know if the degree of the self-initiative of workers is enough to meet the requirement of Japanese automotive manufacturers.

Therefore, the interviewed Japanese managers were not only questioned about the attitudes and capabilities of workers, but also about their involvement in fulfilling customer requests (Japanese-related automotive manufacturers). For example, it was asked if the company received effective suggestions from the workers in addressing the customer requests. The investigated companies were being pressurised by the customers to produce auto-parts at the same level of quality as in Japan, but at cheaper cost than in Japan. In the successful cases, the tangible results of kaizen activities with the self-initiated involvement of workers were checked through the unstructured interviews and the shop floor observations.

Each of the nine sample cases was categorized into one of four groups based on the combination of these two measures (Figure 2). In the upper two groups of Figure 2 (group 1 and 2), the scope of kaizen activities covered the whole plant including the shop floor level. All of these cases had a place in which operators participate in kaizen activities as a team and suggest some improvement concerning their work processes. In the lower two groups (group 3 and 4), however, kaizen activities were implemented only on a small scale. The Japanese managers in these cases noted that they were not in a position to implement kaizen activities on a full scale, or that kaizen activities were only implemented among the people higher than supervisor level.

Concerning the self-initiative of workers, Japanese managers in group 1 suggested that they had already achieved proactive participation in kaizen activities by the workers. In terms of case E in group 3, although they had not implemented kaizen activities on a full scale, Japanese managers noted that the workers showed self-initiative to do their own jobs, and that they had a plan to implement kaizen on a full scale in the near future. In the two groups in the left column of Figure 2 (group 2 and 4), however, the Japanese managers found it difficult to encourage workers to show self-initiative. Finally, only case D and H in group 1 were considered to be successful. This suggests how difficult it is to transfer Japanese kaizen activities to overseas plants in China.

Basic findings

Management practices in supporting type A capabilities

Concerning type A capabilities, case D and H in group 1 and case J in group 2 had a place in which shop floor workers participated in kaizen activities, considered kaizen ideas, and suggested the ideas as a team. This management practice is considered to give workers a place in which workers have a meaningful experience of kaizen through negotiating meanings with each other, i.e. a COP. These three plants also actively used ideas from operators to improve their work processes and equipment. However, although operators in each case made suggestions, it was people higher than team leader level that played a major role in actually improving work processes, tools and equipment.

Case D held small group activities on a regular basis (QC circles and suggestion committees every Wednesday), in which workers considered and suggested kaizen ideas. This plant had a policy to revise work standards in response to the results of QC circles and suggestions from operators as quickly as possible. It was team leaders that played a major role in improving work standards in this plant. Although, the engineers prepared the documents that were used to make work standards, the team leaders judged the appropriateness of the procedures described in the documents, and asked engineers to revise the documents;, i.e. they had a real say in improving work standards.

Case H had a team-based piecework system. In an ordinary piecework system, a standardised way of working determines the upper limit of a piecework rate. In this plant, however, it is possible to achieve a higher piecework rate by improving the tools that the workers use by themselves. The MD of this plant noted that although the Japanese staff initially provided most of the ideas for improving tools, most of the recent ideas were suggested from the workers as a team. In addition, this plant has a policy to share a good idea from operators with the whole plant.

Case J also had QC circles and a suggestion scheme as in case D, and thus gave workers a place to have a meaningful experience. However, this plant had a more individual-oriented suggestion scheme than in case D. In case J, a suggestion was written down by an individual worker on a preset form, and was evaluated in terms of the estimated effects of its implementation. This plant had a monthly board of review in which the management evaluates the suggestions. The worker could get a monetary reward for the suggestion through this evaluation process. In contrast, the suggestion scheme in case D was conducted as a part of small group activities and did not provide a monetary reward to the individual worker who made a suggestion. In this scheme, a leader who was selected from among group members played a coordinating role in a group.

Both case D and J had a management practice that gave workers a challenge or hint to conceive kaizen ideas. As an example of case D, the author saw a noticeboard that had 15 sheets of hints about kaizen written by the MD. One kaizen hint included charts on how to identify the difference between repaired products and defective products. He mentioned his aim of this practice as giving workers clear goals in order to increase their self-initiative to create better things.

Most of the cases in group 3 and 4 did not have a place in which workers considered and suggested kaizen ideas as a team, such as QC circles. Therefore, it was only possible to observe very few management practices concerning type A capabilities from these cases. Case G had a QC circle whose participants were only the people higher than team leader level. In fact, the MD of this plant said that Japanese staff were the main people who conceived kaizen ideas. Case E also implemented kaizen only among people higher than assistant managers.

