Abstract

History and basics of Kaizen

New management principles have been adapted to maintain competitiveness in global manufacturing. These new principles include Kaizen, which is the Japanese term (“Kai” meaning “change” and “Zen” meaning “good”) used to define continuous improvement (Palmer 2001). Ohno (1988) describes Kaizen as one of the pillars of the Lean Management System and a continuous improvement method as it follows the renowned Plan-Do-Study-Act (PDSA) methodology. According to Terziovski and Sohal (2000, p. 540), “Kaizen means ongoing improvement involving everyone, including both managers and workers,” with the underlying principle of serving customer needs. Palmer (2001) cites Kaizen implementation as a way to maintain low cost and less inventory, as well as a practice to reduce waste in processes and obtain continuous change in systems when compared with lean implementation. Kaizen also distinguishes itself from other continuous improvement practices by allowing for team members to implement changes and see the effects of their efforts (Farris et al. 2008), as well as encouraging active participation of company workers in industrial engineering and job design (Wood 1989). The implementation of Kaizen methods and activities is sometimes referred to as a “Kaizen event” (Doolen et al. 2007).

In the 1970s, Toyota was the first company to fully develop and implement Kaizen (Sheridan 1997). As a result, Toyota increased production and competiveness by using small teams of members with different functional skill sets who worked together to meet project goals (Doolen et al. 2007). Bessant et al. (2001), as well as Doolen et al. (2007), identify the importance of such cross-functional problem-solving teams in improving employee cooperation and goal setting. The need for ongoing and active engagement of all members within an organization highlights Kaizen's commitment to continuous improvement principles.

Motivators for Kaizen and continuous improvement

Several studies identify factors that contribute to companies' desires to implement Kaizen methods and other continuous improvement practices. For example, Schuler and Buehlmann (2003) studied how US wood furniture industries compare with other nations that have a globally competitive furniture industry. Czabke (2007) and Pirraglia et al. (2009) highlight the importance of training sessions and trade conferences in affecting companies' decisions to implement Kaizen activities. Kaye and Anderson (1999) identify leadership from company management as an essential motivating factor for the implementation of continuous improvement initiatives. In the analysis of survey data conducted by Smith et al. (2004), “cost reduction” was shown to be a main motivator for implementing lean continuous improvement practices; other important motivators were changes in customer demand, and the desires to remain competitive, reduce lead times, and increase flexibility.

Barriers to Kaizen and continuous improvement

Pirraglia et al. (2009) surveyed wood products companies and identified several barriers to implementing continuous improvement initiatives. These barriers include employee or middle management resistance, lack of implementation expertise, lack of time, lack of labor resources, lack of capital funds, no sense of urgency, and failure of past continuous improvement projects. Kaye and Anderson (1999) cite employee and managerial resistance as obstacles to implementation. Similarly, the case study of Madrigal-Sánchez and Quesada-Pineda (2012, p. 910) noted that the wood products company studied had no “standardized steps to follow up and start the innovation process” after ideas were received from customers or employees. Employees may view this lack of planning and the lack of a formal ability to include employees' ideas as barriers to implementing and sustaining continuous initiatives.

Brashaw and McCoy (2007) identify slow changes and adaptation of new technology, imported manufactured products, waste reduction, high fuel costs, lack of innovation, environmental issues, and raw material costs as additional barriers in implementing continuous improvement practices such as Kaizen and lean thinking in the wood products industry. In addition, surveys of secondary wood products manufacturers also highlighted barriers related to the variability of demand, performance measurement constraints, and long changeover times such as the case of the work published by Smith et al. (2004).

Drivers of Kaizen

Research has also identified several topics and strategies that have been shown to be influential in the success of implementing Kaizen and other continuous improvement initiatives. Table 1 summarizes these drivers and the literature that highlights their importance for continuous improvement.

Table 1 Drivers of Kaizen.

View Fullscreen

Kaizen drivers should have an impact on productivity improvement. Specific components of productivity improvement include cost savings, lead time, labor productivity, cut time, branding time, and on-time delivery as measures in the wood products industry (Liker 1997, Gunasekaran et al. 2004, Czabke 2007).

Although drivers and barriers for the implementation of Kaizen events have been documented, there is no tool that allows managers to measure such implementation and how effective Kaizen is on increasing productivity. Therefore, the goal of this work was to design and implement an empirical tool to evaluate the effectiveness of Kaizen activities as related to motivators, barriers, and knowledge of Kaizen as well as to measure the impact of Kaizen drivers in productivity improvement.

