The evils of mass production 2

I was by turns, pleased, gratified and yes, even dismayed by some of the comments from the first entry. Overwhelmingly though, I’m left with three conclusions:

  1. Cognitive dissonance
  2. Application problems and limitations
  3. What constitutes sub assemblies and where is batching truly unavoidable?

Cognitive dissonance
Let’s be fair. Some criticisms of Ron’s video were a bit over the top, reducing our credibility in the community. As I’ve said before:

According to Aronson (1996), when people are confronted with opposing beliefs or ones incompatible with their own, they are likely to ignore or negate that belief. They do this in order to convince themselves that they have not behaved foolishly by committing to false beliefs. To assure themselves that they have been wise in supporting their position, they often convince themselves that those who oppose that position are foolish and truly objects for contempt and derision (Aronson, 1996 p.184-8).

I think Ron has accomplished two things. One, many DEs already think they are lean because they’re operating on a shoe-string. Lean budget doesn’t mean lean manufacturing and judging from the comments, lean is no longer an abstract. That’s a good thing. Dissent also means you know you’re not. And that’s not a criticism but a goal. Mike (one of our few truly lean DEs) gave me plenty of dissent too.

Application problems and limitations
If one understands that lean is no longer an abstraction, one can either look for small fruitful ways to move forward or look for self-validating reasons for not being there (yet). Not being lean is not a failure. Saying you’re doomed to tread water, is. You don’t get there in a day, it’s a process. Just like you, Mike and Amy started sewing out of a spare bedroom. They had limited space and funds. They didn’t get lean by saying they had limited equipment and space. They had to look for ways to get there. Heck, most of you had a leg up on them. At least many of you have been sewing for years. Amy hadn’t. Everybody has their own obstacles. Real creativity means finding ways around them. Personally, I think that what most people think is creativity, is overrated (rant omitted). This sort of creativity isn’t.

Marguerite brought up some very valid points, the tangible reality for most everyone starting out. She mentioned the necessity of batching due to space and equipment limitations. Eric thought that was an opportunity for SMED (rapid changeover) but I disagreed. Changing needles and thread is such a hassle. It breaks your concentration and flow of operations. You can’t sew with any fluidity if you’re constantly starting and stopping. The best one could hope for was “batching” according to colorways and similarity of operations (which I’m certain Marguerite is already doing, she’s not an idiot). Still, creativity can be helpful. For example, how can one get buttonholes? I can tell you that I’d job it out to someone who did. I’ve already looked into it. I have a button holer but I don’t like it. Ultimately, the sooner you can add machines, the better it will be. It’s funny…most (batching) manufacturers think of it in terms of one machine to an operator. With Lean, it’s one operator per many machines, meaning more machines than operators. Sure, machines aren’t free but they aren’t an ongoing payroll expense either.

Her comments led to a discussion twixt Eric and I. Eric mentioned that she had some points about pressing. That when he made that vest, we pressed each seam as we went along. He thought that logically, we’d need several pressing stations in a cell (if we were to produce these vests). But that was a fallacy; his orientation to the process, led him to this “logical” conclusion. I told him no, we didn’t need to press each seam, just that we did it because he hadn’t learned seam handling skills and we didn’t have the one piece of equipment we needed, a blow press. “Blowing a jacket” means to put it on one of those body shaped steam presses. Having one of those would largely eliminate interim pressing. Using a hand iron after the fact wouldn’t work because lining seams could raise imprints topside. Being that we don’t have one, we have to design a cell with an ironing station centrally located with respect to the rest of the work area. Sill, is pressing really needed? I rarely press when I’m sewing. In industry, the only thing that’s pressed interim, are turned collars and cuffs or anything that takes a paper or metal jig (assuming one lacks automated equipment). But then the discussion of collars and cuffs almost begs the discussion of subassemblies (later).

