This website covers topics on knowledge management, personal effectiveness, theory of constraints, amongst other topics. Opinions expressed here are strictly those of the owner, Jack Vinson, and those of the commenters.

TOC ICO 2006, Workshop Day 2

The second day of the TOC ICO conference was another full day "upgrade workshop" with Eli Goldratt.  He covered a Strategy & Tactics Tree for developing a mutually-beneficial collaboration between a manufacturer and a distributor.  He also talked about an example of TOC issues that surfaced in a large retail organization.  There were some other tidbits throughout the day.

Goldratt started the day discussing the Chinese economy and the unbelievable opportunity and risk that it faces as it grows to match the standards of the western world.  This was used as an entree into discussion of the collaboration S&T Tree.  The interesting thing about the overall discussion is how clearly the collaboration must be a win-win situation for everyone involved.  The distributor cannot be seen to be taking advantage of the manufacturer, and the manufacturer cannot be seen to be trying to do out-price or do end-runs around the distributor partner.  The only way to make this happen is for the collaboration to create a situation where together they can have a far greater impact than they can as individuals.

For those who understand the Drum-Buffer-Rope (DBR) operations solution, I heard new things about how Simplified DBR (S-DBR) works.  The original understanding I had was that it operated strictly by assuming the market is the only constraint, meaning that the plant should take every order that comes its way.  However, if this is not controlled properly, there is a distinct possibility that a spike in orders might overwhelm the internal capacity constrained resource (CCR) - the constraint of the plant.  What S-DBR needs is a mechanism to tie the ordering process to the load of the CCR - to manage the pipeline of orders.  While there is acknowledgment of an internal constraint in the plant, there is no need to explicitly monitor a buffer in front of this constraint because this is effectively managed by the order process.  Cool. 

The discussion of retail organizations is fascinating, as I have not investigated this area very heavily.  One of the key things discussed in the TOC communities with respect to retail and the supporting distribution system is that there is a constant fight between stock-outs and carrying too much inventory.  The balance usually lies toward carrying tons of inventory to ensure the retail positions never stock-out.  However, even with massive inventory (and low inventory turns), most retail shops lose sales due to unavailability.  But how much do they lose?  Retailers don't know.  If pressed, they think it is 2-3% of sales are lost due to unavailability -- or maybe less, if there are similar products available. 

There is a way to check this in most modern inventory management systems.  How many SKU's (stock keeping units) are supposed to be on the shelf?  How many are actually there?  In most cases, if it isn't on the shelf, it translates into lost sales.  Even worse, which SKU's are the ones that are missing?  It is the high movers!  And what is on the shelf?  The low movers?  This means that the sales lost is even higher than the straight accounting for the fraction of SKU's that are missing.  On top of that, the retailer is wasting shelf space and attention on SKU's that aren't moving.  Ouch.

TOC ICO, TOC and Traditional IT

Knowledge workers do more than search