Problems in problem prevention

Jinjer wrote in comments under Nameless #6 -Troubleshooting:

I just read The Machine that Changed the World and it’s fun to see you applying some lean logic to your posts. But a question that’s been bugging me while I read the book bugs me even more now. If you don’t run a factory, but subcontract, how can you expect the sewers to be looking for mistakes and trying to make improvements?

Unfortunately, failing to correct mistakes and failing to make improvements is not limited to subcontractors; it’s more pervasive than that. In fact, the dynamics leading to failures in problem prevention are already present in nearly everyone’s organization, subcontractor or not. The underlying problem is probably the biggest problem in any manufacturing facility, anywhere and in any industry. Still, I’m not a generalist and I’ll explain the depth and issues of this as applied to sewn products.

However unlikely the analogy, I found some great reference material in which to frame the question more appropriately in an article about nurses in Why Your Organization Isn’t Learning All It Should by Tucker, Edmondson and Spear [courtesy of Harvard Business School -get a free subscription here] (all excerpted material is from the latter). I think the biggest problem is framing the question properly. Toward that end, the authors make an important distinction in that they describe “first order processing” as fixing problems in the immediate (a work around) and “second order processing” as “diagnosing and altering root causes to prevent recurrence”.

First-order problem solving allows work to continue but does nothing to prevent a similar problem from occurring. Workers exhibit first-order problem solving when they do not expend any more energy on a problem after obtaining the missing input needed to complete a task. Second-order problem solving, in contrast, investigates and seeks to change underlying causes of a problem.

Yet according to their research, nurses -in spite of being relatively empowered and comparatively powerful when compared to sewing operators and better educated too- are no more likely to engage in second order processing than are sewing operators. In sewing factories, failing to report problems is endemic. In my experience, it’s more likely that a stitcher will say nothing no matter how much you beg them to tell you of problems and in an outside shop where you have little if any control, I can only imagine the response is worse.

For our purposes, here are some barriers that must be addressed in order for stitchers to report problems (assuming the stitchers are valued and encouraged to report problems):

Let’s pretend you’re merrily stitching along and you find that the back lining pivot point is off 1/8″. Well, you’ve got an entire stack of these to sew up and you know that while the problem is significant, there is no way that anybody can immediately provide you with a stack of correct back linings. Rather, the error represents a significant loss to the company if you cannot devise some sort of work-around to complete the seam. You know it’s not an imperative to the extent that the company will reorder lining fabrics, correct the pattern, re-cut the goods and then provide you with a compliant stack of goods in the matter of time that you need them. Your work-around solution is the equivalent of first order processing. Your immediate alternatives are to either devise a work-around or to report the problem (leads to second order processing). Let’s assume you’ve decided you’ll do no more shoddy work arounds (first order processing) and follow the simple mechanics of attempting second order processing to improve the product. The barriers to reporting your problem are these:

1. Obviously you have to stop sewing. However, since you’re paid by the piece, anything you do here on out, is unpaid work. Unpaid work is illegal and against company regulations (and it is). Therefore you must clock-out for piece work and clock in for hourly.

2. To clock out and clock in, you must first stand up. Immediately -before you’ve had time to leave the machine- you’ve attracted the attention of not just your supervisor but every other stitcher in the sewing line. Everyone is watching you. It is highly unlikely you’ll ever make it over to the time clock (on the other side of the plant) before the supervisor catches up with you which means you’ll have to explain to your supervisor why you’re punching out. It is similarly unlikely your supervisor will “allow” you to proceed as your failure to complete your bundle will really gum up the works affecting every process following your own. In making the move to punch out, you’ve basically slowed the entire process to an eventual halt, meaning that every stitcher staring at you won’t be pleased that your actions will be limiting their income. Do that very often and nobody will want to eat lunch with you, not matter how valid your complaints are.

3. Let’s say you’ve successfully run the gauntlet of supervisors and peer approval to make it to the time clock and punch out. Congratulations! Your reward is a 50% paycut! Even in 1995, the average stitcher earned $9.73-$11.27 on piece rate and clocking out means you’re now earning minimum wage. Do that too frequently and you can eliminate frivolities such as daycare and utilities from your family budget.

