“Six Sigma” is a well accepted philosophy for quality improvement across business and industries, and continues to attract organizations seeking process improvement. Though two decades have already passed since Motorola successfully demonstrated the power of Six Sigma, the tool is still very successful and as a result, very popular.
There are a variety of ways to introduce Six Sigma into a company, involving different levels of structure, time frames, costs and management commitments. But all agree on the basic steps of the improvement process. It is the five stages of the DMAIC process. DMAIC basically is synonymous with Six Sigma when talking about the process to follow, the steps to improvement.
In recent years, however, new developments in Six Sigma have led to the amalgamation of the “lean” tools to bring about certain changes in the methodology and expedite the results. In this context “Lean Six Sigma” (LSS) has now become the industry flavor and many applications are being cited to show its popularity.
To date, however, the proponents of Lean Six Sigma have failed to develop a process to prescribe how to apply Lean Six Sigma. Most articles and textbooks on LSS read like Six Sigma texts with the word “lean” added many times, especially in the title, and with some of the lean tools added to the tool kit. But the process followed is DMAIC with no changes. If that is all Lean Six Sigma is, then it is still just Six Sigma, because Six Sigma never looked to exclude any tools from its expanding tool kit.
In this paper the authors explain the advantages of adopting the “lean” approach and suggest “Double LEAN” as a new methodology for better results. The paper briefly discusses the perceived benefits in comparison with the traditional approach and gives a new expansion to the lean approach to convert it into “Double LEAN”.
In the sections to follow, we briefly explain ‘Lean’ and then provide a similar brief explanation of Lean Six Sigma. The next section demonstrates the value of Lean Six Sigma over the traditional Six Sigma. The paper concludes with a description of the Double LEAN method and a comparison between it and DMAIC.
What is meant by “Lean”?
It is assumed that the reader is familiar with Six Sigma and the DMAIC process. Before looking at Lean Six Sigma, however, it is better to understand the meaning of “Lean”. Lean is a separate approach to process improvement. It traces its ancestry back through Just-in-Time all the way to the Toyota production system. As stated in Wikipedia, Lean is the set of “tools” that assist in the identification and steady elimination of waste (muda). As waste is eliminated, quality improves while production time and cost are reduced. Examples of such “tools” are Value Stream Mapping, Five S, Kanban (pull systems), and poka-yoke (error-proofing).
According to Alukal and Manos (2007), lean focuses on value-added expenditure of resources from the customers’ viewpoint. As used by National Institute of Standards and Technology, lean indicates a systematic approach in identifying and eliminating waste by providing the product at the pull of the customer in pursuit of perfection. This leads to improved quality, better cash flow, increased sales, greater productivity and throughput, and improved morale.
Interestingly enough, the concept of lean is not new and dates back even earlier than the Toyota system to some of the practices followed by the Ford Motor Company in 1920’s. Then the question may arise concerning as to why emphasize lean now.
Alukal and Manos (2007) summarize the reasons as follows:
In addition, it has been realized that waste of resources has direct impact on costs, quality, and delivery. Further lean approach becomes a necessity considering the resource constraints experienced during recession and economic downtrends. Lean also incorporates both incremental and breakthrough improvements. In other words, while the tools and concepts of lean have been available for a long time, the current business environment has created a larger and more urgent need for their use.
Note that all of these reasons, or advantages, of Lean can also be used to provide the same reasoning and advantages for Six Sigma. Lean and Six Sigma are derived from parallel tracks of development that had the same aims in mind. The most notable contribution of each is the notion of continuous improvement.
As stated in http://www.dau.mil/educdept/mm_dept_resources/navbar/lean/01rdg-lean.asp key lean principles are:
Before going on to describe Lean Six Sigma, there is one other interpretation of Lean that must be developed, as it is an important selling point for Lean Six Sigma. Lean, in the sense of no waste, has been interpreted as having less levels of management. In the Lean organization, the organization chart is flatter, bringing everyone closer to the work processes being managed. In this regard, the Lean organization is more flexible, and can respond more quickly to issues and change.
What is Lean Six Sigma?
At the outset, Lean Six Sigma (LSS) is understood as integrating six sigma methodology and the lean tools to reap the benefits on a much larger scale. LSS is fast becoming a buzzword among in industry and is gaining a lot of attention among the quality community. While this is not completely clear as of yet, it seems to be replacing Six Sigma as the next generation approach to many.
