Advice on the Focus of a Green Belt Project
Home › Forums › General Forums › New to Lean Six Sigma › Advice on the Focus of a Green Belt Project
This topic has 5 voices, contains 4 replies, and was last updated by 
Mike Carnell 264 days ago.
| Author | Posts |
|---|---|
| Author | Posts |
| August 19, 2012 at 10:41 am #184846 | |
 @  Reputation - 0 Rank - Aluminum
|
I am just starting to scope a green belt project. I want to reduce customer complaints. Errors occur anywhere in the system, from quoting thru shipping. We have two methods of processing orders A & B depending on customer requirements. Over the past 12 months we have processed 971 orders with 90 complaints. My goal is to reduce complaints by 20% from 90 (past 12 months) to 72 over the next 12 months. I want to focus on one of these processing methods. I am leaning toward working on method B even though it seems more robust than method A. I know it’s not much information but, any comments would be appreciated. Thanks |
| August 19, 2012 at 12:22 pm #184849 | |
 CFB @CFB Reputation - 0 Rank - Aluminum
|
Random questions: Is there a statistical difference between the defect rates of each method? Why not focus on defect type instead of method? Where is the business case in determining your project scope? Is there a critical customer or SKU involved? Are the two methods using different resources in different amounts or vice versa? Is the scope large enough to make the financially viable yet small enough to make it practically do-able? etc, etc, etc. Good luck. |
| August 20, 2012 at 11:28 am #184873 | |
 Chris Seider @cseider Reputation - 3008 Rank - Titanium
|
You could run the 2-sample proportion test but don’t fall into the trap of using the lesser of two evils. Find the categories of the errors, find the root causes, and develop a more robust method C. DMAIC GB projects are about using the methodology. Also, you want to use a monthly metric (not a yearly one) so you can get immediate feedback if things are improving. Shoot for at least a 50% reduction in total errors…you’ll probably find 1-3 causes which can be eliminated that account for 80+% of the errors. Good luck |
| August 24, 2012 at 5:32 pm #185021 | |
 Dan @captainamerica Reputation - 31 Rank - Aluminum
|
Honestly, I believe this to be much more simply solved. First things first, you have established a failure rate, which is the first step, but your next step is inherent to MSA (Measurement System Analysis); WHAT are the specific failure modes, or in this case, root causes of customer complaints? In this case, you’d want to use a PARETO CHART to determine the fundamental reasons by which said complaints are occurring. Once you have a full understanding of the failure modes, you can then exact the DMAIC methodology, or even a Kaizen Blitz, to eliminate the most recurrent of the issues. Don’t let the confines of DMAIC wrap you up. While it seems as though you are well within the ‘D’ phase of your efforts, it strikes me as though your precise root cause has yet to be understood. |
| August 30, 2012 at 1:15 am #185124 | |
 Philip Godkin @pgodkin Reputation - 27 Rank - Aluminum
|
Just a point from a newbie, would it be benifical to measure each dept on it’s own ie order process, picking, distaching, invoicing accounts recievable etc. again you could break this down by ordering method A & B At this point you could see which dept you need to focus on more clearly, I have a hunch that we could use the old chestnut that 80% of complaints is caused by 20% of the processes |
| August 30, 2012 at 10:33 am #185139 | |
|
Mike Carnell @Mike-Carnell Reputation - 3168 Rank - Titanium
|
No idea who to send this to so no idea if you really care about the answer. Filling orders does not sound like a particularly complex problem and 10% sounds like a pretty anemic goal. Do them both. Who knows, just maybe the processes are very similar and you can figure out what is wrong with one product by seeing what they are doing on the other line. If you insist on turning this selection process into a science project and you need to cut out 18 defects and Method A only had 22 total, you might want to take Method B. It is probably going to be much easier to improve it some rather than making Method A perfect. I really hope your project is not defined in terms of you taking out 18 defects. If your number of orders processed drops significantly you are screwed. You need to set a goal for each line and it probably needs to be in terms of percent defective. Just my opinion. |
Find us around the web
© Copyright iSixSigma 2000-2013. User Agreement. Any reproduction or other use of content without the express written consent of iSixSigma is prohibited. More »