Considering the above, one can understand why stitchers will never report problems. Rather -like nurses- stitchers will devise a work around. Tucker, Edmondson and Spear state:

…we discerned a pattern of first-order problem solving that characterized the majority (92%) of nurses’ responses when confronted with these obstacles. The pattern was comprised of two heuristics, or rules of thumb, that were embedded in the work system, and can be seen as guiding-either alone or in combination-nurses’ problem responses for all but ten observed problem events. When nurses responded to problems using Heuristic #1, “Do what it takes to continue the care of the patient,” their behavior was characterized by concern for securing the information or material they need to do their jobs and not on understanding what caused the problem to occur. After the nurses were able to resume caring for the patient, they did not expend any further effort on the problem, including communicating that it occurred…

Tomorrow I’ll discuss further barriers to second order processing (eliminating the root causes of problems) which is one of the best ways I can think of -bar none- for a stitcher to get fired, complete with real-life examples. And, I also tell you about a friend of mine and how he ran his plant. He didn’t have these problems. If I do my job half as well as I intend to, maybe you’ll end up more like him.

In the meantime, read the original article. The conclusions there are not to be missed. Lastly, I can’t recall how I discovered this article; it was probably a link through Curious Cat, Lean Manufacturing Blog or Panta Rei.

Get New Posts by Email


  1. Good that people are reading the Machine that Changed the World. It had a profound impact. But to keep learning, keep reading. “Lean Thinking” by Womack and Jones, authors of “Machine,” is very worthwhile. I just read “The Toyota Way” by Jeff Liker. Even if you have to generalize to get past the one automotive company focus, you will learn immensely.

    As for your subcontractor– they are your supplier. Toyota has had a strategy of supplier development for decades. They say, “If you want us to buy from you, you have to improve, and we will help you.” If the supplier doesn’t, they lose the business. Few companies have Toyota’s leverage, unfortunately. Anyway, the reason your subcontractor’s sewers don’t pay attention to quality revolves around practices that Kathleen mentions, but another way of looking at it is that the company’s culture and management are way off what you want. If you could identify a subcontractor practicing lean, in its correct sense, you could give them your business. And that success should motivate them to keep getting better and better, not to mention treating you — their customer — as the most important person in the world. But more importantly, you could start rating such companies and publishing your ratings, thus promising the learning company more business — which they will want if they are lean because they will have the productivity and capacity to take it on. The bad companies, in theory, will become dinosaurs and die off — or learn.

    Do you know of any outstanding companies in the apparel industry?

  2. Toronto Fashion Incubator

    I like the Toronto Fashion Incubator even though they ignore me. You’d think I’d rate a link but no such luck. Still, if I only wrote about people that gave me the time of day, you’d have very little to…

  3. Rocio says:


    Great article!… I have suspected for a long time that the part human nature plays in manufacturing is not so obvious to most people, and finally this article brings it to light.

    The last couple of weeks I’ve seen first hand the effect that the economy has had on factories that didn’t adapt with the times….
    The building we’re in had 4 companies (1 closed down 2 weeks ago, 1 decided to sneak out over the weekend, and my neighbour is planning on closing next week) and now it seems like we are the only ones left standing…. The main difference is that our company culture is completely result driven.
    The hardest part of bringing employees into a culture that is so different from their previous employers is that it takes a while for them to trust that things won’t change as we grow…
    Once they overcome their fears, the results speak for themselves.

    Saying that, it has been a very steep learning curve (as in human behaviour, and psychological learning) and I know that there will always be room for improvement

  4. nosaj says:

    @karen – toyota has the incentive to help breed its suppliers because its product lifetime is a lot longer and they depend less on marketing than actual build of product. in the apparel industry, seasons come and go, new lines are created 3-7 times depending on the market. also, toyota’s subcontractors manufacture parts that are proprietary, complex and r&d intensive. such is not true for the majority of the apparel industry. i can say that nike, despite all its labor woes, does invest in its subcontractors, however this is partly due to potential losses from piracy that can occur if nike does not provide better incentives to the subcontractors. lastly, the difference in potential damage. i believe you remember the ford/firestone fiasco complete with exploding tires. what happens if the stitching on my northface jacket is off by 1/4 inch? and what’s the cost for them to repair the situation?

    so what i’m saying is that unless the brand is focused on producing the BEST QUALITY clothes, as toyota is focused on the BEST QUALITY cars, then the toyota way would not apply. lastly, from porter’s 5 forces, the cost of changing suppliers is extremely low in the apparel industry.

  5. Judy Gross says:

    Another reason why a sewing contractor will not correct mistakes is because it’s not always in their best interest. I’m taking on a contract to make rain jackets, until now, they have been sewn by another company, the finished product looked quite shoddy. When I was given the pattern by the manufacture, I found it to be full of mistakes – I have corrected the pattern and changed the method of sewing. This has cut down the construction time by half. Since the time it takes me to produce the jacket now is only ‘X’ and the previous company took ‘XX’ time, (having to work around all the little mistakes in the pattern) they made more money on the job.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.