Too Many Simple Mistakes

Six Sigma – iSixSigma Forums General Forums Implementation Too Many Simple Mistakes

This topic contains 14 replies, has 9 voices, and was last updated by  Mike Carnell 8 months, 2 weeks ago.

Viewing 15 posts - 1 through 15 (of 15 total)
  • Author
  • #55954


    Okay, Lean Practitioners and LSSBBs, what is the root cause of too many simple, basic mistakes. The parts we make are costly and fragile. So, we use an IATF-16949 quality management system. We’ve got training documents and training matrices. We’ve got supervisors doing job observations and layered audits. Still, people make a lot of fundamental mistakes like:

    * Mixing the wrong chemicals
    * Not calling maintenance when things are only semi-broken
    * Turning pumps on when hoses are not connected
    * Putting 8 parts in a box clearly made to hold 9 parts, even
    * Bypassing Poke-Yoke fail safes

    We are doing lots of things right, but clearly we are doing something fundamentally wrong. I suspect it has something to do with not being disciplined about following work instructions, and/or not enough oversight by managers and supervisors.

    Things I can test?
    Ideas I can go out and check?



    Robert Butler

    The things you have listed give the impression that management is not walking the talk – everything in your list are the sorts of things one sees when quality is just an empty slogan and the only real focus is on timed throughput.

    Have you actually sat down with each of the subsections of your process and built a real fishbone diagram that shows what is really happening? My experience has been that when issues like yours arise a good fishbone will show exactly why they are occurring. The cause of the problems will most likely have very little to do with the people actually doing the work and almost everything to do with the environment in which they are working.


    Chris Seider

    It sounds like you need more engagement of “operators” and their machines (devices). Are basics checked, equipment process conditions in control, etc?

    This is the power of autonomous maintenance with TPM but is one technique.



    Yes, thank you Robert Butler. We have done lots of fish bone diagrams for various CI projects, but I don’t think we have honestly done our due diligence on fixing the basics problems like “Why aren’t hoses connected every time?” These issues are so big that a fish bone, 5Y and Go-and-See are clearly called for.

    Chris, I like your idea of checking and see if the basic processes are in control. We can certainly do specific job observations and put a number on how well people are following established procedures. I agree that getting machine operators more involved in maintaining their machines should help prevent breakdowns. More Autonomous Maintenance is on the list for this year.

    Other suggestions are most welcome.



    Have you seriously done gemba to understand what’s really happening, in a non-threatening way? I’ve often seen when procedures exist people do otherwise, not because they aren’t trained or they’re careless but because the procedures fit an idealized environment that doesn’t actually exist.


    Chris Seider


    Neat idea to walk the floor and see how it’s performing. ….sorry, my tongue’s stuck in my cheek.


    Chris Seider

    @Straydog so we’re clear….we’re in total agreement.



    Strayer and Chris–I tasted, er, I mean TESTED the floor today, and it was still working well. :-)

    Actually, you are both correct. We have not done a good job at teaching folks how to write good work instructions, so, the are routinely ignored because they have accuracy issues, so then you have a program that isn’t a program. Without standards, there can be no improvement! So, maybe this comes back to not-great training and so, not-great discipline, and so lots of variation in methods.

    Mmmm. So, then I need to test better training and better supervision and see if improving training (mostly the work instructions and how we present them) and supervision (I’m thinking job observations or informal audits), actually improves the results. If so, then I can share our New Method.

    Good start, y’all. Thanks!


    Hi @ahiru-san.

    I work with some industries that show similar problems like you pointed. What We have been doing when we cannot get the problem narrowed is take a step back and using tools from Data Science like Cluster Analysis to try come up with problem or sintoms patterns and after that, go on with the analysis.

    In that case if you look at the problems as isolated facts you may not be able to identify its relationships but when you identify clues about them, find the root cause becomes easier.

    So, before study the isolated cause, seek the patterns!