Although most of the cases in group 4 gave workers an opportunity to make a suggestion, managers in these plants noted that the plants were not in a position to link such a scheme to kaizen implementation. As an example of this position, a production manager of case B reported that, although defective products were rare in Japan, nearly half of the products were sent for repair in this plant. He explained that the operators were of the opinion that, “When defective products arise, one only has to send them in for repair.” Concerning the reason why it is difficult to implement kaizen with the involvement of operators, the deputy MD in case C put forward three reasons:

there is a large difference of salaries between management and workers;
it is easy to implement kaizen according to the ways decided by management; and
operators frequently move to other companies.

Management practices in supporting type B capabilities

Concerning type B capabilities, both case D and H in group 1 gave workers an opportunity to participate in the improvement of tools in collaboration with people who had different functions, such as tool engineers, production engineers and managers. This was a place in which participants were able to access other kinds of communities, jobs, knowledge and perspectives.

Case D encouraged active communication between production and tool functions by having team leaders engage in tool design. During the tool design stage, members from various communities, including tool engineers, team leaders, assistant managers, and managers discussed the tool for about 30 minutes. Case H had a room that was specially designed to improve tools. In the room, the operator as the proponent of improving a tool, a team leader and an engineer created a new tool in cooperation with each other under the lead of the team leader. These two practices are considered to be what Wenger (1998) calls “boundary practice,” in which members from various communities negotiate meanings with each other. Such a practice allows the participants to better understand other functions in the plant.

Case D and H also had OJT systems in which workers experience different jobs according to a personnel development plan. As an example of case D, although each worker was able to conduct only one work process at the beginning, half of all the workers were able to conduct all six work processes within the year; the other half were able to conduct about three work processes. In this plant, all Chinese managers had been promoted from within this plant. Case H also had a policy to promote long-term employment. These human resource strategies are able to give employees opportunities to understand the connection between different jobs in the plant.

In contrast, case J in group 2 had a system in order to narrowly limit the scope of each worker's job. This is based on the perception of Japanese managers that it is difficult for the workers in the plant to have a number of jobs. A number of plants in group 4 had a similar strategy. Case G had a policy that each worker basically has only one job, though they developed multi-skilled workers in Japanese plants. This policy aimed to prevent producing defective products by reducing the complexity of each worker's job. In this plant, it was production engineers in the Japanese head office that prepared tools for the Chinese plant.

In terms of the comparison between Chinese and Japanese plants, an operator in case C handled six machines in his/her daily operation in a Japanese plant. However, one operator handled only one machine in the Chinese plant. In case F, although one operator was in charge of two production lines in a Japanese plant, four to five operators were in charge of one production line in the Chinese plant.

Management practices in supporting type C capabilities

Japanese managers in most cases across all groups highlighted the importance of disciplining workers to conform to the company rules, especially work standards. In group 3 and 4, every plant was trying to create well-defined rules, and easy-to-follow work standards in particular. In most of the cases, inspections were strengthened to prevent defects in production. According to the MD in case F, “We are trying to create workplace conditions in which our operators steadily follow work standards. This allows us to prevent producing defective products.” The work standards of case G included a sophisticated means to discipline workers. There was a signature space on the back of a work standard sheet in case G. Every worker had to sign the sheet every day to show that he/she understood the standardised procedure. As another example of strategies to discipline workers, case F had a practice to post the names of workers who produced defects (and also those who discovered defects).

Case J in group 2 had a strategy that prescribed clear rules and standards to discipline workers. This plant had a policy to document all the things that must not be done, such as not to throw away rubbish on the shop floor, and also had four times more company rules than in Japan. This plant also had a practice that all managers go around the shop floor to check the work processes under the lead of the MD three times a month. After checking the work processes, they had a meeting to report the problems in each area and to take countermeasures.

Case D and H in group 1 also conducted the practice of shop floor visits by managers under the lead of the MD. However, there are differences in terms of the ways of disciplining workers through this practice between group 1 and 2. First, in both cases in group 1, a shop floor visit was conducted every day. Second, the MDs of both cases in group 1 noted that Chinese managers began to go round the shop floor on their own initiative after Japanese managers had conducted a shop floor visit for a certain period. The MD of case H said, “Our followers see our everyday practices as of top executives. Our everyday shop floor visit induced them to emulate the same practice.”

