June M. Worley and Toni L. Doolen, Ph.D., Oregon State University
Eileen M. Van Aken, Ph.D. and Jennifer Farris, Virginia Tech
Organizations continue to seek innovative ideas for improving their processes and retaining a competitive edge. Kaizen is a concept that focuses on improving a work area or an organization in incremental steps. Many organizations have begun to incorporate the philosophy of kaizen through the use of kaizen events. Kaizen events typically focus on specific improvement goals. A kaizen team may include operators, managers, or supervisors from both the work area and from other work areas within the organization. Other support personnel may also be included in the event such as engineers or personnel from outside of the organization. Events are typically held for three to five days, but the time period is dependent on the difficulty of the goals and the complexity of the work area. The kaizen team may apply different tools to meet the goals of the event, such as brainstorming, fishbone diagramming, or Pareto analysis. Kaizen teams may also follow particular methodologies, such as Single Minute Exchange of Die (SMED) or Standard Work.
Many organizations report initial promising results from kaizen events, but sustaining the results over extended time periods has been acknowledged as a problem. The variability in event success within an organization is another area of concern for organizational leaders. While some events may yield significant results, others may not yield many positive performance improvements.
Industrial engineering researchers at Oregon State University and Virginia Tech have undertaken a study of kaizen events, supported by the National Science Foundation. The research project currently has partnered with nine different organizations that have operations in several regions of the United States and Canada. The goals of the study are to identify those factors that make a kaizen event successful, as well as those factors that allow the results of the event to be sustained over long periods of time.
A framework was developed for the research that identifies the processes utilized in designing and managing kaizen events. This article provides a summary of the framework and illustrates how the framework can be used to evaluate the differences between kaizen events within an organization. By investigating the differences in event processes and results, organizations can identify strengths within a kaizen program and focus on areas that may require improvement.
The kaizen event framework is organized around four categories related to event processes: planning, implementing, sustaining, and organizational support. Planning processes include identifying an area or a process as the focus of a kaizen event, selecting the area or process, and defining the boundaries of the event. The second category in the framework is implementation. Implementation processes include preparing for the event, conducting the event, and deploying follow-up actions and improvements more broadly after the event. The sustaining category involves performance reviews, dissemination of results and lessons learned, and standardization of improvement activities. The fourth category, organizational support, includes processes supporting all kaizen events-rather than a single event- such as employee training and education, management of the kaizen event program, and employee motivation.
Applying the Framework
Information about kaizen events for this study is gathered primarily through four different instruments: a kickoff survey, a team activity log, a report out survey, and an event information sheet. The kickoff survey is given to the members of the kaizen team on the first day of the event, after the goals have been defined for the team. During the event, a member of the kaizen team records all activities of the team on the team activity log. On the last day of the event, the report out survey is administered to the team, typically after the team has presented its results to management. After the event, the facilitator completes an event information sheet that provides background information on the work area, its history with kaizen events, and the facilitator's perceptions regarding the event.
An analysis of 12 separate kaizen events gathered from the same organization over a six month period is summarized here to illustrate how the framework can be applied. The organization is a specialty equipment manufacturer that has six years of experience with kaizen events. All work areas within this organization have participated in kaizen events, including support areas, such as sales and information technology. Three of the events included in this study focused on sales or customer service, while the remainder of the events focused on manufacturing departments. While the specific goals of the events varied, most of the 12 events identified three or four goals.
Kickoff Survey Results
The kickoff survey measures three factors: goal clarity, goal difficulty, and commitment to goals. Goal clarity measures how well the team members understand the goals of the kaizen event. Goal difficulty evaluates the team members' perceptions regarding the difficulty of achieving the goals of the event. Commitment to goals measures the perception of the team regarding the value of the kaizen event to the work area.
Analyzing the data from the 12 events, there were no statistically significant differences in the results for goal clarity. The mean score was 4.6 (where 4 = "tend to agree") indicating that most of the team members understood the goals of the event to a fair degree on the first day of the event. Such uniform understanding across multiple events indicates that the organization has a consistent and sufficient process for introducing the goals to the team members. The mean score, however, does indicate that the methodology could be improved upon. Some tactics could include evaluating the language used to explain the goals or creating an environment where participants feel more comfortable asking questions. Understanding the goals of the event is critical to ensure that the event moves forward effectively, with full participation from all team members.
The data for commitment to change also indicated that there were no statistically significant differences between the 12 events. The mean score was 4.8, indicating that most of the team members agreed that the event would make a positive impact on the work area. As with goal clarity, opportunities exist for improvement. In this particular organization, to facilitate the use of kaizen events, all managers were required to achieve a set number of kaizen events in 2005. This may have left some participants with the perception that the kaizen events were filling a quota and not necessarily needed. As the organization continues to implement kaizen events, demonstrating the need for the event may help employees become even more committed to the events.
