23 January 2012

QFD and Six Sigma DMAIC

from the QFDI newsletter, January 2012

Efforts to standardize and strengthen product and process improvements are greatly welcomed in modern organizations, and nothing does it better than the DMAIC (define-measure-analyze-improve-control) approach in six sigma. This 21st version of Shewhart’s and Deming’s PDSA (plan-do-study-act) technique is the backbone of ongoing quality improvements in today’s leading companies.

The purpose of quality improvement is to eliminate the costs and losses associated with defects and deviations from targets. The process starts with defining those targets and how to measure them, and then determining what the current level of performance is. Next is to identifying the causes of why current performance fails to meet those targets.

Causal factors can be analyzed as those related to the 4 Ms, or those attributable to workers (men), equipment (machine), processes (method), or design (materials). Advanced thinkers may also include method of measurement (ex. poor gauging or measurement techniques), money (lack of funds to make desired improvements), and management (lack of top level support to invest in and support a quality culture). Let us know if you have identified other “M”s, please.

Once causal factors have been identified, data analysis helps focus on those with the strongest contribution to improving performance. Since DMAIC attends to current products and processes, data can be collected to statistically calculate the correlation between a causal factor and the undesirable effects of the defect.

Improvements to the undesirable effect are made by improving these highly correlated causal factors, and training workers, upgrades or better maintenance of equipment, new processes, or better design are investigated and tested. In addition to efficacy of these improvements, other feasibility constraints such as cost to implement, time to implement, etc. are considered in deciding what and when to implement.

Once the improvement is in place, standardization of the improvement is needed to prevent falling back to old ways. Thus, ongoing data collection helps control any deviations to the new process.

What Role Can QFD Play in DMAIC?

QFD has typically focused on new product development; the large time-consuming “houses” of classical QFD are considered overkill for addressing concerns with local problems associated with production or service delivery failures. Rather, QFD is an approach for identifying customer needs far upstream from production, even prior to design phases in order to define quality from a customer or user perspective and assure it is designed into the new product and quality assured during its build and delivery.

Yet, more and more these days we are seeing six sigma black belts and master black belts studying QFD in order to improve their work. Often, they are manufacturing engineers, service quality experts, and others on the front line of quality. I see them as new customers for modern QFD (our traditional customers are business developers, marketers, product developers, and design engineers), and so we need to adapt our methods to help them do a better job of DMAIC.


In production or service quality, improvement opportunities are identified by customer complaints or internal quality audits. QFD’s powerful Voice of Customer (VOC) analysis front end can offer a more preemptive approach to the define step in DMAIC.  One key VOC tool, customer gemba visits which are typically used for product refresh or next generation models, can be scaled down to identify smaller improvement opportunities in the customer’s experience that could be effected with minimal design changes.  This is especially true when there are fewer mechanical or interfacing system constraints.

In other cases, manufacturing functions see downstream production processes as “customers” of manufacturing engineering. This is wholly consistent with quality thinking known in Japanese as ato kotei wa kyaku or next step in the process is your customer. Gemba visits now revert back to the more common definition of our manufacturing facility, and VOC can focus on the voice of operators, material handlers, inventory or tool crib management, trade skills such as electricians, etc. to both help manufacturing engineers understand the process or equipment needs of these downstream activities. Further, VOC thinking can also help these downstream activities their “customers” sub-processes, etc.


QFD tools such as Gemba Visit table and the Quality Planning table capture how customers measure fulfillment of their needs from their perspective. These tools capture importance, current magnitude of performance, and hoped-for level of satisfaction. Understanding how customers measure their level of satisfaction gives us insight into how they choose what to buy.

New product developers know how critical this understanding is during design and development, but how can QFD add value to six sigma’s “measure” step? 

As previously stated above, the measure step helps the quality improvement team determine how to measure successful resolution of a problem, as well as to determine what is the current and target level of performance. Since the customer is the final arbiter of successful solution to a problem, this added perspective should better help teams focus their improvement activities. This should hold true for both internal and external customers.

The tools cited above are part of our QFD Green Belt® course. For the next public class, please see the Calendar. Of course, in-company courses can be arranged at any time.


  1. Nice blog thanks for the information.

    Six Sigma Training || Six Sigma

  2. Hi
    i'm actually a final year student in one of the local university...
    i got a set of industry data with both numerical and non numerical data... and i need to analyse the data to determine which factor is the crucial towards the output..
    for instance there are 5 factors that might affect the output and i need to determine which factors are important based on the data i have.
    so, may i know in such situation is six sigma applicable?Six Sigma Certification

    1. QFD is a method for determining which design factors are crucial to satisfying customer needs. Usually the design factors are objective and measurable levels of performance or functionality (numerical) and the customer needs are subjective but still measurable levels of satisfaction (numerical or non-numerical). The QFD process allows for a team of combined marketing and engineering functions to evaluate and quantify the strength of the relationship between these two data sets. The Analytic Hierarchy Process is recommended.

      Your question appears to be more about statistical methods. To explain cause and effect relationships, please consider Multivariate Analysis. If both your output data and your input data are numerical, then Multiple Regression may be useful. If your output data are non-numerical but your input data are numerical, than Discriminant Analysis may be useful. When both input and output data are non-numerical, there is a special set of tools created by the late Dr. C. Hayashi of Japan called the Quantification Theory I, II, III, IV.

      There are many, many tools in the six sigma toolbox and part of mastering this method is to know which tool to use in which application.

    2. Hey,
      Very nice site. I came across this on Google, and I am stoked that I did.
      I will definitely be coming back here more often.
      Wish I could add to the conversation and bring a bit more to the table, but am just taking in as much info as I can at the moment.Six Sigma Certification

  3. Hi Guys please tell me what is sigma and how to work it.

    1. Hello onlinebid1, this entertaining video may be helpful to you:



Comments are welcome and will be posted upon review.