Yes, you read that right: “quick-turn” boards cost less than “standard” lead-time boards. That’s because shorter cycle times go hand-in-hand with lower costs. I’m not saying that you can buy them cheaper, but you can make them cheaper with a shorter cycle time. Manufacturers can (and do) sell boards for whatever they can get. That’s because supply and demand controls the board price. But having the lowest cost puts any manufacturer in a better competitive position, and notwithstanding the other competitive advantages of a shorter lead time, one surefire way to have a lower cost is to have a shorter cycle time.

We are all familiar with the many ways to trim costs. Organizationally, you can streamline operations, increase efficiency and reduce overhead for starters. For direct costs, you can reduce labor and/or supply costs. Or you can just have everything built in China. Many of these things are hard to do, and may seem to work at cross-purposes with some of your other goals. Shortening your cycle time is easy to do, and only improves your business. This means that if you are a board manufacturer, you can significantly reduce your costs by producing only “quick-turn” boards. Of course you will no longer consider them “quick-turn” if everything in your shop is moving at the same accelerated pace. Nevertheless the boards will be produced with “quick-turn” lead times.

Beyond the obvious way in which shorter cycle times reduce the A/R cycle (you get paid more quickly), they also lead to lower costs in two significant ways: the first is the way in which shorter cycle times facilitate yield improvement, and the second is through the reduction of WIP, which creates several cost advantages.

A. Higher Yields

Shorter cycle times allow Engineering to increase yields due to improved problem-solving. This is because problem-solving becomes more difficult as cycle time increases, and yield improvement is all about problem solving. Check with your Process Engineers the next time they are called down to Final Inspection to look at a plating problem on some boards that were plated 6 days ago. They will come back with one of two answers: either the problem is still there, or the process is running fine now. If it’s still there, you are looking at 6 days of potential scrap, which may be preferable to the alternative. If the problem has appeared and disappeared before you got a chance to see it, it is extremely difficult to go back and try to re-create the problem so that it can be solved. If that prevents you from solving it, you can almost guarantee that the problem will return, and the next time it may be disastrously worse. Either way, longer cycle times translate into lower yields because of less efficient problem-solving. Since lower yields mean higher costs this relationship is very straightforward.

Less WIP also contributes to higher yields. Since the way to reduce cycle time is to reduce WIP, then reducing cycle time by reducing WIP also improves yield. Consider the fact that, generally speaking, only WIP can be scrapped. Since scrap is the result of process variation or ‘error,’ product cannot be scrapped before it begins the process. Therefore you can state the formula for scrap as the following: Scrap = (Error) X (WIP). Each time an error occurs, it only scraps the WIP that comes in contact with it. A bad driver can only hit the cars that are on the road. It then follows that the less WIP, the less scrap. This may strike you as obvious, but the full ramifications may not be clear. We can test this argument and attempt to illustrate its effects by plugging in a test case. Using the example cited previously, what are the real world ramifications if the engineer who has been sent to check on a plating problem comes back and, instead of saying that the process is now running fine, says that the process is still suffering the same problem? If that process has been running for the last 6 days, then you are looking at 6 days of potential scrap. Now apply this same idea to every process problem that ever occurs, day in and day out. You have to add up all of the WIP that is potentially affected by every potential process problem to estimate the total potential scrap at any given time. This is so unnerving that when I look at WIP now I see only potential scrap. My dream is to build product without having any WIP at all.

Consider how reducing your WIP reduces your potential scrap in the real world. If you have 5000 panels on the floor, and there are 100 ways that they can be scrapped, that means that there are 500,000 opportunities for scrap. If you can reduce your WIP by 30% to 3500 panels, you also reduce your opportunities for scrap by 30% (to 350,000). Without your Engineering group taking advantage of any of the potential for process improvements, you reduce your potential for scrap by 30%. And reducing your potential for scrap reduces your actual scrap. I know when you take inventory, all of that WIP shows up as an asset on your balance sheet, but until you finish it and ship it, it is all just potential scrap. It should now be clear why less WIP equals less scrap.

B. Less WIP (without regard to scrap)

Less WIP means less overhead: When there is less WIP on the floor, the need for Production Control resources like schedulers, expeditors, “hot lists” and the like is significantly reduced. The amount of information that needs to be collected, organized, reported and scrutinized is also significantly reduced. Think how much time and money your organization expends managing WIP instead of completing it.

Less WIP means less un-recovered costs: WIP is essentially an outlay of material, overhead and labor. Therefore WIP represents cost, so the more WIP you have, the more un-recovered costs you have. WIP may show up as an asset on your balance sheet, but it’s not cash, and you don’t get paid until you ship the product.

WIP represents risk: Any process problem, design or engineering change can turn all of your WIP into scrap. The less WIP on the floor, the less dollars at risk.

Less WIP equals shorter cycle time: The next formula I want you to consider is the relationship between cycle time and WIP. Cycle time can be defined using the following formula: Cycle Time = WIP divided by Output. To illustrate this formula, imagine that you have 100 widgets on the production floor, and you are shipping 10 widgets per day. Your cycle time is therefore 10 days start-to-finish. If your WIP increases to 200 widgets but your output stays at 10, your cycle time increases to 20 days. Conversely, if you reduce your WIP to 50 widgets and maintain your output at 10, your cycle time is reduced to 5 days. In short, the less WIP, the shorter the cycle time. In fact reducing WIP is the easiest way to reduce cycle time, since increasing output can be costly or be time-consuming, while much of your WIP is just sitting in a queue waiting for somebody to work on it, and is essentially waste.

So now it’s clear: quick turn boards are produced in shops with less WIP, and therefore have lower costs. Rather than charge a premium for quick-turn boards, shops should thank their customers for demanding shorter cycle times. I suppose offering a discount would be out of the question…