You finally come to the last phase of the DMAIC process which is the controlling phase. This is one of those areas where pure lean people have some divergence with pure sigma folks. WHile I certainly see the value of standardization, you have to make sure that there is still some ability for a vision towards further continuous improvement. I have known a lot of sigma commandos that feel they have managed to mathematically disinclude any further improvements. In fact, they have invested in a strategy that is destined to do so. If you truly are going to implement a solution that gets your error and quality down to 6 sigma, how can you possibly improve any more? At least that is what they tought you in your final week of six sigma academy. It is like a self-fulfilling prophesy. You can’t get any better because there is not better.
Balderdash. Technology will change, environmental concerns will change, even people will change. I remember the first time I was given a laptop, It had a black background and some white letters. Pictures and links? Yeah right. Yet today, here I am writing on a blog that can reach an untold amount of people and I can dang near put anything I want into it. I may have had the perfect back and white television but once the kids found out you could get color, it was all over but the shouting. Heck, I am a dinosaur compared to what the kids are able to do these days. I just hope I make it to retirement before the next great technological leap hits us. But at the speed the world is turning, that probably won’t happen.
The control phase should end up with enough structure to allow you to measure the success of the intervention. Charts and data must be updated and frankly a system must be put in place to monitor the data for a preset period of time. What metrics will be used? WHo will monitor? What actions will be taken in the event that your data indicates the plan is not working? Unless you plan on being in a permanent mode of DMAIC, you also need a plan on who to turn it all over to. This will require story boards, documentation, file transfers and so on. One of the interesting things with modern technology is the ability to create team places on line where people can work. Unfortunately, when you go to transfer the ownership, you may need to rearrange or copy the file to a place where the operating teams are able to use it. This can complicate things so make sure you have a good plan for that time frame.
Finally, don’t forget the post mortem process. What went well, what did not? One of the greatest wastes in corporations is reliving the same problem over and over and not coming up with a common way to address it across business unit and area lines. I don’t want to rob any consultants of their chance to make a living, but imagine how much further a corporation could be in the innovation curve if they stopped wasting so much time, money and resources on solving the same problems a hundred times. Believe me when I tell you that the occurence of this waste in a global company is somewhat staggering.
As you can imagine, this was just a primer in some of the sig sigma methodology. Like lean, it can be deceptively simple. But in the proper hands, it can be dynamic and game changing. Check it out but get some help.
Back soon with some lean learning.
Sorry for the delay in getting to the most fun you will have in both Lean and Six Sigma. This is the part where you really get a chance to explore the results of your previous work. If you happen to be addressing a quality problem, look for the indicators which will validate your solutions. The outcome if its strictly a waste reduction or lean projection should be measurable KPIs and process improvement. In either case, you should end up with a revised stable process that is predictable and conforms to customer expectations.
The first step in achieving these goals os to use your completed tools from the analysis phase to develop potential solutions. I typically involve a number of team members in doing this since it involves some creative thinking. Creative thinking, as has been noted elsewhere, is important to the process since numbers alone will tell you the as is, but creativity can apply that status to a new synthesis. This is almost always improved with a group of people (unless of course your name is Einstein in which case, you wouldn’t need any help from a retired old Navy Chief Warrant Officer anyway). Assuming you are a typical person just seeking help, go nuts with this one. You don’t (and won’t) apply ever crazy idea that comes from your team but some of the best solutions come from some of the strangest ideas. That one idea you had that seems implausible may just contain enough of a spark that will ignite the real fire.
One other thing. In my experience over the past many years, using people who are actually involved with the work being improved is not only a good idea, it is the only idea that will give you credible results. The answer to why is fairly simple. That person or persons have been staring at that process waiting for management to fix it for years. They have probably rehearsed their speech with their wife and had lengthy discussions with their fellow workers over a beer or two about how simple the fix would be if management would just get off its collective ass and do the right thing. You may be the pilot, but they built the plane. The minute you think you are smarter than them, you have already failed. The only thing you are better at is making things more complicated. While that does bring a higher level solution from time to time, it does not guarantee that you will fix the right problem.
