Lean & Six Sigma Methodologies

In this article, we will address an expanded view of DMAIC, DFSS (including its component methodologies), and Lean; how these key methodologies relate to one other; and the emerging concept of business process management.

In addition, we’ll observe how the tools align with each phase of define-measure-analyze-improve-control (DMAIC).

Many times, the discussion of Six Sigma, DFSS, Lean, or even project management with senior management results in stares and a certain level of disinterest. Often this is because they do not understand how these methodologies are separated or used. Consider this: how many times within your organization have all projects been considered Lean Six Sigma projects regardless of which methodology is appropriate?

Figure 1 illustrates how each methodology is chosen.

FIG 1 Selecting the proper project methodology

We will detail in the following articles each of these methodologies.

DMAIC

As all Lean Six Sigma professionals know, DMAIC is the de facto methodology for improving processes using Lean Six Sigma. The reason for this is quite simple. The methodology is:

• Easy to understand
• Logical
• Complete

When belts first learn about DMAIC, they, for the most part, see it depicted in a linear manner as shown in figure 2.

FIG 2 DMAIC in a dream world

Unfortunately, this is an illusion that leads inexperienced belts to plow ahead from phase to phase, steamrolling weak champions/sponsors, particularly if the belts lack the support of a Master Black Belt coach. When this occurs, projects either fail or results are not sustained.

Figure 3 depicts DMAIC in the real world.

Fig 3. DMAIC in real world

As we all know, tollgates occur after each phase. Diligent application of the proper tools and, above all, significant preparation are required to pass them. Passing them is not a given. Notice that up to and including the analyze phase, the failure of a tollgate could set the project back one or two phases. This usually occurs when knowledge is gained in a phase contrary to expectations. In this case, the project team may be forced to retreat all the way back to define in order to restate the problem. Beyond the analyze phase, a failed tollgate usually results in correcting or performing additional work within the phase. However, at any phase, a project may be terminated. Generally, this will happen when projects are no longer synchronized with strategy, champions/sponsors have lost interest or transferred, or the potential savings are less than expected.

Consider this very simplified example:

I have just received my annual performance appraisal, which was based on a 360-degree review process. I am not happy with the result since on a scale of 1-5 (5 being high), I averaged a 2. However, rather than complain, I decide to take the results to heart and do something about them. After careful consideration, I plan to use the DMAIC approach as a means of driving personal improvement:

• Define. I am unhappy with my personal performance and want to move my average performance from a 2 to a 3 by the next annual performance appraisal.
• Measure. The 360-degree performance appraisal is very detailed and allows me to decompose my average result into various components that I can associate with particular behaviors. Each of these component parts has, in turn, an average result determined for them as well.
• Analyze. Since I am able to analyze the results of each of the component parts, I create a Pareto chart to determine which components have the greatest influence on my overall average score. Consequently, I am able to determine the top five behaviors. This is an important consideration since I must balance my actual productive work time against my personal improvement time and work within the time allotted to my annual training plan.
• Improve. With the top five drivers of behavior identified, I set about to develop and execute a plan. In this case, my plan consider primarily of taking company-sponsored training and education courses.
• Control. As I take the courses, I diligently practice what I have learned, and decide that it might be beneficial to keep a journal that details any particular interactions that might influence my performance ratings. This allows me to keep what I have learned in the foreground and to practice it continuously. In addition, I decide to use intermediate 360-degree performance appraisals that “don’t count” as I am the only one who will see the feedback. As long as the feedback on the top five that I am focusing on is improving, I am on track. If not, I will need to revisit my plan and make appropriate changes. Furthermore, I am concerned that behaviors that were “in control” have not moved in the wrong direction. If any have, they will be included in the plan moving forward.

Tables 1 through 5 have been compiled to reflect the current thinking across contemporary authors about which tools should be applied in which phases of DMAIC.

DFSS

While DMAIC may be traditionally viewed as the foundation for Lean Six Sigma, its application is primarily limited to improving existing processes; it does little to address the design of new products or processes.

This concept is illustrated in the flowchart given Figure 1.

Fortunately, several additional structured methodologies exist. Several of the common methodologies are depicted in Table 6. This table has been created to provide a loose comparison between the DMAIC and DFSS methodologies.

Table 7 identifies some of the commonly used tools in DFSS, including the DFSS methodologies shown in figure X. However, this table also lists other tools that are useful regardless of the DFSS methodology chosen. Examples include axiomatic design and TRIZ.