While the term “Lean Six Sigma” is quite commonly used among the business and industry circles, the definition varies across different sources and doesn’t mean the same thing. Some common perceptions about Lean Six Sigma (LSS) are:
Because of these differences in their practice and adaptation, Lean Six Sigma is not having a universally common meaning or implementation procedure. But one thing that is quickly recognized is that Lean Six Sigma takes much less time and other resources so that the results are visible in a relatively shorter span of time. It is this perceived image that has made it popular and attractive to industries. In fact, that is the reason for the order of the perceptions just provided. Most companies are adopting it for the first two reasons, not necessarily the inclusion of the Lean tools.
On the other hand, Lean Manufacturing, which is promoted by Toyota, focuses on improving the “flow” or smoothness of work, thereby steadily eliminating mura (“unevenness”) through the system and not upon ‘waste reduction’ per se. Toyota’s view is that the main method of Lean is not the tools, but the reduction of three types of waste: muda (“non-value-adding work”), muri (“overburden”), and mura (“unevenness”), to expose problems systematically. This focus on exposing problems is the key to what makes it a continuous improvement program.
George, Rowlands, and Kastle (2004) define “lean six sigma” as a combination of two improvement trends, namely, making work better using Six Sigma, and making work faster using Lean principles. But a closer look at their book reads like a Six Sigma book with little more except a name change. The Lean tools are included, but they are just describing DMAIC. It leaves the reader wondering “Just what is Lean Six Sigma that is really any different from Six Sigma?”
Arthur (2007) describes “Lean Six Sigma” as elimination of delays, defects, and variation, associated with the processes, using two important tools namely “value stream mapping” and “Balanced Score Card”. This book makes a better attempt at describing the something called Lean Six Sigma, but the large majority of that book as well is basically Six Sigma.
If the leading texts fail to define Lean Six Sigma as a unique methodology, perhaps it is because no clear methodology has yet to be developed for implementing it. After making the case for LSS in the next section, this paper provides that methodology.
Why Lean Six Sigma?
As stated by Hall (2007), Lean Six Sigma can provide a valuable set of tools and help to establish a culture of disciplined data-driven decision making. But again, Six Sigma would correctly make this same claim.
In practice, it can be seen that the reasons for adopting Lean Six Sigma stem from four fundamental barriers to adopting Six Sigma. These barriers to adopting Six Sigma, lead to adopting Lean Six Sigma in an attempt to get the same results.
The first barrier is just the size and number of roles in the Six Sigma structure. To make it work, management must be heavily involved. There are champions, and sponsors, and master black belts, who have to be touch with the improvement projects conducted at any level of the organization. Setting up this framework will be a big job in itself.
The second barrier is the amount of training involved. Every potential team member is trained at the green belt level, and a core of black belts must be trained to lead the teams. This is very costly and takes time to set up before any results can be achieved.
The third barrier is the time it will take to yield results. No results will be achieved until the project begins, and that will not happen until the infrastructure is in place and the training is completed.
The fourth barrier is the cultural change required to make it work. There is always resistance to change.
Another reason for adopting Lean Six Sigma is that there are some differences between the two methods of Six Sigma and Lean that make a combination of the two very attractive. Stephen (2004) states that the there are differences between Six Sigma and Lean production methodologies as shown in Table 1.
Table 1: Fundamental differences between six sigma and lean production methodologies
| Issues/problems/objectives | Six Sigma | Lean Production |
| Focuses on customer value stream | No | Yes |
| Focuses on creating a visual workplace | No | Yes |
| Creates standard work sheets | No | Yes |
| Attacks work-in-process inventory | No | Yes |
| Focuses on good house keeping | No | Yes |
| Process control planning and monitoring | Yes | No |
| Focuses on reducing variation and achieve uniform process outputs | Yes | No |
| Focuses heavily on the application of statistical tools and techniques | Yes | No |
| Employs a structured, rigorous and well planned problem solving methodology | Yes | No |
| Attacks waste due to waiting, over processing, motion, over production, etc. | No | Yes |
A logical justification for blending Six Sigma with Lean is given by Devane (2004). He states that a pure six sigma approach lacks three desirable lean characteristics:
He further states that on the other hand, a pure lean improvement effort has the following shortcomings:
Smith (2003) comments that when six sigma and lean production methodology run separately they will collide with each other and in contrast, a combination of lean and six sigma will have a positive impact on employee morale, inspiring change in the workplace culture because teams see the results of their efforts put to work almost immediately. Similar opinion is expressed by George (2002) who says that Lean Six Sigma directly attacks the manufacturing overhead and quality costs more effectively than any previous improvement methodology because it comprehends both quality and speed. Hence it can be stated that an integrated approach blending lean tools and Six Sigma will produce greater solutions in search of business and operational excellence.