As stated earlier, 5S practice is used to make muda or problems easily noticeable by disciplining workers. Managers in case D, E, G, F and J made an intensified effort to implement 5S. Concerning this practice, interesting similarities were discovered between case D and E, both of which belong to the right column of Figure 2. First, the MDs of both cases noted that their workers practice 5S on their own initiative. Second, when asked how such a situation had been created, they emphasised the significance of the MD himself to pick up rubbish. Third, both cases had an assessment system and an opportunity to announce the results of 5S. In case D, Japanese managers, including the MD, had checked every workplace once every month to decide the best 5S achievements. According to the MD, Chinese managers asked him to leave the check of 5S achievements to them eight months after the assessment by Japanese managers started.

Discussion

This section begins by examining the similarities across the cases. All the investigated companies were under pressure from the Japanese-related customers to meet difficult requests, such as ensuring high quality at low cost. In addition, most interviewed managers across the cases shared the opinion that the initial standards of work performance and employee commitment in China were much worse than in Japan. Therefore, great management efforts were needed not only to implement Japanese kaizen activities, but also to meet the customer requests. In fact, all of these companies were vigorously trying to implement Japanese quality and cost management methods. These companies also roughly shared a “quality first” policy.

However, such management efforts did not necessarily lead to the implementation of Japanese kaizen activities that involved workers' active participation. Of all three types of capabilities, only management practices connected to type C capabilities were actively being applied by all the companies. In particular, most companies clarified their company rules. This suggests that in Japanese overseas plants in China, there is a higher priority on ensuring high quality by the use of management practices that discipline workers than on the implementation of Japanese kaizen activities through management practices that support type A and B capabilities.

As the previous section showed, various ways of implementing management practices that support the different types of organisational capabilities were identified among the nine cases. However, it was possible to sum up the common characteristics between the successful cases as shown in Table II. The following section examines the theoretical and practical implications of the findings by focusing on the characteristics of the successful cases.

General implications for the studies on kaizen

First, this analysis raises some doubt about the effectiveness of individual-based suggestion schemes as a part of kaizen activities. Although most of the cases implemented suggestion schemes, few positive results of individual-based suggestion schemes were encountered. In contrast, successful cases actively used team-based suggestion schemes, and gave a team a certain degree of autonomy. This result matches the perspective of COP that emphasises the role of other participants in a COP.

However, this analysis illustrates that it was the team leader level that played a major role in actually improving work processes and equipment even in the successful cases. The main roles of operators in case D were to follow standards, to identify problems, and to report them to their team leaders. Most of the interviewed Japanese managers had the opinion that in order to enhance quality it is necessary for everyone to follow the rules. This suggests the necessity to further note the roles of disciplining workers in Japanese kaizen activities. In fact, discipline on the job is considered to be a virtue in Japan. In general, Japanese consider disciplining employees, or shituke in Japanese, as a part of corporate education. Shituke, whose meaning is to teach employees good manners, is sometimes considered to be a part of corporate responsibility.

The successful cases used methods which emphasise the effects of COP and were also connected with disciplining workers. As Wenger (1998) illustrates, a COP plays a critical role in disciplining the participants to conform to the ways that the participants in a COP have developed through negotiating meaning. However, as stated earlier, a COP is likely to develop its own perspectives that do not always match company goals. Therefore, management practices that bridge type A and C capabilities are needed to align the perspectives of a COP to the goals for the whole organisation. In the case of small group activities in case D, they had a group leader meeting every week, in which group leaders who were selected from among group members, a Japanese director in charge, and Chinese managers participated. This meeting was a place in which management gave the participants advice, and through which activities from each group could be aligned.

It is important that this meeting should not be the place where the management requires the workers to unilaterally conform to the management policy. In that case, this meeting would cause detrimental effects on type B as well as type A capabilities by impeding workers from talking openly to the management. In order to avoid such a situation, the MD of case D always let Chinese managers accompany him on shop floor visits, and tried to teach them how to coach the workers so that they could gain a deeper understanding of the workers. Thus, it can be stated that QC circles and suggestion committees in case D, which are used in combination with group leader meetings, are able to support not only type A capabilities, but also type B and C capabilities, or each mode of belonging by Wenger (1998): engagement, imagination, and alignment.