On the other hand, I told Eric that what pressing there is, might need to be batched to keep expenses low. Keeping an iron going all day for occasional use in a limited cell, seems too wasteful. In such case, batching pressing is more mindful of energy consumption. Lean means lower costs so how can it be lean if it costs more? There’s any number of ways to look at the problem. One of us also mentioned a custom solution of smaller pressing stations, designed to the size of the work piece but this would still involve multiple irons being left on all day. In sum, no one suggests the solutions are simple. If they were, everyone would be a lean manufacturer.

What constitutes sub assemblies and where is batching truly unavoidable?
I truly do not know the answer to this question. How do we define a subassembly? I’m holding out for chest pads and linings as subassemblies, in part because they’re more interchangeable than shell/self pieces. I also think fusing could be a batching subassembly. Believe me, I’ve had plenty of spirited discussions with other lean adherents regarding subassemblies and the difficulty of avoiding batching in apparel production.

Using Ron’s video as an example, (he’s posted a follow up entry) what if he were assembling a multi-part mailer that consisted of four pieces of paper stapled together, a business card, a brochure, and a CD? I think a business card, brochure and CD would all be subassemblies regardless of whence they came. But how about those four sheets of paper stapled together? Should those be collected and stapled as a subassembly or should they be aggregated during the process of “manufacturing” and “packaging” the “product”?

Really, I’d love to get feedback on what you think defines subassemblies and where you think batching is unavoidable. I don’t mean unavoidable in your particular situation, I mean over-all, assuming you had every resource at your disposal. Textile production comes to mind -you can’t weave a yard of fabric, not with today’s technology. But you can make one-off sweaters. Likewise, single ply cutting is also largely untenable (and arguably undesirable from a cost perspective) but that wouldn’t be true of leather CAM cutting. Where do we draw the line? Is it cost, time, ROI? What do you think?

Get New Posts by Email


  1. Mike C says:

    Fit Couture is partially lean. Our sewing room is in pretty good shape, but our cutting room has a long way to go.

    Until we get our cutting room to the same level as sewing, we’re forced to do a ton of sub-assemblies. Basically all long pants, some shorts, and some cropped pants are done as sub-assemblies. With, say, style 5009 we’ve got 96 individuals SKUs. To avoid having to inventory all that, we’ll make up sub-assemblies consisting of incomplete pant shells that will have trim attached and be hemmed to length as orders come in.

    We batch some operations that require very little time and are themselves subassemblies. Waistbands are a good example. The total time allotment for a waistband is 42 seconds – the time required to move a single piece at a time from a serger to the finishing station is longer than the time required to pick them up and put them down in batches. They are unique in that there is only one cut piece and only one sewing operation done. I need to do some timings on a part of the operation that might be improved upon. Next time I do a batch of them, I’ll see if we can remove some of the batching.

    Once we bite the bullet and get a good auto-cutting solution in house, the only inventory we should be forced to carry is that which is needed to keep our staff working through the slow summer months when we have excess capacity. And, even that I hope to be able to solve by finding people who’d like to have the summers off and only work during the school year.

    (Along those lines, I discovered a way to batch orders into our Optitex marker software. I could theoretically write some code that would take import a list of pieces required to fill orders for the day from our order management system and then automatically generate the required markers, by color, that would then feed it to an auto-cutting station where it would first encounter a human being as they loaded up the appropriate fabric rolls. *That* would be cool.)

  2. 42 seconds for a waistband? Oh, I wish I wish I wish.

    Thank you, Kathleen, for defending the needle/thread changing hassle to Eric (maybe this could be your next lesson with him! LOL).

    I continually fight the batch/one-piece production. I work mostly with straight pieces, and with some styles if I don’t iron them when everything is still straight (usually before the finishing step), then ironing takes way too long. So I do batch ironing, and then batch that last step.

    I’m not sure any process can get totally, 100% lean. There are always subassemblies, there just are. But, with the lean model in mind, any manufacturer can become more efficient. I think it’s the commitment to batch assembly that’s the stumbling block.


  3. Lisa Bloodgood in Portland says:

    Well, I, who am not an expert, say that subassemblies and batching depend on what you’re sewing.

