Video není dostupné.
Omlouváme se.
ONE THING you are doing WRONG when Improving a Process | fkiQuality Lean Six Sigma Green Belt 2.0®
Vložit
- čas přidán 14. 08. 2024
- Improvement requires stability. Eliminate all causes of instability first through PDCA / PDSA / Kaizen efforts. Then improve the operation through fishbone analysis, process improvement, design of experiments.
Timestamps:
0:35 - Road to Improvement
1:00 - DO NOT IMPROVE an unstable process
1:08 - Why this order?
1:31 - How to tell if a process is stable (Control Chart Explanation)
1:57 - I-mR & x-mR Chart Explanation Slide
2:10 - Control Chart Example
2:23 - UCL & LCL Explanation Slide
2:44 - Range of Natural Variation
3:19 - What the Control Chart shows us Summary Slide
3:33 - Unpredictable process Explanation & Slide
4:24 - Step 1 to control Process: Remove Special Causes
6:09 - Is the process capable?
6:28 - Upper & Lower Specification Limits
7:33 - Upper & Lower Specification Limits Summary Slide
7:48 - Step 2 to control Process: Improve Process Performance
8:32 - Step 2 Explanation & the Road to Improvement
9:45 - How to begin the Road to Improvement (PDSA & Fishbone Diagrams)
11:23 - Road to Improvement Summary
Thanks for showing that stability is a necessary first step. Not only that, but it's possible to make processes stable by studying them and seeing how they became unstable. This isn't something that just happens but it's something that is worked on.
This shows the importance of addressing critical failures in your process before moving on to incremental improvement.
Excellent "play by play" of how an organization or team can get from an initially unpredictable/not in control process to one that becomes predictable and measurable (with only common causes of variation) to finally a process that is capable in a sustained way and that is free of any surprises. To achieve this progression the tools of Kaizen and PDSA move the process from unpredictable to predictable, and then by utilizing the fishbone and process redesign achieving the coveted step of having a sustained capable process.
Extremely logical--process stability before process improvement, yet in our rush to fix a wrong, the trial and error method still prevails. PDSA is essential and a reminder to slow down/think first.
It is so important to understand you have to be stable before you can make meaningful and measurable improvements. I wish the leadership at my workplace understood this concept!
I appreciate the knowledge behind how to find anomalies and how they are a sign of an uncontrolled process. Where as you can have an 'outlier' but still be within predictable variability of the process. I think that's an important distinction to make before you freak out.
First from unstable process to stable process and then to fit the stable process as per the boundary laid down by the Voice of customer(s) . Very clear teaching . Thank you very much .
This lesson shows how it all can come together to problem solve. This is a key video in my understanding of the process. Thanks! -Rob S.
Seeing the three stages plotted out next to eachother was great for conceptualizing the whole system from the start.
Really cool how everything ties in! It sounds obvious but I like the "do not improve something that is unstable". You need to take a more holistic view and understand the root causes before diving into improvement. I feel like this video perfectly summed up everything we've learned. Thank you!
Stability, then improvements. Also nice to see the problem solving added between charts to show when to use the proper tools for evaluating the processes.
Great clarification on the difference between UCL/LCL and USL/LSL in that the specifications are set by the customer and have nothing to do with the process itself. I also liked the visual going from an unpredictable process due to a special cause to a stable and predictable process and then finally to the capable and sustainable process showing which types of methods to use (PDSA and Fishbone diagram).
I liked that the video included tools to assist with each step. The visuals also offered a good representation.
Great step by step summary of how to determine if a process is stable and then determine is it working and all of the tools we can use to make the changes needed to achieve our goals
Stability and Continuous Improvement! Hands Down
Im always shocked by how simple it is to see a process that isn't in control but yet so many companies that I worked at don't bother to look at something so simple as a chart which can tell them so much information.
Makes sense that the control chart would show if the system is in control. I agree that the system needs stability or a firm foundation before attempting improvements.
I like how you compare to building a structure, make your foundation stable first then you can build solid improvements.
It's interesting to see how much information we can truly extract by charting and how we can use this information to see the results that projects are having and lead us in the right direction.
A great video about the steps of process improvement and how important it is to stabilize the process first before trying to improve it. Also liked how you gave some examples of techniques that could be used to both stabilize and improve the process.
Really good video demonstrating the goal and making customers happy
Stabilizing and improving the process. Charts are helping to see if the process is working- to see if you need to keep improving
Road map to success and improvement. Build the foundation to build from, use I-MR, X-MR charts and find the value and identify the range of natural motion.