The Kaizen tool was implemented in a wood products firm and statistical procedures were used to evaluate the impact of Kaizen barriers, drivers, and its effectiveness on the basis of the perception of production and nonproduction employees. This study provides a tool to evaluate the effectiveness of Kaizen events as part of the continuous improvement initiatives for practitioners and researchers. The implementation of the tool through a case study can be used as a guideline for practitioners.

Methodology

The methodology used builds upon the initial case study research of Atkinson (1994) and Bessant et al. (2001) on continuous improvement by determining the specific drivers of Kaizen and their effects on continuous improvement and developing a tool to measure the effectiveness of Kaizen events. The first step was to review previous research on Kaizen events to determine barriers, motivators, and drivers that could predict Kaizen effectiveness. Findings from the literature review were used to design a tool in the form of a questionnaire to measure the effectiveness of the Kaizen events. The Kaizen tool could be used to measure production and nonproduction employees' perceptions and to understand how the upper management supports Kaizen events. The tool included two different sections: section or questionnaire one was used to measure perceptions of production and nonproduction employees and included demographic aspects, Likert statements (to measure perception on motivators, barriers, Kaizen effectiveness, and Kaizen's drivers), and open-ended questions. The second section or second questionnaire was designed to capture Kaizen events' strategic aspects from the Kaizen manager. This second questionnaire was designed to better understand how the upper management supports the implementation of the Kaizen events at the case study firm.

Once the two questionnaires were designed, the following step was the validation of the Kaizen tool by implementing it in a real setting. Two companies in the wood products industry where Kaizen had been implemented were contacted and invited to participate in the implementation phase. After explaining the purpose of the research, confidentiality issues, and potential impacts of the implementation, one company agreed to participate in the implementation of the tool. An interview with the Kaizen or continuous improvement manager was conducted over the telephone and a visit to the company was scheduled to better understand the production process and how the Kaizen strategy was being implemented.

The next step required collecting data from nonproduction (administrative) and production employees. The employees participating in the survey were chosen by convenience (they were asked to voluntarily participate) by the continuous improvement manager. Descriptive and inferential statistical procedures were used to analyze the responses. Comparisons based on the type of employee (production and nonproduction) on the perceptions of knowledge, barriers, motivators, and Kaizen effectiveness were conducted using analysis of variance procedures. In addition, multiple regression analysis was used to determine which of the Kaizen's drivers (quality, teamwork, Kaizen awareness, and training) could be used as predictors of productivity improvement (Brayfield and Rothe 1951, John and Reve 1982, Santos 1999). The resulting multiple regression model is outlined as follows:

$$\text{Productivity improvement = quality planning and control construct + teamwork construct + employee awareness and training construct + error}$$

Results from the questionnaire applied to the upper management were used to provide additional insights in helping to understand how barriers, motivators, and drivers were affecting the effectiveness of the Kaizen process.

Kaizen effectiveness tool design

The findings of the literature review were used to formulate the Kaizen tool. The tool includes two questionnaires and has the following structure:

• Questionnaire for upper management: Before implementing the Kaizen tool with production and nonproduction employees in the case study firm, an interview with the continuous improvement manager was conducted with the goal to obtain general information on the strategic aspects related to Kaizen. The specific topics that were asked of the Kaizen manager were

• • Frequency of Kaizen events

  • Design of Kaizen activities

  • Performance metrics

  • Support from upper management

  • Participation and attendance to Kaizen events

  • Training related to Kaizen

• Questionnaire for production and nonproduction employees: This questionnaire had the following sections:

• • Section I: Demographic questions about position, knowledge of time that the company has been involved in continuous improvement, and the type of continuous improvement initiatives being implemented in the company

  • Section II: Likert statements to measure perception on the constructs of Kaizen knowledge (4), effectiveness (9), motivators (9), and barriers (10). The number in parentheses indicates how many items per construct.

  • Section III: Likert statements to measure perception on Kaizen driver's constructs: employee awareness (6), employee training (3), teamwork (10), quality planning and control (10), and productivity improvement (5). The number in parentheses indicates how many items per construct.

  • Section IV: Closed questions on frequency of training activities, frequency of employee participation in Kaizen events, frequency of communication of Kaizen goals by upper management, frequency of communication of quality-control goals by upper management, and frequency of quality-control data collection.

