Cutting Lead Times for Custom Goods

My first real job after college was at a company that made electron beam deposition systems for the then-fledgling semiconductor industry. These coaters, as they were called, vaporized small metal ingots - primarily gold - in a vacuum chamber lined with wafers that would eventually be etched and sliced into individual logic or memory chips. Every coater shared a number of major components. They all had a vacuum chamber, a couple of vacuum pumps, an electron beam generator and a control panel. But there the similarities ended.

Some needed spindle assemblies to rotate the wafers through the cloud of vaporized gold. Some needed heaters to pre-heat the wafers. Some needed to inject argon gas into the vacuum chamber at some point in the process. And so on.

After working at this company a few months it became clear that the business was in trouble. The parent company apparently considered selling it. Why else would those strangers with clipboards be wandering around the building counting the number of desks, chairs and saleable fixtures? But instead, a consulting group was hired to fix things. And the first mandate of that group was: Don't ever start building a system until you have a signed purchase order.

As  naïve beginner, the idea of building a customized machine that cost close to a million dollars (inflation adjusted) without a firm commitment from the buyer seemed crazy. If that was a common practice, no wonder my employers were in trouble!

Later, however, I overheard a conversation between the VP of sales and the VP of manufacturing that revealed a method to their madness.

It turned out that my company won many of its orders by promising significantly faster delivery than its competitors. They did this by starting to build coaters before they were actually sold. If the deal fell through, they just reworked the machine to fit the requirements of another customer. And although this resulted in a significant WIP inventory, it gave the company a huge competitive advantage.

What the consulting group should have recommended was a way to formalize the system of pre-building to maximize efficiency and minimize waste. Waiting for signed purchase orders would have killed the company.