    Best Regards



    Chris Walker

    I work in commercial roofing and see the same issue. The installers (and even office staff) know what to do and how to do it, but don’t always do things they way they have been trained and agree is best. I’m frustrated. I don’t know if they just DGAF, if they lack pride or incentive, and/or if there is a lack of accountability.

    This is my first experience where I physically cannot control the environment (machines, software, accountability, etc.) and eliminate the opportunity to do things against procedure. You’re not alone in this search!



    I have found that getting Management “In the Game” will increase the odds in your favor.
    To directly answer, “… clearly we are doing something fundamentally wrong. I suspect it has something to do with not being disciplined about following work instructions, and/or not enough oversight by managers and supervisors.” It might be a case of defining who ‘we” is, and the cause & effect of the lack of oversight. I think you’re seeing the effect.
    Maybe get a copy of “The Work of Management”, by Jim Lancaster,
    [ ], sorry about the ginormous hyperlink. Get your management folks to read it, then get them to DO it. If they care about what is happening in Gemba, they will start showing up there on a predictable basis (read: daily scheduled Mgt Walk About Reviews), asking tough (but fair) questions, and providing the level of help only they can deliver.

    Combine with Training Within Industries (TWI) methods, to address things from both sides of the spectrum. [1] If your management does ‘Management by Magic Wand’ (they wave their hand and The Magic Happens), you’ll forever be frustrated. [2] If employees don’t have good training (using Standard Work) they’ll find ways that work for them, but not always the best for the company, like the results you’re experiencing now.
    Hope this helps a little.



    Hi @ahiru-san

    The hard but powerful way to address some problems you mentioned might be proper process description (e.g. with SIPOC) and using FMEA for it. Team effort to assess the risks and discussion of all aspects (potential failures, root cause analysis and all detection and prevention controls) might be better than any training. Having FMEA as book of knowledge may help to prioritize focus areas for continuous improvement.

    Just as an idea.



    Your suggestion @ssobolev may be right on point. When work instructions aren’t followed it’s often because they’re idealized rather than practical, not due to poor training or carelessness. This can happen even if people who actually do the work are involved in writing them. An FMEA exercise, if done right, can get everyone to understand what isn’t working right, and why, and lead to practical process improvement.


    Marilyn Barber

    First, map every point in the process that has a potential for an error or omission. Make a list. Have it laminated and posted at those point locales. I would tie quality to performance. Require each employee and manager to OWN the process and report errors. Tie performance to a positive incentive as well, for example, for an employee with no errors during a month, provide a monetary incentive, and make their performance known to the entire team. Have each employee be actively involved in process improvement–also tied to performance. Require them to come up with at least two process improvements per month, and record/track these. Look at your tools, materials, training, and process.


    Mike Carnell

    @ahiru-san You had all the information you needed to fix this issue once you had responses from Robert Butler, Chris Seider and Strayer. Everything else is redundant.

    Your QMS isn’t going to solve problems like this. That isn’t what it is for. This is sustainability so if QMS were the solution then everyone that was Q1, ISO, Pentastar, etc certified would not have sustainability issues. This also has nothing to do with top management involvement. If your top management has to solve this stuff then they don’t need you.

    Lots of talk about standardized work. That is about half of it. The part of standardized work that gets skipped is going from documentation to implementation. That is training. People generally want to do what is right. If you have something documented and they aren’t following it you can ask them. Typically if they do something different there is a reason and they will normally tell you why. All that Gemba talk. Mostly people trying to sound very avant-garde. Visiting is pretty superficial. Sit down and try to do the job yourself and talk to the operator about what you believe are “simple mistakes.” That already sounds pretty judgmental.

    That gap between documentation and implementation is training. Here in the US we normally fill it with something fast and cheap. Fast food training what woman I met called “give, tell, dump, shove it to them” training. There is a reason we do not train our Olympic athletes on fast food. We are trying to optimize performance. You don’t get it on fast food and you don’t become an efficient organization of cheap and fast training.

Viewing 15 posts - 1 through 15 (of 15 total)

You must be logged in to reply to this topic.