The questionnaire for production and nonproduction employees is included in the Appendix. Likert statements were chosen to measure the employees' perception on the knowledge, motivators, drivers, barriers, and effectiveness of Kaizen. A scale from 1 to 5 was designed for the users to indicate their level of perception using the following format: 1 = strongly disagree, 2 = disagree, 3 = not applicable, 4 = agree, and 5 = strongly agree. The direction of the Likert statement was positive in all cases.

Implementation of the Kaizen effectiveness tool

The manufacturing facility used for the case study is one of several manufacturing facilities across the United States owned by a parent corporation. With a daily production capacity of about 800,000 ft², the case study plant produces kitchen cabinets using soft maple, hard maple, and cherry lumber. Rough lumber is purchased from suppliers within a 250-mile radius and kiln-dried in the facility. The operations include ripping, cross-cutting, moulding, machining, and assembly. There are approximately 200 employees working at the company over two shifts, with an almost equal number of male and female employees.

Interview with upper management

Table 2 shows the key findings related to the interview with the upper management. The interview was conducted with the Kaizen manager, who dedicates 100 percent of his time to the implementation of Kaizen events. The Kaizen manager highlighted several Kaizen and other continuous improvement practices used by the case study firm. The use of Kaizen events was identified by the continuous improvement staff interviews as a positive way to create targeted increments in productivity and change. The interview also helped to identify specific aspects of the implementation and continuity of Kaizen events.

Table 2 Key results of the interview with the Kaizen manager.

View Fullscreen

The case study company combines Kaizen with value stream mapping (VSM) to fit Kaizen within a larger vision for the company and to build upon and sustain past improvements. VSM is a graphic technique to represent the flow of materials and information for a production process. VSM also includes critical performance metrics such as lead time, inventory levels, and value-added time. This graphic technique helps managers and employees to understand and visualize where to perform process improvements. These practices follow the suggestions of Liker and Meier (2006). The company actively worked to include production-level employees in getting their support for and participation in Kaizen events. The firm did not cite language barriers between production and nonproduction staff as existing to a large degree, which likely increased their initial ability to communicate and get “buy-in” from production employees.

The Kaizen manager in the case study firm indicated that VSM was used as a strategic tool and starting point to conduct Kaizen events. VSM was used to develop a strategic vision; the gap between the current and future state of the VSM generated tension that led to an effort in closing that gap. The Kaizen manager indicated that VSM is valid for 6 to 12 months and after that period, the VSM should be revised. After the gap between the current and the future state is closed, the Kaizen leadership team should get together to generate a new current and future VSM. The support from upper management to Kaizen events is fundamental. This includes the hiring of a 100 percent dedicated Kaizen manager who oversees the strategic plan and implementation of Kaizen events to close that gap between the current and future VSM. As the manager of the case study firm indicated, the most important outcome of the lean journey (in the form of Kaizen events) is the change in culture of the company's employees. New values that are developed include discipline, creativity, responsibility, and alertness. More important, the impact on cost and productivity will come as the culture and values are being changed.

In regard to the Kaizen implementation itself, there are several important take-away points resulting from the interview with the Kaizen manager. The manager indicated that Kaizen events are conducted at least once a week and that they last a few hours in most of the cases. The recruiting process for the events includes sending invitations to all employees (production and nonproduction) and providing sign-up sheets. The composition of each Kaizen event should include the following participants: a subject matter expert, a member of the leadership team, the operator or production staff involved with the actual problem or opportunity, at least three members of nonrelated areas, and one upper manager. The structure of the Kaizen teams ensures that everyone is involved and the problem can be addressed in the proposed timeline. As far as the time for the events, it was recommended by the Kaizen manager that the event be held during working hours and not finish until the problem is finally addressed. Participants in the Kaizen events should receive appropriate training and be recognized for their efforts with lunch, T-shirts, or a certificate to recognize their participation.