Specifying the difference between UCL/LCL (voice of the "process") and USL/LSL (voice of the "customer") is helpful towards understanding what a control chart is communicating. Since USL/LSL is most likely immovable, achieving a process that is both in control and capable requires improvement to processes.
Ted, the first responsibility of a process owner is to keep the process predictable, which is another way to say, in control. If a process is not in a state of control (any of the Western Electric rules is triggered) then it cannot be reliably trusted to produce results that will meet any kind of specification. If it's happening is because we are lucky or because we are reworking half of what we are producing!
Once the process is in control, then the process owner (or the #Greenbelt) will seek to move the process #average where it is required - to the midpoint between specification - and reduce process #variation to make the spread of values narrow enough so that it doesn't spill over the specs.
Practice makes perfect. 2 ways to put a process in motion and then make imporvements as needed to be consisitant
My understanding is the control chart helps to predict if a process is stable and predictable. There are many causes that can effect an operation but it is up to management to review the data using the control chart to determine what process needs to be corrected.
I agree with the position that all organizations should use x-mR charts, even if they think their processes are stable. Charting the data allows you to see the trends and link them to special causes and common causes with the goal of eliminating the special causes and controlling the common causes reducing the variability within a process.
Good point Sherry. The x-mR chart must be used especially when processes are not stable -- how could you tell otherwise? Rest assured that the math and empirical experience behind the x-mR chart make it suitable for use with non-perfect or non-normal data.
road to improvement - instable process -> stable process -> improved process.
It is important to stabilize the process before improving it. If you improve without stabilizing, then you just improved that particular occurrence, but not the following one and the one after is still unpredictable.
Lack of productivity, to a predictable process on to a capable process!
Charting ranges shows how to help with process and staying in the limits.
The charts help to create a common truth as to how the process is performing and its level of stability. That is the basis for brainstorming of how to improve the process to be in line with the Voice of the Customer. Staff generally take pride in a well-run process, but if it isn't working for the customer then it has little value.
Don, this is correct; a well-run process must also be customer-satisfying.
The trick is that because of a mis-understood #customer focus, we tend to measure our processes only against the #specification, creating what #Wheeler calls a binary view of #quality that switches between a "benign neglect" when we are inside the spec and a "crisis mode" when we are outside.
Unfortunately, measuring against specs does not help find out the reasons why we are outside specs! This is, because the specs have nothing to do with the behavior of the process -- they are just the voice of the customer separated from the voice of the process. Specs do not provide useful information to fix the process.
The #control #chart, however, provides a way to understand what is happening with the #process, and very important, it also gives us an "early alert" that the process may drift outside specs through the use of the Western Electric rules.
Identify, stabilize, and then improve.. you’re ultimate goal is to improve processes for a better outcome, but you must first find the need for change, make a stable process, and then continues improvement from there.
Just because process is in control does not mean it is predictable. Once a new process is implemented it is important to stabilize the process to create the new standard
I see it too often that teams try to make improvements without any planning or stabilization. This will be a great takeaway for my teams' focus.
This is a good overview. Trying to adapt this approach to my workplace (engineering services) is going to take some thought since our products don't necessarily fit a "standard" mold.
This chart in this video allows people keep track of what field they are doing good in and what other area they will need to improve in.
Very informative.
two major steps: stability and how can we continue improving
Interesting how changes made to an unstable process are steps to stability not improvement. It makes more sense to me now how improvements can only be achieved on a process that has record-able consistency.
thx this was really helpful
Its interesting to think about this in current business settings, am I rewarding changes to an unreliable process that get more stable or try to improve towards the spec limits but remain unreliable.
Joseph Tshulos, as you make changes to process you need to make sure that the results you get are going in a positive way.
I would be interested to see on average how many processes that are in control listen or actively work with the employees. It seems that the closer to the VOE and VOP you get, the more in sync you are.
Need to eliminate instability in processes to have improvement that can be sustained.
eliminate all causes of instability and improve the operation.
Remove special causes of variation and improve the process.
Our processes remain predictable as long as we are in sync with the VOC. We have to be willing to change with customer standards.
Präzise und systematische Beschreibung.
Eliminate all causes of instability = improvements. We should be able to predict our success
How are the UCL and LCL decided? By policy?
Loren R Elks, not at all, control limits are calculated from the data produced by the process that you are monitoring. Control limits form a range within which the vast majority of results will be found when the process is stable, that is, they represent the voice of the processs.
So, when a value falls outside the LCL-UCL band, you must find out what happened, because it's very unlikely that it happened "by chance."