Perfection by the Numbers: Key Learnings from Six Sigma

Six Sigma, at many organizations, simply means a measure of quality that strives for near perfection. Six Sigma is a disciplined, data-driven approach and methodology for eliminating defects (driving towards six standard deviations between the mean and the nearest specification limit) in any process — from manufacturing to transactional and from product to service.

imageThe statistical representation of Six Sigma describes quantitatively how a process is performing. To achieve Six Sigma, a process must not produce more than 3.4 defects per million opportunities. A Six Sigma defect is defined as anything outside of customer specifications. A Six Sigma opportunity is then the total quantity of chances for a defect.

(www.isixsigma.com)

What is “Six Sigma”

Good question. Imagine you order room service at a hotel, and you are told the service will arrive in 30 minutes. One major hotel chain found service that arrives too soon dissatisfies a guest nearly as much as service that arrives too late. So a hotel that wants to satisfy guests would strive to deliver the room service within some reasonable window — around 30 minutes. Because of variation in the cooking time, staffing levels, elevator speed, distance of room from kitchen, complexity of order and so on, there will also be variation in the resulting delivery time for different guests. However, in order to maximize guest satisfaction, the hotel needs to not only hit the 30-minute target on average, but also to do it consistently.

Statisticians use the lower case Greek letter sigma to signify the standard deviation. Standard deviation is the variation, or deviation, of a set of numbers such as the arrival times of room service orders. When we say that a process is a Six Sigma process, we mean that the window of acceptable performance (from the customer perspective) is six standard deviations from the average actual delivery time. In terms of a probability, that means that a room service order would arrive outside of the acceptable window only three times out of every million orders, essentially perfection.

Six Sigma means both a measure of performance as well as a method of process improvement to achieve that desired performance. It includes a sound philosophy, a step-by-step methodology, and analytical tools. Even if your process output is nowhere near the three defects per million mark, the Six Sigma method takes you in the right direction. The Six Sigma method breaks through the barriers that hinder your performance.

Six Sigma represents the evolution of decades of thinking about quality and process improvement. Your first reaction to Six Sigma may be that it is simply a twenty-first century repackaging of familiar, old ideas and tools that have been around for decades. Some have called Six Sigma “structured common sense”, or have discounted it because of that familiarity, but it is precisely that structure that makes Six Sigma so effective. Six Sigma combines the best of past improvement efforts into a comprehensive package that includes alignment with business goals, support from senior management, measurable financial impact, and data-based decisions.

Where and How Do You Use It?

You may be thinking to yourself – what has any of this to do with software development, in general, and my project, specifically? Well, Six Sigma can improve any process that has an output. When you stop and think about it, that really covers most of the things we do on a daily basis. From the time we get ready for work in the morning to the time we wrap things up and head home for the evening, we participate in and develop processes. Designing, building and manufacturing a product is obviously a process with a product (actually, a series of processes), but so are marketing, selling and servicing that product. That is why Six Sigma has been successfully applied to a wide variety of operations, from manufacturing plants that stamp automobile fenders to banks that process loan applications.

Regardless of the nature of your business, the ideas and tools of Six Sigma hang on a common framework or methodology. That framework comes in two basic flavors:

  • The first and most common, DMAIC, is best for improving existing processes. DMAIC stands for Define, Measure, Analyze, Improve and Control.
  • The second flavor, DMADV, follows a methodology of Define, Measure, Analyze, Design and Verify. DMADV is better suited for designing new products and processes.

The two methodologies differ in small but meaningful ways, and share many of the same ideas and tools.

To use Six Sigma, you must first define a problem to be solved. Six Sigma then takes on the form of a project lasting several months, during which the project leader, usually known as a “black belt” or “green belt”, follows one of the methodologies described above and applies the appropriate tools along the way. After successfully completing the first project, the project leader becomes certified in Six Sigma. Criteria for certification vary from company to company, but often also include completion of several days of formal training and an examination.

Begin and End with the Customer

When defining the problem of poor quality, lost revenue, or high costs, it is not uncommon for employees to blame the customers. They might say things like, “our customers are not sophisticated (i.e. “intelligent”) enough to use our product correctly” or “it’s another case of user error.” If the customer is internal, they might assert something like, “the department we hand off to doesn’t understand the way we work” or “it’s a training issue, not a functionality issue.” Such statements are counterproductive, and only serve to entrench companies in their inadequate ways, making real improvement and customer satisfaction that much more difficult to attain.

Six Sigma, on the other hand, begins and ends with the customer. That includes internal customers as well as the more obvious external customers. That’s because the customer ultimately defines quality. Granted, the voice of the customer can be emotional, contradictory and subjective. Green belts and black belts must refine that voice of the customer into objective, measurable characteristics that are critical to quality, or CTQs for short. CTQs become the basis for primary and secondary metrics by which we measure the success of a Six Sigma project.

Examine the Process

There is a reason for everything! In other words, the quality of a product or service doesn’t just materialize out of nowhere. The process and the inputs that go into that process determine the results. By characterizing the process in the Analyze phase using process data and experimental data, Six Sigma practitioners uncover the relationships between causes and effects that allow them to make improvement breakthroughs.

Compare this thinking to software modeling. Before you build, you need to understand your actors, your use cases, and the interactions and relationships between all objects within the system, allowing – as in Six Sigma – your development team to make improvement breakthroughs.