As for activities to be conducted during the Kaizen events, the Kaizen manager responded that the process or areas of concern should include a walk-through by all of the events' participants. The goal of this walk-through is to identify the improvement opportunities with focus on the quality, safety, self-proof devices (poka-yoke), and waste identification. In addition, participants should be able to take basic productivity measurements such as travel distances, times, and production rates. Once the team has conducted the walk-through, the team will identify the appropriate lean tools to solve the issues. The Kaizen manager recommended focusing on the “low-hanging fruit” and start with the easy-to-implement lean tools in this order: 5s (sort, set in order, shine, standardize, and sustain), total productive maintenance (TPM), standard work, visual controls, and quality in the source. As the need arises, the manager suggested incorporating other higher-level lean tools such as Six Sigma, just in time, one-piece flow, and cells. The manager highlighted that 5s and TPM are perhaps the most critical tools, as a stopped and unorganized production line are perhaps the biggest sources of waste in the system.

The Kaizen manager also provided interesting insights regarding the sustainability of Kaizen events. The manager should conduct daily, weekly, monthly, and quarterly audits to make sure Kaizen events are sustained over time. Daily audits should be conducted by the process owners or operators using a list of items that should be inspected or checked. These daily audits should focus on quality, safety, 5s, maintenance aspects, and productive issues. The results of the audits should be reported immediately to the Kaizen leadership team and the issues found should be addressed immediately. Weekly and monthly audits should be conducted by the line production supervisors and by the Kaizen manager. These weekly and monthly audits should be focused on medium to long-term aspects of the Kaizen process. Quarterly audits should be conducted by the upper management with the goal to check on the strategic long-term goals of the Kaizen events.

Effectiveness of Kaizen events

This section of the tool was designed to measure the employees' perceptions of the effectiveness of Kaizen events. Production and nonproduction employees were asked about the impact of Kaizen events in helping to achieve different internal metrics and goals such as profits, cost reduction, lead time, product quality, and customer satisfaction among others. A Likert scale with the following values was used: 1 = strongly disagree, 2 = disagree, 3 = not sure, 4 = agree, and 5 = strongly agree, and the values shown in Table 3 correspond to the median as a measure of central tendency. The results in Table 3 indicate that production and nonproduction employees have similar perceptions regarding the effectiveness of Kaizen events. A nonparametric median test was used to compare each group on each question and none of the comparisons were found as significant (using an α value of 0.1).

Table 3 Median of statements related to effectiveness of Kaizen events.^a

View Fullscreen

Knowledge, motivators, and barriers of Kaizen events

Employees were also asked about their perception on the knowledge, motivators, and barriers of Kaizen events. Using the same type of Likert statements, Table 4 shows the results by type of employee (administrative [nonproduction] and production). In the case of Kaizen knowledge, it was found that both administrative and production workers have similar perceptions. A nonparametric median test (with α = 0.05) was conducted and no significant differences were detected between the two groups.

Table 4 Median values of knowledge, motivators, and barriers of Kaizen events.^a

View Fullscreen

Similar results were found for the Kaizen motivators where no differences were found when comparing both groups (α = 0.1). When employees (production and nonproduction) were asked about cost efficiencies as a motivator, both groups indicated that they were not sure, and similar results were found for the statement related to their knowledge using kaizen in other companies.

When examining the perceptions on barriers for Kaizen events, it appeared that some differences between the groups might exist. However, when a nonparametric media test was conducted, none of the differences were found to be significant (α = 0.1), except for the statement “There is resistance to generating new measurements of improvement for Kaizen activities” that yielded a P value of 0.02. In this case, the statistical evidence suggests that production employees disagree (median = 2) with this statement, whereas nonproduction employees are not sure (median = 3). A statistically significant difference was also found for the statement “There is a lack of technological capability to be able to implement Kaizen activities,” with a P value of 0.08 (α level of 0.1). The median for nonproduction employees was 4 and for production employees it was 2, which indicates that production employees perceive that technological capability does not influence the implementation of Kaizen activities. For the rest of the statements in the barriers constructs, the results were not found significant among the groups and the whole sample tended to disagree (median = 2) or strongly disagree (median = 1) with the different statements.

Effect of knowledge, motivators, and barriers on Kaizen effectiveness

From the previous section, the perception of nonproduction and production employees was found to be the same (α = 0.1) except for one statement. Therefore, the groups can be treated as one for posterior analysis. Four different constructs were designed to measure the perceptions of production and administrative employees on the knowledge, motivators, barriers, and effectiveness of Kaizen. Instead of doing a comparison based on each individual Likert statement, a subscale was created by adding all Likert statements in each construct. This approach allowed for a multiple regression analysis to be conducted, where the new variables knowledge, barriers, and motivators are continuously independent, and the new variable effectiveness is the continuous dependent variable. The model is:

$$\text{Effectiveness = knowledge + motivators + barriers + type of employee + error}$$

Before creating the new subscales of variables, it was necessary to test the internal reliability of each construct by conducting the Chronbach's α procedure. For this test, α values above 0.7 indicated that the internal reliability of each construct was acceptable; therefore, the subscales or new variables could be created (Table 5).