Ramaswamy (2007) has described an approach for integrating training and application of Lean and Six Sigma methodologies that captures the key principles and philosophies of each method. This approach will work in both the situations when an organization is deploying Lean Six Sigma from the scratch and when an organization wants to overlay on an existing Lean or Six Sigma program
However, LSS is not without its detractors. Micklewright (2006) comments that six sigma and lean tools are time consuming and require special efforts to achieve success. He contends that six sigma itself takes longer time duration to put into practice. Quoting the Quality Guru “Kaoru Ishikawa” he states that 90% of the quality problems can be solved by using Seven Quality Control Tools and hence a different program or approach is not necessary.
Guarraia, Carey, Corbett, and Neuhaus, (2008) remark that not every company would reap success by adopting LSS and passing every process through LSS would be a waste. They argue that LSS fails to detect the hidden sources responsible for waste reduction, cost savings or revenue generation and hence it is first necessary to decide where to focus before embarking on LSS journey. They contend that for every Lean Six Sigma success story there are tales of dissatisfaction, and state that in a recent survey of 184 companies, 80 percent say their Lean Six Sigma efforts are failing to drive the anticipated value
Swartwood (2003) states that although by combining Lean and Six Sigma, it is possible to achieve highly effective improvements in a company’s operations, there are weaknesses in this approach. First, the project selection process is not well defined and projects may or may not be aligned with the corporate business strategy and goals. Second, Lean Six Sigma efforts tend to be aligned by organizational functions (silos) rather than by supply chains. This can result in departmental improvements, but fall short in achieving end-to-end supply chain improvements. He further comments that Lean Six Sigma depends primarily on brainstorming for project ideas, and it can be difficult to sustain long-term momentum once the initial waves of projects are complete.
The truth is that most companies adopting this Lean version of Six Sigma are doing it for cost reasons. They know that Six Sigma is popular and effective, but they are not willing to invest in setting up the program and all of the training that is required. The Lean version is sold as having little or no infrastructure. All you have to do is to set up a few improvement teams, task them with a process to look at, and get them going. In fact, I insist on groups being formed before my classes when I do live classes. And the classes focus on those groups, their projects, and the tools that they need as we progress. So, Lean here actually means “cheap” as opposed to the meaning we have been learning.
This leads to the another reason beyond lower cost, faster results. In the Lean version, we go straight for the projects and the results, and within a month or two can show significant savings or improvement to a process. I usually end my Lean classes with presentations to management, and they are always impressed. In the traditional Six Sigma approach, it takes months to get things set up, and then a typical project team can take months more to get results. So here the word Lean is used to mean “quick” again different than the true meaning.
But something is missing here when using Lean Six Sigma. It fails to address the culture changes required. As a result, these efforts are often short lived, and no program survives. When companies are doing it for these reasons, I stress that these are pilot projects, and management has to be setting up a longer term structure for this as we go.
Another reason people choose this approach is that people have heard the benefits of the Lean methodology, but assume it is just for manufacturing. Sometimes we emphasize that the Lean Six Sigma brings the Lean concepts from manufacturing to the service industry. In this context, it is not a Lean version of Six Sigma, but a service oriented version of Lean.
There is no need to elaborate on the best reason for adopting the Lean Six Sigma approach. We believe that the tool sets of these two approaches come together to complement each other. The result is an approach to improving business processes that is better than either one by itself.
Proposed Approach for Adopting LSS
In the previous paragraphs, the LSS approach and its merits and demerits have been described. Based on these observations, we see that the LSS method will give faster results and that it will be less expensive to implement. Since no one argues these conclusions, our recommendation is always to begin any improvement effort with LSS, not Six Sigma.
On the other hand, the claim that LSS provides better solutions is questionable. It should be obvious, however, that a larger tool set can provide better results. Where one question comes in though is that LSS fails to look below the surface at underlying causes of problems in quite the way that Six Sigma does. A second issue is the long lasting effects, where Six Sigma seems to be a more sustainable approach relying on a structure to support it in the organization, not just a few projects here and there. And the third issue is the strategic one, where Six Sigma again provides a better global view of the improvement efforts and links them with the goals of the organization.
Based on this, our recommendation is to begin to build a Six Sigma organization, a structure for sustaining the improvement efforts in the long run, as soon as the initial LSS projects provide valuable results. In this way, LSS builds the momentum, and Six Sigma can then sustain it. The linkage between the two comes from the governance of the LSS projects.
Another key issue is resources. Six Sigma, with its hierarchy of Green Belts, Black Belts, Champions and so forth, provides the resources to make the program work in the long term. LSS claims that training at the Green Belt level can solve most problems. While this is correct, the support provided by more knowledgeable leaders is important. Once success comes from the initial LSS projects, a few individuals should be selected to be provided with more training and groomed for leadership roles, much like is done in Six Sigma.
