Advice for Managing Large-Scale Project

Six Sigma – iSixSigma Forums General Forums Implementation Advice for Managing Large-Scale Project

This topic contains 3 replies, has 2 voices, and was last updated by  Aaron Olson 4 months, 2 weeks ago.

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #56058

    Daniel Sims


    I’m about to begin a new project (that I suggested) but it’s admittedly rather large in scope. Perhaps too large to be effective. I’ll give a brief bit of background info:

    Project that is just finishing looked at increasing the output of an automated stitching cell. Metal pins in, plastic mouldings in, components out. Many process/mechanical improvements were made and there is still too much variation going in. The company is happy with the output in terms of volume, but we are to continue seeking scrap reductions. I have made the case (with common sense as I see it) that we can’t reduce this variation and scrap AT the process, we have to go upstream. We have all capabilities on site, so raw material comes in and finished components go out. As you can imagine we have numerous value added steps in various departments in between.

    I have pitched that to really improve this (and all) processes, we should start at the source and work our way down. This is obviously a bit washy and not defined enough to really do anything, so the first thing I’m going to spend the time on is a detailed process map from warehouse in to warehouse out. I have a quality engineer green belt who will be working with me and he’ll focus on collecting the value stream as we go. I think what I should start looking at in each area is 1) What are the procedures in the department 2) What is the department doing 3) Are the procedures still relevant or can they be improved. Once these questions are answered it will be a case of bridging the gap between what is happening and what should be happening. I foresee some issues arising when quantifying improvements. For example to prove out improvements I’ll be monitoring performance at the end of the chain quite closely, but there are obviously many factors and noise between the start and the end.

    This is as far as I’m allowing myself to think ahead at this time. I have a MBB coming on site in a couple weeks to spend a day with me and get things clear in my mind, but I thought I’d draw upon the experience here in the meantime. Some of my initial questions: Do you have any advice on tackling a project with this scope? Is it best to break up an initiative like this into smaller projects, for example ‘Goods in project’, ‘Metal stamping project’, ‘Inspection area project’? I don’t have many green belts to draw upon really, so I’ll be taking one along for the whole initiative and drawing upon some others at each stage, any advice on a team for a project like this?

    Sorry if it’s a bit vague, or if some of it is obvious. I know in my mind this is the obvious route to a real improvement but I’m finding the scale of it a little daunting.



    Aaron Olson


    your scope does, in fact, seem quite large. Not that it won’t be worth it but it will be daunting. Your thinking is right but make sure that your evidence/maps/charts/etc. actually take you up that river.

    My SUGGESTION is that you pick a single final product or final product family that your are seeing unacceptable levels of scrap. Take that product and map the process. Find a way to quantify defects and only follow up the stream if the map takes you there. If you can’t statistically prove that defects in final products are directly due to defects in upstream processes, don’t go there. When you fix a product/product line, take that knowledge and then spread it across the floor.

    My assumption may be wrong but you don’t seem to be presenting evidence as to where the source of defects actually originates. Don’t go on the “witch hunt” if there isn’t anything there.

    Just my opinion. If you’ve already proven where the defects originate, then please disregard.


    Daniel Sims

    Hi Aaron,

    I didn’t want there to be humungous wall of text in the original post but your thinking did align with my own. The previous project has highlighted the variation is coming from one of two places, the metal stamping area or the metal plating area. These are sequential processes, data collected so far suggests the stamping is the root of many problems initially. To gauge change i’ll be focussing on one product, which was the focus of my just finished project, as I have a deep understanding of it and what to expect. There are internal politics to consider also, so to keep people onside I realistically have to start at the warehouse in, which is fine as it’s literally the process step before stamping. It will allow me to review the processes in place and ensure material is handled correctly before being released into the stream, hopefully ruling them out.

    We hold monthly reviews on scrap and over the last few months by carrying out root cause analysis on many defects we have three independent analyses all come to the conclusion that the stamping/stamping inspection is the root of all evil so to speak.

    But to return to the wide scope this points me in a direction, but in terms of a six sigma project it is admittedly vague at the moment. I expect (and am hoping) by injecting improvements into the start of the chain defects upstream in the middle of the chain will decrease naturally, thereby the project should get easier as time goes on. I just don’t want to wade into an ocean of problems without direction and get lost I guess.

    Thanks for the reply


    Aaron Olson


    That makes sense now.

    If you don’t mind the asking:

    -What is your current scrap rate vs your target scrap rate?

    -How significant of an improvement are you seeking (%)?

    I would limit the scope to the stamping area with ‘Y’s being measured at the end of production after stamping and after final assembly. With your last project being successful, I think it’s fair to keep that process out of the scope. You can probably keep the ‘warehouse in’ in your scope. It would be fair to say that it is possible that there is special cause variation that could, in fact, be due to raw material or handling. [we have a part that we make that often gets rejected due to cosmetic issues – the kicker is that the customer purchased raw material from a vendor that does not properly handle with any thought towards cosmetics; due to that we have additional machining, labor, inspection, etc. involved.]

    Politics can be so damaging to a project. Is it that they don’t want to admit that a certain process may be under-performing? I’d tell them: “This isn’t a witch hunt. The process doesn’t reflect on the individual. Nobody wins if everyone ‘looks good’ but garbage is still being produced. If it’s not the actual problem, we’ll move on; if nothing else, we can verify what is being done well.”

    If it is a “all options need to be considered”, the counter is that they were considered and they pointed to a potential source of defects.

    I’m very intrigued by the project though. Especially since it was identified through your previous project. Hopefully my opinions are at least mildly helpful.

Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.