Inventory Management Review

More on Wal-Mart, Bullwhip, and Vinson

Last post, I discussed the bullwhip effect and wrote that I would soon be responding to Jack Vinson and his comment.

Jack pointed out that "Wal-Mart is (or should be) providing their suppliers with excellent information about the movement of their products through distribution centers and into stores," and went on to say that this should be keeping inventories "DOWN" for suppliers, not up.

Jack, you're definitely right, the more information a supplier has about his customer, the lower inventory levels SHOULD be.  BUT, there are some other issues at hand.

First are the batch sizes. Wal-Mart, as we know runs a pretty efficient JIT system.  This means small batches.  This makes sense because the factors that determine batch size, as derived from the EOQ, are ordering costs, demand, and holding costs.  Because the ordering costs are so low, it makes sense for Wal-Mart to order at the drop of a hat and force suppliers to hold onto the rest of the inventory.

But, suppliers, even if they have the same demand data, have higher ordering costs.  In this case, the cost of ordering may include manufacturing setup costs, or it may be the result of economic efficiencies derived from larger batch sizes.

That's all a bunch of fancy talk to say that yes, Jack, the information helps, but the fact remains that ordering costs for Wal-Mart, with their EDI systems, are almost non-existant, but ordering costs, in the form of setup costs, for producing less at a time for the suppliers are not so low.

So, if the EOQ for a supplier is 50000 units, even if the data Wal-Mart suppliers them with data that shows that those 50000 units will be used up over the course of 1 year, that is still 25000 units in average inventory.  Now, if Wal-Mart didn't have a JIT system, then they would perhaps be more likely to buy those 50000 at a bulk discount, thus lowering average inventory of the supplier.

The end result is, because Wal-Mart places such high demands on service, and because they now refuse to buy in bulk, suppliers are forced to hold onto more inventory than in the past.  The data provided by Wal-Mart to their suppliers definitely helps to keep that excess inventory to a minimum.

However, as I mentioned in my original article on the topic, Wal-Mart's no tolerance level places very high expectations on its suppliers.

Clearly there are lots of what-ifs to discuss here.  Yes, Jack is correct, the data helps, and without it, Wal-Mart's demands would result in even higher inventory levels.  But it doesn't account for batch sizes, nor for the, as Jack put it, "hiccups" in in demand.

Those "hiccups" are not exactly something to sneeze at either.  Statistically speaking, providing a 95% service level, which is common for most suppliers, is a fairly attainable practice.  However, Wal-Mart requires near perfect, and statistically, this results in MUCH higher inventory levels.  Until the demand data provided by Wal-Mart results in no hiccups, which will never occur, the 99% service level Wal-Mart requires will have a substantial effect on inventory.

Conculsively, Jack is absolutely correct that Wal-Mart's data helps to lower inventory levels across the supply chain as it keeps the bullwhip effect to a minimum.  But, Wal-Mart still forces suppliers to hold the lion's share of the inventory, which results in higher inventory levels for suppliers, just as Ms. Lacefield wrote about in her article in the October 2005 Logistics Management (Vol. 44, No. 10).

The Bullwhip Effect

Much discussion has been stirring within the IMR community regarding my post on Wal-Mart's suppliers.

As I mentioned in my last post, I have decided to devote an entire article to the bullwhip effect.  This will be a solid foundation for next week's post which responds to long-time reader, Jack Vinson of Knowledge Jolt, and his recent comment.

But first, the bullwhip effect.

I will refrain from entering a mathematical discussion regarding the bullwhip effect and instead will stick to the basics.

So first of all, what is the bullwhip effect?

The bullwhip effect is a wasteful phenomena that occurs due to a lack of information across the supply chain.  The result is excessively high inventory levels increasing as you get further from the end customer.

Basically, the bullwhip effect is safety stock for safety stock, because suppliers hold extra stock for their customers the same way retailers hold extra stock for their customers.  Suppliers need safety stock, for the safety stock.

Great!  How does it work?

Essentially, it's like I said.  Incongruent information across the supply chain.  If a retail store that typically sells 100 units a week all of the sudden sells nearly 200 per week, then this is going to result in the supplier producing more than 200 in order to have a safety stock for its customer.

Now the supplier is producing 200+x.

The supplier's supplier now needs to ramp up in order to have a safety stock that results in 200+x+y.

As you go down the supply chain, more variables are tagged onto the end of that equation.

The problem is, the supply chain as a whole needs to be able to satisfy the same demand.  If the retail store needs 200 units, then everyone in the chain should be prepared to supply 200 units.  If this isn't enough then it isn't enough, but if it is, then it is.

With the bullwhip effect, the retailer wants 200.  If it is enough, then good...except, the supplier has too much, and the supplier's supplier way too much.  If 200 isn't enough, then it's good that the supplier has extra, but if 200 is the optimal inventory level determined by proper ROP formulas that is needed to supply the actual demand with as little excess inventory required to satisfy a particular service level, then 200 is the amount the entire chain should use.

This is a very simplified version of the bullwhip effect, but ultimately, this is what is occurring.  Suppliers ramp up in order to prevent stock-outs.  Unfortunately, they are ramping up to prevent a stock-out for their customer, not to prevent a stock-out for the supply chain as a whole.

The result?
Waste.  Small movements at the end of the supply chain trigger exponential movements down the chain.  Oh, I get it, like moving the handle of a whip.