Finally, we want to provide one suggestion that carries the LSS idea to its extreme in terms of fast and inexpensive. It is a good idea to start a LSS program with no prior training. Our experience is that training can be provided on an as-needed basis (just-in-time, or true Lean) with groups from a company that have improvement projects to work on. The “trainer” then guides the improvement teams, providing training in the tools they will need at each stage of the project. This yields immediate results without the costly investment in training up front.
The approach we describe is, in our opinion, the best way to implement LSS. Perhaps more important though, is that it minimizes all resistance to establishing an improvement system. That resistance usually stems from the cost of initial training, and from the cost of the management structure required, and from the long time needed to see initial results. This approach to implementing LSS avoids all of those problems, allowing larger funding decisions to be made later after valuable results are achieved.
How to Conduct the Improvement Project – The Double LEAN Method
A Six Sigma project follows the DMAIC process steps. But there are problems with this approach. First, the five phases take too long to complete, delaying results. More importantly, it misses key steps and is unclear in where to finish one phase and begin the next.
For an example of a missing step, the Define phase fails to look at the organization and define its processes. It also fails to help choose the process to study for improvement. This is all done in some vague process outside DMAIC, and DMAIC starts with this process chosen.
For an example of how the delineation of the phases is not clear, look at the line between Define and Measure. Define should state the goal for the project, but too often this cannot be done until results are in from the Measure phase. So Define is not complete until much of Measure is complete. The lines are blurred.
Another example is between the Analyze and Improve phases. It is not specified anywhere if identifying the improvement is in one or the other. Most would say it is in Improve, but the results of the Analyze phase identify the improvements. And then it is not clear whether the actual improvements are put in place in Improve or Control. IT is done differently on each project.
In response to this, a new approach is proposed here.. The authors have coined the “LL-EE-AA-NN” acronym for the proposed Double LEAN Approach. Each of the letters represents a particular action to be taken up by the group or the team responsible for quality improvement through the Lean Six Sigma approach. Further every two letters of the acronym constitute a phase and thus four phases are involved in the proposed methodology.
By having four phases instead of five, the time to completion is already reduced. By having clear lines between the phases, or phase gates to apply the term used in project management, the conduct of the projects is made more clear and easier.
The four steps of the Double LEAN method follow:
PHASE 1 – Look and Locate
In the first phase, the LL stands for Look and Locate. Here we “Look” at the organization and its customers and processes and “Locate” the process in need of improvement. The organization is described in terms of its Critical-to-Quality customer-based measures of success, and the set of processes in place to achieve those measures. The organization is modeled in a way that leads to customer satisfaction.
The Look and Locate (LL) phase seeks to identify the process to be studied. The main outputs of this phase are the justification for the selection of the project chosen and the charter for that improvement project. While these outputs are similar to the Define phase of DMAIC, the justification is has much more in it and the lines between this and the next phase are more clear.
Begin by Looking. Look at the business organization. Describe that organization in terms of its business processes. This is the Six Sigma view, or the Process View (the LOOK) of the organization.
The first approach to this step is to use brainstorming. Find the right people and start to list the processes. A convenient approach used is to list the high level processes first, and then list the processes under each of those. Continue this until the 3rd or 4th level of processes to get the necessary level of detail in the LOOK.
A common Six Sigma tool is to use the Voice of the Customer. The suggestion is to use this in conjunction with a high level SIPOC. Start with the Customer, the “C”, and see what the main needs of the customer are. Then, go backward to see what processes, the P’s, are in place to provide those needs.
Once all of those processes are identified, it is time to LOCATE the issues and select the process to be studied. Process mapping can identify where most problems or delays are occurring in a process. A Pareto chart can identify and rank the urgency of looking at any particular process. Financial analysis can help determine which process improvements can save the most money for the time invested. Multi-voting can be used for the selection where multiple factors weigh in on the decision.
Ultimately, the process is selected and a Charter is written. The charter should list the team members and the goals of the project, while also outlining a work plan and a schedule. It should be approved by management, or a sponsor, prior to continuing, or beginning, the actual process improvement work.
PHASE 2 – Explore and Establish
In the second phase, “EE” represents “Explore” and “Establish”.
The basis for improving a process is to understand the process, how it works as well as how well it is working. EXPLORE the process to understand it. Use detailed process mapping or value stream mapping here. The more detailed version of the SIPOC done earlier can also be used. Tools like swim lane flowcharts and spaghetti diagrams are also useful.
