1. What Lean and Six Sigma tools are a good place to start focusing on continuous improvement?
The biggest challenge that most organizations face center around bridging the gap between strategic vision and strategic execution; this is often the best place to start. In the C-levels of many organizations there is a perceived failure of strategic execution. They feel that they have set the vision, but the people executing are working against that vision. At the front lines of many organizations people feel that they are working diligently to solve problems in their processes, but they are not clear on the strategic vision.
While it is not necessarily a Lean or Six Sigma tool, I like to “Start with Why.” An organization must have a unified focus if they are to drive toward something. Organizations that fail to define their why will exhibit behaviors where it seems that departments are working against each other. I believe that people want to drive toward a common goal in their organizations. Finding their why is a great place to start. Sustaining a continuous improvement effort will be difficult at best if this is missing. To get started I’d recommend watching the TED Talk video by Simon Sinek, “How great leaders inspire action.” A great resource to discover implementing this for yourself or for an organization is the book by Sinek, Mead, and Docker, “Find Your Why: A Practical Guide for Discovering Purpose for You and Your Team.”
Once this unifying vision is set then an organization can align strategic execution with the vision. My favorite process for this is Hoshin Kanri. Hoshin Kanri is from the Lean toolbox. The brilliant Wes Waldo of BMGI (Now the Lean Methods Group) wrote a great article on implementing Hoshin Kanri, “The Seven Steps of Hoshin Planning.” Finding Your Why is a great way to complete step one of Wes’s process. An important concept here is that this must be iterative. Do not approach it as a waterfall process.
Deciding on breakthroughs for your Hoshin Kanri requires an understanding of the future. Overemphasizing the past can be a devastating mistake since the past is something we cannot change. Your deployment should be focused on improving the future of your organization. Hoshin Kanri will help align strategic execution with strategic vision; this, by definition, is future focused. Unfortunately anything future focused will introduce risk.
There are many methods that try to simplify understanding the future, but one I have recently discovered and find helpful is the Anticipatory Organization by Daniel Burrus. The key to this method is to reduce future risk by breaking down future trends, as described by Burrus, into Hard Trends, Soft Trends with hard assumptions, and Soft Trends with soft assumptions. I won’t try to describe the process here, but I would recommend picking up a copy of the book.
Following this process is a good road map for discovering where to start focusing. Michael Porter once wrote that, “The essence of strategy is choosing what not to do.” This is an important point to decide what to focus on. New deployments will have a lot of low hanging fruit. The challenge will be in how to utilize the limited resources you have at your disposal. Use Finding Your Why, Hoshin Kanri, and Anticipatory Organization to improve your odds of success.
2. Which type of charts do you find most useful?
I will assume that this question is about control charts. Most of my experience comes for the services industry and I am sure that has influenced my choice here. In the services industry we are faced with some unique challenges. You will encounter much more special cause variation than what will typically be found in manufacturing. This is because people are typically involved on both sides of the process.
I find that the Laney P’ and U’ are great control charts to combat the variability found in services processes. These variations of the traditional P and U charts were developed by David Laney. An interview between the folks at Minitab and David Laney can be found here: “On the Charts: A Conversation with David Laney.”
The Laney charts will continue to grow in importance in the coming years. Obtaining data is becoming cheaper. As more data is pushed into control charts there is more opportunity for the traditional charts to display subgroup or sampling variation. They can look quite chaotic. The Laney P’ chart is great at overcoming this problem.
3. What is a differentiator between Lean & Six Sigma continuous improvement versus other methods?
To best answer this I would first say that I am method agnostic. I believe that implementing a philosophy and culture of continuous improvement is far more important than the specific method used. With that said I do feel that Lean and Six Sigma used together are an especially effective method to execute continuous improvement projects.
Lean is great because it is fairly easy to understand. Many gains in Lean can be realized through observing processes and implementing solutions with minimal understanding of mathematics and statistics. This is not said to downplay the impact that Lean can have on an organization. Additionally Lean is a tool-set that is very difficult to truly master. I recommend implementing Lean early in a continuous improvement deployment.
Six Sigma is great because it can reveal less obvious root causes. Unfortunately, truly implementing Six Sigma will require a deep knowledge of statistics and analytics that can be difficult to procure in an organization. I believe that Six Sigma will continue to evolve and improve as a continuous improvement method. The coming wave of machine learning, artificial intelligence, and big data can be utilized in Six Sigma to maximize outcomes. These skill sets will also become more difficult to procure.
The greatest benefit to a continuous improvement deployment comes when Lean and Six Sigma are combined. The two overlap well and can bring maturity into a deployment. The best way to think of this is in terms of continuous improvement. Launch your deployment with the understanding the perfection is the unending unattainable goal. Work each day to improve your deployment.
