In response to: “How Manufacturing Software Should Adapt to Support Lean Principles”

In response to: “How Manufacturing Software Should Adapt to Support Lean Principles”


ERP Analyst, Software Advice
December 16, 2011
Nice position article; we appreciate thoughtful discussion on issues critical to optimizing value chain effectiveness and profitability. I think productivity is even critical today given current competitive and regulatory environments.  That said, I am not sure I agree entirely with your point of view: I think you miss the other side of the equation which, together with your position, would actually create the cohesion you espouse.

So a couple of two or three points to consider.

First, one consistent between the two philosophies is the need for production requirements i.e. demand; it goes without arguing that the more accurate the production requirements, the better both will perform. So one way to bridge the gap between the two is to agree that both approaches are made better through improved demand planning and specifically improved forecast accuracy at the execution level.  It does no good to plan production, in either approach, to 4 decimal places only to have the forecast or production requirement end up being off by 30 or 40 or 50% at the execution level.  I am sure this point does not surprise you coming from me.

Second, this is an important discussion because it focuses on execution which is important to us.  No matter how accurate a forecast, it is worthless unless used to improve specific supply chain activities within the flow that you rightly highlight in your discussion.

Lastly, the core hypothesis is that software can and should work to accommodate “lean” and you highlight three areas within which to do so.  And, coming from the perspective of lean, these are solid focus points that I would agree should be address; I would go so far as to say they are being addressed by certain S&OP and manufacturing software vendors.

However, I would argue that lean also needs to grow and adjust to accommodate the principles of MRP.  There are some who argue that lean was developed in the vacuum of driving an accurate forecast and demand plan – hence the revival of demand driven supply chains by the people you mentioned as well as Gartner/AMR and others.  in general is a reactive dynamic; given a specific production requirement, what do we need to make and then how do we do so most efficiently.  And in a perfect world that is great.  However, what if the needed production requirements change?  What if they change by 40% (up or down) in a 2 week window?  What if the vendor does not show up with the needed WIP until 7 days later?  Not everything goes perfectly and this is where MRP and it related and perhaps even more important cousin DRP come into play.

When done well, DRP/MRP take into account variables such as forecast error, vendor performance, inventory costs (full costs), order fill rates and consumption patterns (to name a few) in order to help anticipate issues and drive supply chain operating decisions not only in optimizing performance for today but also for various time frames into the future based on lead times and customer promises etc. and in so doing, help minimize the risks of disruption.  Practically, this helps drive decisions such as what specific inventory to carry and who much which would actually work to make lean perform that much better in a more flexible manger.

There is no question that software can always look to improve ease of use, flexibility and value add.  But it should do so within a framework that takes advantage of what technology can do while supporting operational philosophies such as lean so that both sides can progress.  This is true for expansive ERP systems as well as useful spreadsheets.  And, if they are expected to execute based on a forecast with measurably decreasing forecast error, there is no need for these two approaches to be mutually exclusive.


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title="" rel=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>