It seems hard to accept the fact that you can cut the inventory in a company and run out less often. The answer lies in how we think about and order the inventory.

In the "good old days" of high-inventory manufacturing, the cost of money was relatively cheap. Therefore, the costs associated with holding inventory also were relatively low. That meant that companies were a lot more inclined to hold inventory just in case they needed it.

Let's take a moment to be clear on one fact. Any inventory, except for what you need to process orders in hand, is an insurance policy. It's designed to protect you from differences between what your customers demand (sales lead time) and what you and your vendors can provide (manufacturing lead time). The question becomes how can we maximize the protection of this "policy" while paying the minimum premiums for it?

Unfortunately, most batch sizes aren't designed with the idea of protecting the customer; they're designed with the idea of minimizing changeover costs for manufactured items and purchase prices for purchased items. Both of these goals have the same effect of encouraging the purchase and processing of large batches of product to minimize the costs per part.

So if you have huge batches of everything, why do you still run out? The answer is that you can have "everything." Because you're running huge lots, the inventory levels of parts in the plant will vary widely. You may have months' or years' worth of supplies of some items, and completely run out of others. And it's an interesting paradox of manufacturing that the larger the batch sizes in your plant, the longer it takes for any individual object to move through the process.

Consider the following example. Each line has the exact same five operations, with the same setup and run times. The batch quantity for Line #1 is 100 pieces, and for Line #2 it's 10 pieces.

If you were to look at only the minutes per piece, as we have traditionally done, you would think you're incurring a much higher cost, since your labor minutes per piece are almost twice as high. It's easy to see how this thinking would encourage a company to run bigger lot sizes. Bigger lots mean less cost, and less cost means more profits. We'll save the argument as to why this isn't the case for another day.

The part I want to focus on here is that we used to never consider the effect of the lot size on the number at the end of each line -- the total cycle time. If I start a part through on Line 1, it will take me four times as long until those parts will be available for shipping, assembly, etc. And as you increase your batch size, your time through the process will increase.

So often I have heard companies complain about long lead times, and the usual response from the rough end and machine room department foremen is that they need to run larger batch sizes to increase the amount of run time on their equipment. In their minds, if they run more, and setup less, they will catch up. Sadly, the opposite is the case. As you increase batch sizes, you decrease your ability to respond to changes in demand.

On the other hand, companies that work hard to reduce batch sizes intelligently find that they can react faster to the inevitable crises that come up. Because they "fan" through their product line faster, a broader range of parts are available at any time, but just not in the huge quantities that they may have been in the past.

That brings us to the last argument we hear for large batch sizes: that if we get hit with a huge order, we'll be covered. It has been my experience that companies making this argument have average inventories on the shop floor of at least three to four months. How often does a customer place an order for that magnitude for immediate delivery? And when they do, what are the odds that all the parts needed for that order will be available for immediate delivery? Most of the time, you'll have to bring some parts through from scratch. And, as we have shown, your large run quantities make it more difficult to bring parts through quickly.

But let's just say that by some miracle you get a huge order and are able to fill it in a week. You win a large new customer. Wonderful! What kind of delivery performance are they going to expect from now on? And what will they do when you can't deliver every time like you did on the first order? We started out saying that inventory is an insurance policy. It also is a security blanket for production.

There was a day not too long ago when a lot of inventory on the floor and a big backlog of orders were signs of a healthy company. Today we know that they are really signs of terminal thinking and a failure to recognize the realities of today's marketplace.

It's hard to give up that security blanket, but the fact is that the only real security in today's business world is to remain better than your competition.