PDCA (plan do check act) is a great tool at this point. You need to evaluate the effectiveness of your ideas and that will require some experimentation. Up until this point, a Sigma way of thinking is the dominant process tool. From now until the end, you will find that Lean thinking benefits the team a bit more. Get folks involved in finding alternate paths and then use the other tools mentioned to validate.
As you move forward. make adjustments to your value stream map. Hopefully what you have been doing is actually eliminating non-value added steps and enhancing the value add where possible. I think that most projects will involve the first more than the second until you reach the point where you have cleared enough waste out to actually innovate. But I like to keep an open mind about that too. More than one time Tom on the engine line pointed out that a small adjustment would reap larger benefits on value add.
Speaking of Tom, do you know the people in your organization who add that value the most? Not everyone on the shop floor can claim that title other than as a participant. These are the real deal makers and wisdom of your organization. If you engage them and help them overcome the stupid processes we have stuck them with in the past, you will make quantum leaps in your improvement activity. But it’s up to you to find out who they are and get them engaged.
Now that you have done your homework, its time for an actual pilot project. Keep the parameters in check so you can create data which is rich with facts and not too out of scope. At this point, you must put systems thinking into its proper perspective (not completely ignoring it but keeping it contained). This will give you data that is helpful and useable. As quickly as possible, make sure your project delivers the results you anticipated and if so, execute the full-scale implementation plan you should have already been working on. (Come on now, you really need to keep up which means anticipating the next step as quickly as possible). If your pilot goes completely wrong, find out why and modify. PDCA again and again until you are either hopelessly deadlocked (meaning you had the wrong root cause) or until you are assigned to another project where you can do less harm. In either case, someone will be ready for the final phase of DMAIC which will be discussed next.
Just as the name implies, in this stage you are taking the collected data and trying to make sense of it. This gives the team the opportunity to identify and validate the key input and output variables which are related to the project goals.
When you finish this phase, you should have some good documentation of the potential causes that are part of your analysis. A couple of points in this phase. The team should have good communication about what is important and what is not. Plus, there needs to be an agreement that analysis paralysis is not allowed to creep in. I have seen more projects flounder in this stage than any other. It normally occurs because of the unnatural mix of cross functions that occurs. engineers and accountants by trade delight in spread sheets and tons of data. Support staff may want to drive to a quicker path to glory so they can get back to their real jobs. It is up to the Black Belts to maintain the flow of data and analysis.
Finding what adds value and what does not is another key part of this step. Validating Process Cycle Efficiency is also a key player and should be benchmarked if possible against what can be identified as world-class. Next you put together some theories that can possibly lead to root cause. Any number of the standard tools can be used during this part of the process including brain-storming, Cause and effect diagrams, and so on.
As soon as feasible, narrow it down to a few possible root causes. Then of course bring the actual root cause into focus with the substantiated data.
Team building, conflict management, leadership and root cause analysis skills are needed in this phase in order to be successful. If conflict becomes the key driver, it may spill over into areas that are not related to the problem. If that is the case, re-examine the charter and deal with it directly. Conflict that does not add to the creative process is nothing more than waste.
Some last-minute thoughts:
If you find low hanging fruit and the answer is intuitively obvious, assemble a quick Kaizen and execute. You may lessen the impact by gaining quick wins and in some cases hasten the search for the real root cause. While the steps are important to drive to a solution, keep common sense at your side at all times. As the little boy said when asked why he was so happy to shovel crap out the stall he replied , “Well, this proves there’s got to be a horse around here someplace!
This part of DMAIC is important for measuring and understanding the current state of the process you have selected to improve. During this phase, the collection of data which is reliable is done. That data would include things like quality stats, costs, process KPI’s and so on.