Our natural tendency is to focus on the results or outputs. A typical result of such thinking is to reward employees for achieving a desi
red goal without regard to how they reach that goal. Such improvements often turn out to be unsustainable and impractical in the long run. Over time, such programs seem to come and go and often end up being an exercise in futility. A better practice is to reward employees for achieving improvements to overall process, and those things that are repeatable and sustainable.

Six Sigma goes below the surface to the root of the problem, which more often than not is the process and its inputs, not the people. By first defining what the important process outputs should be (the CTQs), and then understanding the nature of the relationships between process inputs and outputs, you can use those relationships to your advantage. The focus of Six Sigma improvement efforts rests on the process itself and on the process inputs, not the outputs. By manipulating, controlling and compensating for the process inputs, you ensure acceptable process results that meet the critical to quality requirements. .

Show Me the Data!

Six Sigma requires data. One large financial institution drives this point home by asking Six Sigma trainees to yell, “Show me the data!” Without data, your analysis and conclusions are just opinions. Many past quality initiatives based their analysis on subjective judgments and gut feelings. Not surprisingly, the resulting improvement efforts often fell short. Six Sigma doesn’t totally discount personal experience. Following best practices is key to ongoing process improvement. In fact, Six Sigma includes tools for gathering and classifying that sort of information, but in the end all such claims need to be backed up with hard facts.

The tools for examining data fall into two broad categories: graphical and statistical. Much of the required Six Sigma training is devoted to learning how to apply these tools correctly. Graphical tools start with a process map, a diagram that shows not only the relationships between process steps, but also the inputs, outputs, and current performance for each step. Process mapping alone has revealed the source of many a seemingly impossible problem.

Statistical tools include descriptive statistics, hypothesis testing, regression, design of experiments (DOE) and statistical process control (SPC). As an example of why such statistical tools are needed, consider the following scenario:

You poll one hundred customers and find that seventy of them are “very satisfied”. You then go on to spend money, time, and other resources in order to improve that satisfaction level. Six months later, you poll another one hundred people and find that eighty are “very satisfied”. Did you really make an improvement, or is the difference within the margin of error?

One of the statistical tools, hypothesis testing, answers that question. As it turns out, the difference between 70 out of 100 and 80 out of 100 is not statistically significant. In other words, the difference could be due to random chance. That being the case, do you really want to make intelligent business decisions based on dumb luck? Probably not!

Now’s the Time

Six Sigma is a method whose time has come. If you are considering implementing Six Sigma, keep in mind that it requires a significant investment of time and resources. You’ll need to excuse the participants from their normal responsibilities so that they can concentrate on learning and applying the methodology. Although results begin to show in a few months, it may take several years before you see the return on investment you seek.

The future will probably bring some fine-tuning to Six Sigma. After all, one of the underlying assumptions of the method is that there is always room for improvement. However, the heart and soul of it won’t change drastically. That’s because both the individual components of Six Sigma as well as the complete package have proven so successful in yielding tangible financial results. In the end, Six Sigma affects the bottom line, and every business in the world understands that.

Finding More Information on Six Sigma

If you are interested in finding out more about Six Sigma, here are a few links:

  • Your first stop for information should be within your own organization. Most large corporations have a group or vendor responsible for Six Sigma training and information.
  • If your company doesn’t provide this kind of resource, you may want to start with Quality Progress Magazine and Six Sigma Forum Magazine (www.sixsigmaforum.com), both distributed by the American Society for Quality (www.asq.org). While ASQ does not focus primarily on Six Sigma (and some feel that they are a late entrant into the space), they are a leading industry group, and provide great information.
  • If you’re looking for a comprehensive site by an industry group that is dedicated to Six Sigma, then take a look at iSix Sigma (www.isixsigma.com). Their site contains content, an activities calendar, a bookstore, a newsletter, online surveys, and discussion forums.
  • Another great site and magazine is Quality Digest (http://www.qualitydigest.com), which contains links to numerous white papers and case studies, such as how Motorola received the Malcolm Baldridge National Quality Award with the help of their Six Sigma practice (http://www.qualitydigest.com/dec97/html/motsix.html).
  • Take a look at your local Six Sigma vendors, or find out who trained your team or that of a customer or partner. Most of these vendors are highly qualified, and happy to provide you with detailed information on getting you trained or your company certified. For example, Thomas A. Little Consulting (www.dr-tom.com) provides training and train-the-trainer, and auditing, implementation, and certification services.
  • Sith Sigma (http://sithsigma.wordpress.com/), while it has little to do with the formal quality process, is billed as “a fun way of helping technologists incorporate the business perspective for total success” using the wisdom of the dark side of the force (particularly good is the Sith Business Philosophy).
  • And then there’s Six Sigma, the band (www.sigmamusic.com). While we haven’t actually heard their music, with a quality name like that, they have to be good, right?

 

(Co-authored by Carl Ashcroft, former SVP of Quality and Productivity at Bank of America, now a Business Operations Manager at Cisco Systems)

Christian Buckley

Christian is a Microsoft Regional Director and Office Servers and Services MVP, the Founder & CEO of CollabTalk LLC, an independent research and technical marketing services firm based in Salt Lake City, Utah, and CMO of revealit.io, a blockchain-based video technology company.