DMADV

DMADV is a well-recognized Design for Six Sigma (DFSS) methodology, the DMA portion of DMADV is same as DMA of DMAIC, so we’ll concentrate on the remaining DV portion:

• Design. Quite simply, this means carrying out the process of designing a product or process. Many organizations have well-established policies and procedures for their respective design processes. One valuable Lean Six Sigma technique that supports the design process is quality function deployment (QFD). Additional tools useful in this phase include pilot runs, simulations, prototypes, and models.
• Verify. This phase is directed at ensuring that the design output meets the design requirements and specifications and is performed on the final product or process. Basically, verification speaks to the design meeting customer requirements and ensures that the design yields the correct product or process. By contrast, validation speaks to the effectiveness of the design process itself and is intended to ensure that it is capable of meeting the requirements of the final product or process.

Both verification and validation are necessary functions in any design process. As such, this suggests that DMADV might be more appropriately named DMADVV.

DMADOV

The basic difference between DMADV and DMADOV is that “O” for “Optimize” has been added. This sounds like a trivial observation, but many organizations’ design processes do not include this refinement action. Oftentimes they produce only a minimally workable product or process. DMADOV forces attention on the need to optimize the design. Additional tools useful in this phase include design of experiments (DOE), response surface methodology (RSM), and evolutionary operations (EVOP). These methods help the design team establish and refine design parameters.

DMEDI

The define-measure-explore-develop-implement (DMEDI) methodology is appropriate where the limitations of DMAIC have been reached but customer requirements or expectations have not been met, or a quantum leap in performance is required. Unlike DMAIC, DMEDI is not suitable for kaizen events and it is generally more resource-intensive, with projects taking considerably longer.

Let’s explore ea ch phase of DMEDI:

• Define. This phase is the same as in DMAIC
• Measure. Unlike measure in DMAIC, this phase must emphasize the voice of customer and the gathering of critical customer requirements since no baseline data are available. The quality function deployment (QFD) tool is valuable during this phase.
• Explore. This phase focuses on creating a high-level conceptual design of a new process that meets the critical customer requirements identified in the measure phase. Ideally, multiple designs are developed that provide options to be considered.
• Develop. The options developed in the explore phase are considered, and the “optimal” one is chosen based on its ability to meet customer requirements. Detailed designs are developed.
• Implement. During this phase, the new process may be simulated to verify its ability to meet customer requirements. Also, a pilot is conducted, controls are established and put in place, and the new process is transferred back to the process owner.

IDOV

Unlike most of our DFSS methodologies, IDOV contains only four phases and stands for identify-design-optimize-validate.

Let’s explore each phase of IDOV:

• Identify. This phase links the design to the voice of the customer via the customer, product, and technical requirements, establishes the business case, identifies critical-to-quality (CTQ) variables, and conducts a competitive analysis. Quality function deployment will be particularly useful tool during this phase.
• Design. This phase focuses in identifying functional requirements, deploying CTQ’s, developing multiple design concepts, and selecting the best-fit concept.
• Optimize. This phase focuses on developing a detailed design for the best-fit design chosen in the previous phase, predicting design performance, and optimizing the design. Additional work in this phase includes error-proofing, statistical tolerancing, prototype development, and optimization of cost.
• Validate. During this phase, the prototype is tested and validated, and reliability, risks, and performance are assessed.

LEAN

As with Six Sigma, the tools of Lean are not new, only rediscovered. Table 8 lists many of the common tools used by Lean practitioners.

While Six Sigma focuses on reducing process variation and enhancing process control, Lean-also known as lean manufacturing-drives out waste (non-value-added) and promotes work standardization and flow. These methodologies are complementary, not contradictory as some organizations and practitioners of each believe and defend.

In recent years, the demarcation between Six Sigma and Lean has blurred. We are hearing about terms as “Lean Six Sigma” with greater frequency because process improvement requires aspects of both approaches to attain positive results. It is for this reason that Six Sigma practitioners need to be well versed in the tools of Lean.

Many successful implementations have begun with the lean approach, making the workspace as efficient and effective as possible, reducing the eight wastes, and using value stream maps to improve understanding and throughput. When process problems remain, the more technical Six Sigma statistical tools may be applied. Furthermore, both methodologies require strong management support to make them the standard way of doing business.

Some organizations have respond to this dichotomy of approaches by forming a Lean Six Sigma problem-solving team with specialists in the various aspects of each discipline but with each member cognizant of others’ fields. Task forces from this team are formed and reshaped depending on the problem at hand.

Unfortunately, other organizations have decided to adopt one approach to the exclusion of the other. Perhaps this is due to a strong influence by misinformed senior management and/or possibly both weak skills and leadership on the part of the deployment leader. Master Black Belts must be evangelists and passionate about Lean Six Sigma. They must be able to carry forth the word and communicate the symbiosis that exists between Lean and Six Sigma, or any other methodology for that matter, and how those relationships can be exploited to the benefit of the organization.

 Best regards

Youness LAAMIRI

PhD, LSSBB, PMP, EMBA