Then, collect data on the process to ESTABLISH a baseline, or the before improvement position, on how well the process is performing. These measures are mostly on process outputs. Tools like the Pareto chart, the run chart, the control chart, and the tools of descriptive statistics can all apply.
The main output of this phase is the knowledge and the measures of the degree of problems in the process. At this point, the problems can be seen more clearly, hopefully bringing the solution to those problems into sight.
The fact that this phase ends with the establishment of a baseline is valuable in three ways. From a prescriptive viewpoint of what to do, it provides a clear phase gate in the project. A review can take place to refine what needs to be improved, what needs to be further investigated, and the goals for improvement can be established. It also allows for a decision not to continue the project, if the data does not justify continuing.
The second item of value at this point is the understanding of the current process that is gained from this step. Most times when a process is conducted, especially if done by many people, there is no clear understanding of it. Everyone is doing it differently. This Establish phase yields improvements just by getting everyone doing the same thing.
The third item of value is the baseline itself, the main goal of this phase. The baseline provides a point to look back at and compare to prove later that improvement was achieved. This baseline will be needed in later phases to reach that conclusion of success and to identify the degree of success.
PHASE 3 – Analyze and Apply
The third phase is denoted by “AA” which stands for “Analyze” and “Apply”. In this phase the effects that potential changes in process inputs will have on the process outputs related to customer satisfaction are analyzed. This leads to a list of process improvement actions. These actions are then applied to the process to check the validity of the changes, and to assess the feasibility of implementing them. Again, a clear line is established to end this phase.
To ANALYZE the problems, or the processes, it is necessary to identify the possible process parameters or steps that lead to those out measures established in the previous phase. The linkage between inputs and outputs from the process, the causes and effects, is the most critical step in this entire methodology. By linking the causes of problems to the problems, we can then APPLY that knowledge to identify the improvement steps to take and try out those improvement actions. The goal is to improve the outputs by modifying the process inputs.
The most common tool in this phase is the cause and effect diagram. Many times the Pareto chart is used here, most often looking at the items listed in the cause and effect diagram. Another approach is the use of the 5 Whys. Some of the Lean improvement tools can apply in this phase. For most processes, these tools suffice. But the statistical tools of correlation analysis, especially regression for quantifying the relationships between inputs and outputs of a process may be helpful. In situations where there is a lot of data available, especially manufacturing, design of experiments can be used.
The output of this phase is the identification of the improvement actions to take, and their use on a trial basis to test them out. The tools of hypothesis testing and ANOVA can be used to verify that improvement is achieved.
Another key phase gate is introduced here. In the DMAIC, the improvements are really identified in Analyze, but not mentioned until the next phase. But since these cannot really be separated, they are combined here. At the end of the phase, another phase review can take place. The improvements are reviewed and any required investment decisions can be made. The decision to go ahead or not with implementing the improvements is made before going to the next phase.
PHASE 4 – New and Navigate
The fourth phase consists of “NN” indicating “New” and “Navigate”. In this phase a “New” process is developed based on the evaluation and improvement implementation of the existing process. Schedule and budgets are prepared to implement the desired changes and improvements. Besides the improvement tools, project management and change management tools are introduced to facilitate the move to the improved process. Plans are introduced here to “Navigate” the use of the new process, to ensure that the improvements achieved are monitored and maintained.
The improvements have been selected, and the process improvements are ready to implement. The NEW process should be described with a new process map. The tool of process mapping should be used. The major tools needed now are the tools of project management to implement the new system. Change management may also be needed.
The improvement process is now ready to use with the new system. Finally, a plan to NAVIGATE the new system needs to be prepared. The poka yoke can be used to prevent problems from occurring. New operating procedures (SOP’s) are developed. Control processes like user logs or control charts are used. The intent is to make sure that the system continues to show the improvements and that they are not just temporary. To some degree, this Navigate is similar to the Control phase of DMAIC, but more fully developed.
At the end of this phase the improvements should be visible and the cycle can be repeated with other processes.
Conclusions
This paper has clearly established that there is good reason to adopt a LSS approach, either in place of Six Sigma or as an introductory step toward adopting Six Sigma. It also shows that no methodology exists for conducting a LSS project, separate from a Six Sigma project.
An approach to implementing a LSS program is developed, providing suggestions for how to plan and manage that evolution in the company. It can be done independently or in conjunction with a Six Sigma program.
The main result of this paper is the method for conducting a LSS project. This is the Double LEAN method. The four phases are each described. They are:
LL — Look and Locate
EE — Explore and Establish
AA — Analyze and Apply
NN — New and Navigate.
The Double LEAN method is more prescriptive and easier to use than the DMAIC. In addition, the establishment of phase gates is a major improvement over the DMAIC.
References