    I recently made about 20 handbags and several pairs of pants. I could only batch within colorways. Subassemblies were the straps, shell, and lining for the handbags and the inseam pockets on the pants. But I put a drawstring in the pants as I sewed the casing, which was the last step. Was that a subassembly or not?

    On both I batch ironed in that I sewed a bunch of seams, pressed them, sewed more, etc. like Marguerite did because it would have been too hard to do it after the items were completed and they wouldn’t have looked as good.

  4. J C Sprowls says:

    Defining what is a subassembly is tough. A lot depends on whether a dedicated setup is needed. That said, let’s stick with waistbands; but, a different flavor: tailored waistbands w/ a curtain.

    In most cases, the wb curtain can be purchased already assembled to spec and on a roll from a supplier. So, the need to create that subassembly is already built into the cost of materials. That leaves the wb self, the belt loop carriers, the hooks/eyes, label affixing, interior buttonhole, button attaching and felling the curtain steps.

    The beltloops would typically be a subassembly because it requires a dedicated setup. In my shop, I have one blindstitch machine that setup with folder to make the loops. Then, I re-purpose that machine by moving it to another location after I’m done batching the loops.

    For one-offs, I sew one continuous strip (1 yd) w/ double-faced fusible inserted into the folder. When it’s done, I take it to the clamshell press for fusing and then cut it apart into individual lengths after it has cooled. A template is affixed to the table edge to ensure consistent lengths. For multiple units, I roll the narrow strip onto a cardboard insert and mount it onto a strip feeder over the machine head, run it under the needle and back out onto another insert.

    The loops subassembly, fusing and cutting is usually done before I pull the cut units into bundles. My focus is typically setting up each machine with the needed supplies for that round of production.

    When I move to (sit down) the sng ndl machine pre-made loops, labels and hooks/eyes are in bins. The wb curtain is on a roll over the machine head, making it easy to insert it when needed or roll back out of the way when not needed.

    I set the self fabric, folding up the pleats and inserting the prepared loops at appropriate intervals. Then I set the wb curtain to the self, cutting it off the roll as I approach the required length. The label is set to the curtain and the hook/eye are set with pliers before stitching closed the ends, trimming and turning.

    Since I don’t have a bar tacker (nor do I like them on tailored pants) I use the sng ndl to tack the tops of the loops to the waistband. On some styles, the loops are inserted into the wb curtain, wrapping to the interior of the pant for a clean finish – they’re still tacked on the interior edge while the curtain is flat, though.

    The bundle then moves to the buttonholer for its buttonhole (batching not required if standing). And, then the button attaching point is marked before progressing to the button attaching machine where the buttons are stored in bins (batching not required if standing). After all this is done, it moves to the blindstitch machine to fell the wb curtain (again, batching not required if standing).

  5. Andy Chang says:

    this is more a question than a comment.
    We are an apparel manufacturer located in China and has been using a traditional production flow for the last 33 years. We have a total of 50 production lines and 2000 sewing workers.

    Our management firmly believes that lean manufacturing requires a fixed product where you can define time for each operation and arrange the flows accordingly.

    However, most of our customers are Fashion leaders, meaning they don’t have large quantities. Sometimes orders are as small as 500 units. So before the production flow starts, it has already ended, which makes the entire engineering process inefficient.

    Due to recent cost increases in labor, fabrics and well pretty much everything, we are looking into the concept once again.

    Like to get some feed back and see if Lean Manufacturing can also be applied to sewing lines that are making frequent changes to styles and etc.

  6. Kathleen says:

    Hi Andy, LTNS.

    I’m familiar with some of your products and I can see where you’re going with this. I don’t think this is insurmountable but considering the size of your operation, I’d urge you to contact America’s 21st. I think a limited roll out for test purposes is well worth exploring. Btw, the average production lot size of the largest manufacturer in the world (Zara) is only 300-500 units.

  7. Kursat says:

    Hi all,

    I came across this page while searching for SMED applications in sewing lines. I’d highly appreciate anyone sharing their experience, how-to and do’s and don’ts?

Leave a Reply to Kathleen Cancel reply

You have to agree to the comment policy.

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