The end product in this phase is a current state value stream map which was discussed previously. This is a process to identify where your VA (Value Added) NNVA (Non Value Added but Necessary) and NVA (one of the 7+ wastes) exist within your current processes.
Collect the data, map the processes, validate the data (is it repeatable or predictable). Create a data collection plan using one of the data collection tools and finally assess the capability and performance of the process itself.
There are a number of steps and tools employed in this phase but their importance is critical to the DMIAC tool. Accurate data is the only way you can use the tool to achieve true improvements. In DMAIC, you don’t just assume or believe you have a problem, you have to statistically prove it with real facts.
There are normally three main areas of measurement:
Output: what do you expect as a result of the process you are mapping. This focuses on immediate results such as deliveries, complaints, or defects. Outcome also measures longer term results such as profit and customer satisfaction. In many cases these statistics are already available, but the team should validate them in any event.
Process: how is the product or service made? these are actual steps or events within the normal cycle that can be measured and tracked using any of the customary statistical or observational processes. This step can help the team to identify the actual causes of the problems.
Input: What are you putting into the system that will help to create the output. Bad inputs (supplier quality issues for instance) can lead to bad outputs so they need to be examined as well. On one of the product lines my team examined, the pumps were leaking after they left the shop. Our data revealed that internal to the shop they passed our existing test, It wasn’t until they went out into the field and were exposed to higher and more random pressure that an internal seal began to show signs of failure. The examination we conducted showed that a switch to a new supplier (to save costs) resulted in a product made in a workplace that did not have the same level of quality control. Microscopic traces of grit were discovered upon closer examination which would cause the seal to fail.
An important reminder at this stage is to know which data to collect, how much, and how often. The purpose of a lean six sigma event is to move quickly to solve a problem and create a continuous improvement solution. It is possible to get stalled out at this point by demanding too much data over too long a period of time. The trick is in understanding how much data is enough and that comes with practice as a green belt.
If you are accurate and your data is readily available, you can quickly develop an early understanding of what your early sigma is (the amount of errors you currently are experiencing. In any event, this phase will lead to the next step: Analyze.
One of the key tools for a Six Sigma project is the tool commonly called DMAIC
Define ♦ Measure ♦ Analyze ♦ Improve ♦ Control
In this section, we will only cover the Define phase.
The work to do a six sigma project starts before the team is assembled in many cases. The leadership (sponsors) and the Black Belts that will lead the project start to create the rough draft of the charter. This charter will help later to define the scope of the project and determine the need and types of resources. Having these in place can lead to an increased possibility of success early in the process.
Since defining the project is typically the first step (and you are cautioned not to skip steps) understanding what the process is and the outcome should be are important at this stage.
First of course, there should be a sponsor. This person should have a stake in the outcome, the ability to provide resources (whether directly or indirectly) and in most cases, enough understanding of the issues at hand to be a good judge on the component parts of the rest of the project. It would also help if this person was not a micro-manager but that is something that will need to be negotiated between the sponsor and the team.
Once the proper sponsor is aligned, its time to look at the team. Depending on the type of project, the team should have the appropriate cross-section to adequately analyze and identify problems and solutions that are within the scope. Team dynamics play a critical role in any projects but are magnified in a project where collaborative work needs to occur. That doesn’t mean you don’t put people together who normally have issues or concerns, it just means setting up the right framework for them to operate under. Conflict can actually improve the outcome if it is contained and nurtured by the leadership towards creative thinking. Frankly, a little conflict will guarantee that the group does not develop a mindset of group think.
Next, the combined group should re-look at the charter and refine its scope. Charters are a funny thing. When they are first started, they can be an accurate reflection of what the authors wrote, However, turn a group of non-aligned people loose on them and you will quickly see that the original thoughts worked well inside the heads of a few people but may not stand the test of a group of people. Re-defining is not a sign of weakness. Rather it is a healthy sign that tells the group that real progress can be made and real environmental impacts can be identified.