Disciplining workers based on concrete situations

In terms of disciplining workers, this analysis discovered interesting similarities in successful cases. To summarise, successful cases had a tendency to employ more shop floor-based methods to discipline workers. Such methods were carried out through a shop floor visit by managers under the lead of the MD. This suggests that a lot of top-level management effort and their direct involvement are needed to achieve successful kaizen implementation. This assumption is supported by many studies on continuous improvement that emphasise the importance of active commitment by senior management (Bateman, 2005; Bateman and Rich, 2003; Kaye and Anderson, 1999).

In case D, the MD visited the shop floor for two to three hours every morning based on his philosophy of the “three gen” principle. In Japanese manufacturing plants, this principle normally refers to the three “gens”: genba (actual place), genbutu (actual things), and genjitu (reality). However, at this plant, he used the concept of genkohan, a word indicating that one is caught “red-handed” at the scene, instead of genjitu. According to him:

It is important to stop someone who is not doing things as established, at the scene. Such a situation constitutes an opportunity to convey our true purpose to the employees. If we explained things to them elsewhere, they might wonder why we were talking of an incident that took place in the past […] In order to establish a trust relationship with local people, we must explain things to them in a concrete situation.

This practice not only disciplines workers to conform to the company's rules, but also provides an opportunity to learn the ways of Japanese kaizen activities based on a response to a problem at the actual scene. This learning style has a significant meaning in the perspective of COP. As stated earlier, a meaning is always situated in the local context and is difficult to be applied beyond the local situation. Therefore, communication that is not based on a concrete situation is likely to lead to misunderstanding. However, this practice allows management and workers to use the concrete situation as a context for learning, and thus to share understanding.

The practice of shop floor visits by managers is not only used for the unilateral communication channel from management to workers. It is possible for workers to discuss some problems concerning their work processes with managers. In case D, managers were given authority to call suitable people to solve such problems, and to have a meeting at the scene. This means that this practice also provides an opportunity for cross-functional communication based on a concrete situation. Regarding this, the MD of case D noted that if a problem was not resolved at the meeting, the same problem or “red-handed” situation would always be caught as long as the everyday shop floor visit continued to be conducted. Therefore, this practice is able to give workers the legitimacy to suggest something concerning their work processes by letting them know that management addresses the problems suggested by workers.

This suggests that disciplining workers is not achieved by the application of unilateral power over the workers by management. A shop floor visit by managers is able to give workers an opportunity to have an image of the plant as “the plant that makes serious efforts for kaizen or not.” In the example of 5S in case D and E, if the managers pass through an area without stopping to pick up rubbish that is visible to workers, a plant image as “the plant that makes serious efforts for 5S” could be unsettled immediately. This means that actions by managers on the shop floor are always watched by workers, and that these actions are able to give workers the legitimacy to engage in kaizen activities. Managers as well as workers should follow 5S practices on the shop floor. Managers must show a lot of self-discipline if they want the workers to show the same self-discipline.

Kaizen and the evolution of equipment

Concerning type B capabilities, both successful cases provided a place in which workers participated in the improvement of tools in collaboration with people who had different functions. These cases also had a scheme that trained workers to do different work processes, though unsuccessful cases were likely to narrowly limit the scope of each worker's job. As the comparison of multi-skilled operators between Chinese and Japanese plants in case C and F illustrates, Japanese overseas plants in China are likely to use simpler equipment than that in Japan. The reasons given were that in the case of complex equipment, it was difficult to do maintenance and to fix problems with just the people in overseas plants. This suggests the ineffectiveness of the methods that try to introduce or transfer advanced equipment to overseas plants without support systems for their introduction.

The improvement of tools in case H was a continuing process of trial and error by the people from different functions. According to the MD of case H:

The integration of different tools allowed one worker to do two different processes. However, this also broke the line balancing, and brought the increase of intermediate stock […] In the course of fixing line balancing, someone suggested the improvement of other tools […] We have done things like that using ideas from various people.

This illustrates that case H was able to develop support systems to maintain the equipment in the course of the kaizen implementation. The workers are able to gain knowledge about tools; the engineers are also able to gain knowledge about the conditions of equipment on the shop floor through the experiences of trial and error and communication between different functions. According to the MD of case H, this Chinese plant had developed a better production line than the Japanese main plant. In the assembly shop, the author saw a unique production line that used a circular rail to hook the products. The circular rail turned over at a constant rate (63 seconds), which decided the takt time of each work process in the production line. Moreover, the external defect rate of this plant had decreased from 100 to 10 ppm in the past year.