The results from goal difficulty indicated no statistically significant differences between eleven of the events. The mean for these eleven events was 3.6 (where 3 = "tend to disagree"), indicating that most of the participants did not perceive the goals of the event to be very difficult to achieve. The mean for the twelfth event, however, was 4.8, indicating that most of the team members perceived the goals of the event to be moderately difficult to achieve. This event was focused on a non-manufacturing department. The goals of the event included reducing process steps by 50 percent, reducing cycle time by 50 percent and reducing errors by 50 percent. Measurements after the event indicate that process steps were reduced by 44 percent and cycle time was reduced by 30 percent. The goal of reducing the errors was not addressed. The results of the event indicate that the participants were able to accurately assess the difficulty of the event.
It was not unexpected to find differences in goal difficulty, as the complexity of event goals would vary. It is encouraging that the team members participating in the majority of the events agreed that the goals would be achievable. Beginning the event with the perception that the goals can be achieved indicates that the goals have been well thought out and positively presented to the group. Beginning a kaizen event with a negative perception of the realistic achievability of the goals could hinder the event process.
Report Out Survey Results
While the kickoff survey measured team member perceptions before the event began, the report out survey measures factors related to the event itself. Nine separate factors are measured in the report out survey: attitude, impact on area, skills, understanding of continuous improvement, team autonomy, management support, action orientation, internal processes, and overall success.
Attitude is defined as the extent to which the event increased team members' liking for kaizen events. Analysis of the twelve events found that the mean from one event was significantly different than the other events. The mean value for the other eleven events was 4.5 (where 4 = "tend to agree"). The mean value for the outlying event, however, was 5.4 (where 5 = "agree"), indicating that the team members of this event felt more strongly that their recent experience had increased their liking for kaizen events. This event was focused on a kitting area. One of the team members had participated in more than 100 events over his/her tenure with the organization, but the rest of the team members had participated in three or less kaizen events. The goals set for the event included reducing throughput time by 80 percent, reducing travel distance by 60 percent, reducing cycle time by 20 percent, and reducing errors to zero. The first three goals were exceeded, with throughput reduced by 82 percent, travel distance reduced by 70 percent, and cycle time reduced by 21 percent. The goal of zero omissions was not achieved by the end of the event, but was still in the process of being implemented.
Impact on area measures the team members' perception regarding how valuable the event will be to the performance of the work area. Analysis of the twelve events found no statistically significant difference between any of the events. The average value for impact on area was 4.8, indicating that the majority of the team members agreed that the events would be of value to the work area. As with commitment to change (measured in the kickoff survey) the quota system utilized in the previous year to manage the kaizen events may have impacted participant perception regarding the need for events. In order to sustain kaizen events, the organization should focus on emphasizing the need for an event, as well as sharing the results of the event with all employees.
The skills variable measures the team members' perception regarding the new skills the participant gained from the event. Understanding of continuous improvement measures the extent to which the event increased team members' knowledge of the philosophy of continuous improvement. For both Skills and understanding of continuous improvement, no statistically significant difference was found between any of the events. The mean for skills across all events was 4.6, indicating that the majority of the team members found that the events continued to provide opportunities to add to their skill set. The mean for understanding of continuous improvement across all events was 4.7, indicating that most of the participants agreed that the events helped them increase their knowledge base. While the scores indicate that learning is occurring, the organization should focus on how the skills learned are applied in everyday settings. To sustain the knowledge gained in the events, the organization must continue to provide opportunities to use the new skills.
Team autonomy measures how much control the team members had over the event. The analysis found no statistically significant difference between any of the events. The mean for team autonomy across all events was 4.8, indicating that the majority of the participants agreed that they were given a fair amount of control over the event and the ideas that were to be implemented to achieve the event goals. To further increase the perception of team autonomy, future planning of events could look at setting clear boundaries at the beginning of the event and decreasing the facilitator's time with the team. The average facilitator spent 83 percent of the event with the team. Allowing the team more time to work on its own may increase the participants' perception of team autonomy.
Management support measures how well the team members felt supported by management. The analysis found no statistically significant difference between any of the events. The mean for management support across all events was 4.7, indicating that the majority of the event participants agreed they had received adequate equipment, supplies, facilitator help, and resources to achieve the goals of the event. To increase the perception of management support, the organization, as part of its commitment to organizational support, could increase the level of management involvement. Three of the events did not have any contact with management. Three of the events did have work area supervisors involved with the event, but higher levels of management were not involved. Asking managers from different levels of the organization to attend report out sessions could help employees perceive a commitment to kaizen events by all levels of management.
Action orientation measures the extent that team members believed their team focused on implementing changes versus planning changes. The analysis did find statistically significant differences in the means, with the events divided into two groups and one outlying event. One event was not included in the analysis because of inadequate response rates.