Table 5 Structure, reliability analysis, and normality test of Kaizen constructs.

View Fullscreen

In addition, a normality test was conducted on each of the newly created variables to make sure the individual new variables followed a normal distribution. In the case presented, the null hypothesis (the data follow a normal distribution) was not rejected for any of the new variables at a significance level of 0.1 (Table 5).

The means of each type of employee on the dependent variable Kaizen effectiveness is shown in Table 6. Therefore, a statistical test is necessary to test the significance of these differences when controlling for Kaizen knowledge, Kaizen barriers, and Kaizen motivators (covariates). A preliminary analysis of covariance (ANCOVA) test was run to make sure there was no interaction between the continuous independent variables (covariates) and the factor (type of employee). This test indicated that for a significance level of 0.1, there was no significant interaction between the factor and the covariates. In addition, a Levene's test was conducted to test for the equality of variances and the results indicated that there was not a significant difference (P value of 0.289 with an α of 0.1). Therefore, the variances are considered equal.

Table 6 Descriptive statistics for Kaizen effectiveness.

View Fullscreen

The results of the ANCOVA test in Table 7 indicate that there were no significant differences by type of employee when using a significance level of 0.1. Also, the test shows that there is no influence of the covariates Kaizen knowledge, Kaizen motivators, and Kaizen barriers on the variable Kaizen effectiveness when using the same significance level.

Table 7 Univariate test for Kaizen effectiveness.^a

View Fullscreen

Drivers of Kaizen activities

The same procedure to prepare and validate data that was conducted in the previous section was also performed for the analysis of Kaizen's drivers. The new variables employee awareness (EA), employee training (ET), teamwork (TW), quality planning and control (QC), and productivity improvement (PI) were constructed by adding their individual Likert items (Table 8). A reliability analysis showed that for all constructs, the internal reliability of the Likert items was acceptable (>0.7). A normal test was conducted on each of the new variables. Results show that all variables follow a normal distribution except for TW. The data for TW were revised and an outlier was identified. Once the outlier was removed, the normal test was rerun, now showing that TW fits a normal distribution. The P values of the normality test in Table 8 are the recalculated values after the removal of the outlier.

Table 8 Reliability and normality test for Kaizen driver's constructs.

View Fullscreen

Table 9 shows the descriptive statistics of the Kaizen's drivers. The standard deviations are similar except for variable QC. The means are different because they are dependent on the number of items on each subscale. For example, variable ET only includes three Likert items. Table 8 shows the Pearson correlations of the Kaizen's drivers against the variable PI. These correlations are considered moderate and some of them are significant (at the 0.01 and 0.1 levels), as shown in Table 10.

Table 9 Descriptive statistics of Kaizen's drivers.^a

View Fullscreen

Table 10 Correlations of Kaizen's drivers with construct Kaizen effectiveness.^a

View Fullscreen

An important question that Kaizen practitioners need to know the answer to is if Kaizen's drivers influence the productivity of the company, variable PI. The correlations in Table 8 suggest that there might be some level of association for these variables. Therefore, a multiple regression test was conducted to know if PI can be predicted by any of the Kaizen's drivers as follows:

$$\text{PI = EA + ET + TW + QC + error}$$

Given that some of the correlations are significant, a collinearity diagnostics test was conducted. Observing the values of eigenvalues and a condition index, it was determined that there is a critical problem with collinearity. To mitigate that potential problem, factor analysis was conducted to reduce the number of variables in the model. Using a principal component as the method of extraction, varimax as the rotation method, and a cutoff value of one, two factors were extracted. Factor one is composed of variables ET and EA (called A_T) and Factor 2 of variables TW and QC (called T_Q), and the two factors explained 68 percent of the variability. Therefore, the final model looks like:

$$\text{PI = A\_T + T\_Q + error}$$

To run regression analysis, factor scores are generated for each observation. The results of the multiple regression analysis in Table 11 indicate that both factors are significant for an α value of 0.1.

Table 11 Regression analysis coefficients.^a

View Fullscreen