Case D also had improved their tools through the active communication between team leaders, tool engineers, and Chinese and Japanese managers. Finally, this plant had produced equipment that eliminated the intermediate stock between six assembling processes, i.e. they had achieved a single piece flow production system. According to the MD of case D, this Chinese plant was better than the Japanese main plant in everything at that time. The external defect rate in case D had decreased from 300-400 to 15-20 ppm in the past year. When the author visited the company's Japanese main plant in March of 2007, the plant was trying to implement a similar single piece flow production system under the lead of the same production director who was the MD of case D in 2005. Thus, case D had become a model plant for the Japanese main plant.

The kaizen activities in these cases make it possible both to continuously improve their equipment, and to create the support systems to maintain the equipment through the communication between different functions. These two successful cases suggest that the transfers of Japanese kaizen activities to overseas plants in China were not achieved by imposing Japanese methods on Chinese people unilaterally. These plants were able to outperform the main Japanese plants by continuously implementing the kaizen cycle (shown in Figure 1) based on the communication among various people including both Japanese and Chinese people.

Conclusion

In conclusion, the implications for the studies on management practices in transferring Japanese kaizen activities, or organisational capabilities that facilitate an incremental organisation-wide innovation, to overseas plant are summed up in the following three points:

Management practices that try to support one of the three types of capabilities that encourage workers' self-initiative, facilitate cross-functional communication, and discipline workers are able to achieve significant effects in combination with the support of the other two types of capabilities. It is necessary to create a healthy balance between the three types of organisational capabilities.
Shop floor-based disciplinary methods play a major role in transferring Japanese kaizen activities to overseas plants. These methods are able to give people in overseas plants not only an opportunity to learn kaizen activities, but also a source of understanding that provides the legitimacy to engage in kaizen activities.
Communication between different functions makes it possible not only to improve their equipment, but also to create the support systems to maintain the equipment with just the people in the overseas plant.

As stated earlier, in Japanese kaizen activities, management had to make a trade-off between type A and C capabilities. In order to address this issue, this study concludes it is necessary to further examine shop floor-based disciplinary methods. The successful cases made good use of such methods. As the “red-handed” practice in case D illustrates, managers are better able to make the linkage between company goals and workers' operations comprehensible, when it is based not only on canonical descriptions (Brown and Duguid, 1991), such as work standards and company rules, but also on a concrete situation or actual practice. This suggests that such shop floor-based methods may lead to the sharing of company goals between workers and management, and may also produce a trust relationship between them.

In order to further examine this hypothesis, it will be necessary to develop a framework that is useful to investigate the relationship between management practices and trust relationships between different actors. Using Wenger's (1998) concept of COP, it is possible to address this issue by focusing on the three modes of belonging (engagement, imagination, alignment). As Wenger (1998) suggests, an essential duality is inherent in COPs that not only create meaningful experiences for the participants, but also hold them hostage to such experiences. COPs create discontinuities between those who have been participating and those who have not (Wenger, 1998, p. 103). However, few studies have addressed this issue in the context of the implementation of actual management practices, such as kaizen. Further research should be focused on issues such as how individual actors achieve a healthy balance among the three modes, and how the combination of the three modes leads to trust relationships within and between COPs.

Concerning the research methodology, some future tasks exist. This study has not fully examined social and historical aspects of management practices, such as conflicts and power relations among various plant members. To address such issues, it is necessary to conduct diachronic studies on single cases that cover interviews with various members, such as workers, supervisors, engineers and managers. Moreover, in order to approach the complexity of individual management practices in depth, more participative methods, such as participant observation, are needed.

Some future tasks also remain concerning the generalisability of the results. This study has not fully examined the influence of cultural factors in China on the adequateness of management practices. For example, it can be questioned if Japanese methods that emphasise discipline work better in countries, such as Europe and the USA with a longer history of implementing Japanese kaizen activities than China. Thus, there is a need for cross-national case comparisons in order to examine the cultural differences of the implementation of Japanese kaizen activities.

Figure 1Relationship between kaizen process and three types of capabilities

Figure 2Four groups of sample cases

Table IList of surveyed plants and interviewees in research visits in China

Table IICharacteristics of management practices in successful cases

References

(1994), in Abo, T. (Eds),Hybrid Factory: The Japanese Production System in the United States, Oxford University Press, New York, NY, .