Sunday, October 07, 2007

Creating Resilent Supply Chains Without Depending on Forecasting

Many years ago, I remember having a discussion with one of my company's Sales Managers. He was responsible for an important, but difficult, customer. "Jim," I said, "I'm having a real problem with with your forecast accuracy. You've been forecasting the same order for four months; and I have been carrying component inventory for nearly two. When are you going to close this deal?" Jim replied, "Hey, Sunny, we got a new order last month." "Yes," I said, "for a product that was not even on your forecast. I'm paying expedite fees to meet your promised ship date. I have calculated your forecast accuracy at 50%. Jim, that's no better than a coin toss. I need better forecasting from you." Jim mused, "Sunny, there are only two kinds of forecasts, 'Bad' and 'Lucky'. I can't always be lucky."

Does this sound familiar? Chances are it does. Almost everyone I talk to, and in every department (Sales, Planning, Purchasing and Manufacturing), complains about the forecast. Yet, many producers heavily rely on forecasting even when they admit it is not accurate. Worse yet, most companies have intermediate processes to review and revise the forecast because they do not trust what comes from the field. This step adds latency to the process, further causing the information to be less accurate.

Reducing Supply Chain lead-time can reduce forecast dependency. There are many ways to accomplish this, but ultimately, lead-time reduction offers multiple benefits including reduced inventory, improved responsiveness and better on-time delivery. It also offers the same benefits to your suppliers and their suppliers. Before you run off and start negotiate lower lead-times with your suppliers, make sure that the reductions are real. It's critical to understand the "physics" of your supply chain. Producers must work closely with their suppliers; building trust is essential. Arm-twisting a supplier to sign-up to a lead-time they can't possibly meet is a receipe for disaster.

Yes, buffering inventory may be a necessary evil, but inventory risk can be dramatically reduced if effectively managed. I am not talking about buffering finished goods or in-process inventory. Instead, I propose going deep into the supply chain to the lowest possible cost point. Your company will dramatically reduce lead-time, even if it is necessary to share some risk with your suppliers (or their suppliers). Analyze the liabilities carefully, because if executed properly, the risks will likely be offset by big benefits. You may even find that you've gained new business opportunities through your organization's increased responsiveness.

Lead-time reduction is also an effective process during New Product Introductions (NPI) when a forecast may not even exist. I will discuss some of my ideas on implementing lead-time reduction in my next post. Stay tuned.

In the meantime, feel free to comment on your experiences.


wms said...

I absolutely agree that reducing lead-time and processing time can reduce certain amount of required safety inventory.

rmichaelorr said...

Great in concept, difficult in practice. I'm on the business development/sales side of this and couldn't agree more with the comment on forecasts being bad or lucky. Especially in the aerospace and defense business where the end customer is facing massive budget shortfalls...