I mentioned in my recent email newsletter and a recent blog post, an online discussion I took part in regarding push and pull manufacturing.  It was interesting that it was difficult for the participants to agree on what was push and what was pull.  (I'll have to go back and get some excerpts to illustrate but I'll save that for another post.)

I think one thing that happens in this sort of discussion is a bit of the "apples and oranges" problem.  Different manufacturing processes might call for different approaches.  What looks like push from one view, looks like pull from another.

The whole conversation got a bit testy at times.

As I said, my view is I don't care if it's push or pull, so long as there is flow that maximizes customer service and minimizes cost.  (Mind you, flow tends to be better when product is pulled rather than pushed everything else being equal.  The problem is...everything else isn't always equal.)

Let me give an example from the discussion.  Imagine a product that can be painted any of ten different colors but, up to the point of being painted, is not differentiated.  So it's one product through one process up to painting.  Then it's ten different products.  Strict pull (best flow) might be to get the orders for the different colors, then pull the product right from the beginning.  Get an order for 100 red and 100 blue, you make the 100 red, then the 100 blue right from the beginning of the manufacturing process.  This might even make sense if we assume that the manufacturing lead time up to painting isn't long, the order quantities for each color are large and/or paint shop changeovers aren't a big deal.  But what probably makes more sense is to put an inventory of product ahead of the paint shop, especially if any of the "ideal conditions" I just mentioned don't hold true.

So product will "stop" right in front of the paint shop, breaking the flow.  But we can still look at:

• Flow up to the supermarket,
• Flow from the supermarket through the paint shop to the customer dock, and
• The typical amount of  time that the flow is "broken" (product sits in the supermarket).

Whereas the participants in the discussion debated what was push and what was pull in the above scenario, our response is...Who cares what you call it?  (The way I look at it, the plant will pull to the supermarket in a "take one, make one" approach.  It will also pull from the supermarket through painting to the customer dock in a "order one, make one" approach.  In the meantime, the plant will keep an eye on and study that supermarket to make sure it is the optimal size.) 

In this scenario, if we have too much push, either to the supermarket (not so bad) or to final, painted inventory (worse), we risk having lots of stuff sitting around that we're not getting paid for and maybe even that the customer doesn't want.  If we use too much pull (no or too small supermarket) we risk hurting customer service.  But there's no such thing as "too much flow".  There's just "it flows well" or "it doesn't flow well".

Tune in again for more on this topic.