With the charter re-defined, the next natural step is to look at the scope. Does it still reflect the intent of the charter? Will it bring a need for different resources or time elements? What are the actual deliverables or at least the framework for them. An important part of this process is to find a way to hear and understand the Voice of the Customer (VOC). As simple as that may seem, many projects are launched with only the VOC of the owners of the process with no real intent for the betterment of the customers. Does the customer expect or need anything related to specific improvements in cost, quality, or delivery? What are the driving forces behind those needs? Are there changes in the market conditions that have made this an urgent project?
Since this is six sigma which is the reduction of the waste of errors, what is the financial impact? One of the key points in any six sigma project is to enlist the aid and support of the CFO or financial person such as controller at a local level. Their definition of success may have a completely different meaning at the end of the project if you are not aligned from the beginning. There are hard and soft gains that will be made by any efficient project. But if the balance is out of alignment and not enough hard savings can be identified, it can impact the way a project is perceived.
Another critical element to success is that you need to be very aggressive about getting all of this done. Lay out a clear plan of action and execution even at this stage. Many people who participate in projects like these will have other duties (like their real jobs). Anything that distracts them from your team’s efforts will result in delays and lost objectives. You can’t afford either in a project that will probably be expected to gain some quick wins. The Black Belts must be the driver for action and execution. Setting this example will make the team more effective, make your project a priority, and gain the support needed for the final outcomes.
Next up: Measure
Lean and Six Sigma Toolbox
As businesses and industries become more complex, the need for better ways to improve their processes and ultimately their products or services are the only true competitive advantage besides innovation in meeting the customers needs and demands. A lean journey is normally mapped out with three phases that start with an implementation phase. During this phase, change management and basic disciplines are built and developed. At some point in the maturity, the journey goes into a more circular phase that I call integration. Lean is no longer seen as something extra that is done, everything is done as a lean process. The next natural phase is using the learning and improved discipline to enhance innovation. With stable processes, a path for continuous improvement regularly being used, and waste brought into a manageable framework, the ability to innovate becomes a way to take the next leap forward.
The tools for lean and six sigma are often similar. Problem solving, standardization and so on play similar roles in both disciplines. But it is important to remember that there are often competing philosophies at play in the world of lean and sigma. Lean is based on the belief that it truly is a journey and those phases to maturity are often interdependent upon each other. Sigma on the other hand is based on data driven logic that is sometimes better used in situations where a more immediate effect is needed for a particular outcome. While both serve the customers needs, true change and continuous improvement is not achieved with tools or processes. True change is driven by our ability to have the people involved having complete ownership and embracing the changes as a path towards even more improvement. Just as PDCA is shown as a circle so is the lean journey.
Tools are also deceptively simple. The more you learn the more you realize how much you still need to learn. Tools in and of themselves are wonderful in skilled hands and when used as part of a system. Used alone and by amateurs will certainly produce the expected results. Reading about a tool makes you no more an expert than a person reading a medical journal becoming a surgeon. Yes they are important. Yes you should seek guidance from a craftsman before banging away at your problems. But if you succeed in that fashion, you will be the first I have seen do so in my nearly forty years of professional life.
One of the silliest descriptions I have ever seen on a chart measuring lean progress was the choice of the word “Perfection” as the highest state an oganization could achieve. The word itself assumes no further improvements can be made and that is absolutely ludicrous. You may have the perfect process of service anywhere for today but be assured that someone is sitting at a workbench or keyboard or in their kitchen saying “I can do this better if I …” That is the first step towards obsolescence of your perfection.
Last thought for the day: This journey can be hard and depending on the size and complexity of your organization, it can be rather long as well. Make sure before you undertake it you have a clear understanding of what it is you are signing up for. Get a good road map and make sure you have a guide who has been on a similar journey to help you navigate. I have a list of over 150 reason lean implementations fails. The list of why it can be successful is a mirror and reverse image. High on the list in any event is leadership commitment. It is one of the pillars of success and failure in any endeavor but I think more so in a lean journey.