The days of building inventories as a protection against purchasing department snafus and unreliable vendor deliveries are long gone. Gone, too, are the aisles and shipping docks full of in-process and finished goods inventory.

In their place, many companies have embraced just-in-time inventory concepts. But JIT, as popular as it has become, doesn't automatically translate into bottom-line improvement or a greener cash-flow situation.

I recall visiting with a company management team that was patting itself on the back for the success it achieved in trimming its inventories. By implementing JIT, it reduced raw materials, work-in-process, and finished goods inventories by $7 million.

The operations team proudly presented charts, tables, and trend analysis pointing to an inventory reduction yielding a cash flow improvement of $7 million. Based on a 10% cost of carrying inventory, they claimed the $7 million would translate into a $700,000 annual boost to net profit.

It fell to the company controller to deflate their balloon. He concurred the operations team had met its goal of trimming inventories by 70%, but added, “I don't see any cash flow or profit improvements. If anything, our bottom line has slipped and costs have increased.” He showed them the financial reports to prove his comments.

That's when I stepped into the conversation, asking the operations people: “Tell me, as you were reducing inventories, did you notice a change in your plant's performance, especially in the plant's output?” They immediately pointed out that they had more difficulty meeting throughput goals. “And what about your on-time delivery to customers?” They said that at first it slipped backward and then improved, but with a substantial increase in cost of overtime. The more we explored these issues, the more it became clear that as inventory levels dropped, other areas of performance were affected, especially the company's throughput and shipping to customers.

They did not realize that there is a strong correlation between inventory and throughput in any manufacturing flow system – and their organization, being a flow system, obeys these rules.

As inventory levels decrease, it affects the flow process. For example, if the inventory level between two machines is too low, a disruption in the feeding machine may stall it for awhile, causing the other machine to starve and hence lose throughput. The same example can be applied to the organization as a whole – from vendors through raw material storage to the manufacturing process itself, and all the way to the customer.

The company, as it decreased inventory, unknowingly created waves of starvation throughout the entire system. In order to overcome these problems and still meet customer requirements, they had to invest a lot of money in overtime and fire fighting. This excess cost ate up the gains created by the inventory reduction program.

It's obvious we can't go back to the good old days of “just-in-case” inventory. However, we must realize that inventory management based on JIT concepts could adversely impact the organization's performance.

What we need to do is to take a more scientific approach to inventory reduction. First, we need to understand the overall flow process as it connects from resource to resource – vendors to raw material, raw material to machining operations, and all the way to the customer. Second, we need to identify and qualify all of the uncontrollable disruptions. Third, we need to identify key resources whose output is most affected by disruptions. These resources should be protected by strategically locating inventory buffers near them. These buffers act as shock absorbers overcoming accumulative disruptions, much like an accumulator in a hydraulic system. Now we can reduce inventories in most areas while keeping a few inventory buffers at key control points, and hence achieve high output at low cost. For example, the buffers should be located in an assembly operation, in front of a gating resource (i.e. first step of machining), in raw materials and finished goods, and in front of a critical machine.

We have to realize that in a complex system like manufacturing, any change may impact other areas unexpectedly. We have to understand these cause-and-effect relationships and, before we jump on the bandwagon, make sure we do not fall off and hurt ourselves seriously.