The first group included three events, with an average value of 2.9 (where 2 = "disagree"), indicating that the team members participating in these events felt that the majority of their time was used to plan rather than implement changes. The team activity logs indicate that the teams spent an average of 76 percent of their time on analysis and supporting activities and 6 percent on implementation activities. All of the events had a different facilitator. Two of the work areas had previously experienced three or more events. For one of the work areas, this was the first event in this work area. Two of the events were focused on support areas and the other event was focused on a peripheral manufacturing work area (repair). Event goals included reducing process steps, reducing cycle time, reducing errors, reducing throughput time, implementing standard work, eliminating rework, reducing standard time, and reducing work in process.
The second group included seven events, with an average value of 3.9 (where 3 = "tend to disagree"), indicating that the team members participating in these events felt that more time was spent in planning and supporting activities rather than implementing changes. The team activity logs indicate that the teams spent an average of 49 percent of their time on analysis and supporting activities and 44 percent of their time on implementation activities. Two of the events had the same facilitator. Five of the events focused on work areas within manufacturing and two of the events focused on peripheral work areas in manufacturing (kitting and re-work). For three of the work areas, this was the first kaizen event. Event goals included improving work flow, reducing batching, reducing throughput time, achieving 100 percent first pass yield, reducing set-up time, implementing standard work, reducing travel distance, implementing one piece flow, reducing cycle time, reducing touches, and implementing 6S.
The outlying event, which did not fit into either of the two groups, was focused on an area in manufacturing. It had an action orientation average of 4.7, indicating that the team felt they spent more time in implementation activities rather than analysis or supporting activities. The team activity log indicates that the team members spent 31 percent of their time in analysis and supporting activities and 58 percent in implementation activities. This was the first kaizen event to target this work area. The event goal focused on reducing scrap rework errors by 80 percent. It may be possible that the single goal created an opportunity to spend more time implementing ideas, rather than gathering data or analyzing ideas for multiple goals.
Due to the different types of events, the goals, and the focus on different work areas, it is not unexpected that differences would be found across events for action orientation. Each event is unique, and as demonstrated by the result for team autonomy, in this organization the team members have control over the direction of the event.
Internal processes measures how well the participants worked together as a team. The analysis found no statistically significant differences between any of the events. The mean across all events was 5.1 (where 5 = "agree") indicating that the team members felt their team had worked together well, in a positive atmosphere with good communication. This result represents the organization's investment in implementing kaizen events that respect new ideas in an environment that encourages open communication.
Overall success measures the team members' perception regarding the overall success of the event. The analysis found no statistically significant differences between eleven of the events. The mean across all eleven of the events was 5.0, indicating that most of the participants believed the events were successful. The twelfth team had a mean score of 3.6 for overall success. This was the first kaizen event to focus on this particular work area. Two of the three goals were not achieved, which may have impacted the team's perception of success. The facilitator indicated that the lack of team member motivation was an obstacle to success. The team members cited several different obstacles to success including difficulty in accessing information and lack of problem definition. One team member also reported that the work area had little room for improvement.
Utilizing the Results for Future Kaizen Events
The results from the kickoff surveys and the report out surveys indicate that this is an organization that has invested many resources in all aspects of its kaizen events. As demonstrated by the results, however, even within organizations with successful events and established processes, opportunities exist to improve. During the planning phase, more emphasis on creating clear goals and decreasing the facilitator's participation in the event could allow employees more opportunities for creativity and participation. During the execution phase, demonstrating the need for the event could increase employee commitment. Providing more opportunities for employees to utilize the skills and knowledge gained through the events could help increase sustainability of event outcomes and further help the culture to accept kaizen events. Organizational support is also necessary for a successful kaizen program, including the participation of upper management. A visible presence at report out meetings will emphasize the importance of kaizen events to all employees.
As demonstrated by this case study, factors that impact a kaizen event may be varied. By continuing to study events across multiple organizations, insights will be gained that can be applied by leaders in a variety of industries to their kaizen programs, providing opportunities for short-term and long-term performance improvement.
This research is supported by the National Science Foundation under grant Number DMI 0451512. The authors gratefully acknowledge the support leaders and team members in the organizations currently participating in this research.
About the Authors
June M. Worley is a Ph.D. candidate in the industrial and manufacturing engineering department at Oregon State University. Her research interests include lean manufacturing, information systems engineering, systems analysis, and human systems engineering.
Toni L. Doolen is an assistant professor in the industrial and manufacturing engineering department at Oregon State University. Her research is focused on manufacturing systems design, lean manufacturing, work group effectiveness, and mobile technology in education.
Eileen M. Van Aken is associate professor and associate department head in the Grado Department of Industrial and Systems Engineering at Virginia Tech and director of the Enterprise Engineering Research Lab. Her research interests include performance measurement, organizational transformation, lean production, and team-based work systems.
Jennifer A. Farris is a Ph.D. candidate in the Grado Department of Industrial and Systems Engineering at Virginia Tech. In addition to lean production and kaizen events, her research interests include performance measurement, product development, and project management