4. Which internal teams should be involved in Lean and Six Sigma continuous improvement initiatives?
In short, every part of an organization should be involved in continuous improvement initiatives. The best organizations are moving toward a unified vision or goal. They realize that they are part of system with nested subsystems. Each subsystem must, by definition in this analogy, work in unison with every other subsystem to maximize outcomes.
Imagine the organization as an internal combustion engine where the pistons are operations, the fuel is marketing, and sales are the spark plugs. Tuning all of these separate components to work in unison is the key to delivering power. Now imagine that the fuel is adjusted without consideration of the pistons or the spark plugs. What will happen? Will the engine produce more power or will it sputter and die? There is a chance that ignoring the other components will produce more power, but it is less likely than working them in unison.
An organization operates in much the same way. Thus, every part of the organization must be involved. This is obvious to small organizations, but it is exceedingly difficult in large organizations. Methods to overcome these difficulties will depend on the structure, culture, and size of each organization.
5. How do you address human error improvements?
Human errors are typically a symptom rather than a root cause. Address these by first understanding the true root cause of the problem. There are many root cause analysis tools that can be used, but one I especially like is 5 Whys.
Using 5 Whys to understand root cause should be used with some caution. Some common mistakes in application come from strict adherence to the number, following rabbit holes, and asking why question. First understand that there is no magic number of questions and second understand that mastering the method will require intuition. Jon Miller recently gave a great interview on the Gemba Academy podcast on “How to use 5 Whys.” A continuous improvement practitioner should also heed caution with how the question is asked. People may respond emotionally when asked Why rather than What or How. I cover this on a Gemba Academy podcast that will air next week.
Once you truly understand the root cause you should then classify it as confusion, anxiety, resistance, frustration, or false starts. Confusion results from vision, anxiety results from missing skills, resistance results from lacking incentives, frustration results from missing resources, and false starts result from failed action plans. Address the root cause to solve the issue.
6. Would you ever use tools used during the Define phase again? (i.e. SIPOC, Value Stream Mapping, etc)
Yes! Tools used in the define phase can, and should, be used again. A SIPOC or Value Stream Map can be used to understand the current state, the perfect state, and the future state. Stakeholder analysis should be used throughout a DMAIC to continuously understand the changing needs of stakeholders.
A common mistake in the DMAIC or DMADV methodologies is that they are iterative…that they should follow a waterfall pattern. This can be catastrophic to the improvement goal. These methods are typically shown as a one-way pattern with specific tools used in each phase. This method of teaching is necessary so that a practitioner can understand the methods and the science behind them; however, strict adherence here will take away from the art of applying DMAIC or DMADV to solve problems.
7. How do you manage determining an Alpha versus a Beta error?
Alpha and Beta (or type 1 and type 2) errors are necessary components of hypothesis testing. These are best thought of in terms of risk. In theory one could incorrectly reject the null hypothesis (alpha risk) when the null hypothesis is actually true. Alternatively one could fail to reject the null (beta risk) when the alternative hypothesis is actually true. The key to managing for this risk is to first determine what an acceptable risk is for each.
After determining an acceptable risk the next step will be to produce repeatable tests. Ultimately, when dealing with samples, you may never truly know if you’ve produced an Alpha or Beta error, but sound application of the scientific method will help eliminate unnecessary risks.
New practitioners will often seek to increase the sample size unnecessarily. Increasing the sample size beyond what is needed for your acceptable risk is rarely beneficial, and oftentimes it can introduce unnecessary risks. In theory, however, one could eliminate all alpha and beta risk by testing a population. The problem is that a population is always backward looking and always limited to what has already been measured. This means that measures of the population will ONLY be applicable to the population that was included in the measurement and will yield limited value.
8. In what type of scenarios would you use an RPN?
An RPN (Risk Priority Number) should be used in any quality improvement project. It is typically used as part of an FMEA (Failure Modes and Effects Analysis). The RPN is calculated by multiplying the Severity, Occurrence, and Detectability of an outcome.
I would recommend using an FMEA and calculating the RPN on every continuous improvement project. This tool is wonderful for uncovering risks and developing methods to overcome or reduce risks.
9. What do you use to know what to focus on first?
Assuming that question 1 (above) has been followed and we are now focusing on individual projects, I use the FMEA and RPN as described in question 8 (above). High RPN scores should be focused on first to best improve the process.
10. What metrics or strategies do you use to convince management continuous improvement is needed when the process is in-control?
A process that is in control can still be well outside of specification. This is best imagined from the manufacturing world. Imagine a machine that is designed to produce washers. This machine may be reliably producing washers that are 10 centimeters in diameter. Imagine further that this machine is producing the washers within one nano-meter greater than or less than 10 centimeters 99.9999999% of the time. This machine is in control; however if the specification is to produce washers that are 9 centimeters in diameter then the machine is producing a defect nearly 100% of the time. The goal of continuous improvement in this example would be to remain both in control and in specification.