Creating flow in complex environments can be challenging, especially in operations where processes or machines are shared across different product lines. In these situations, the concept of the interval is critical.
The interval tells how long it will take to cycle through the mix of products that go through a shared resource. Once it’s established, an operation can created a guaranteed turnaround time (GTT), which promises that if a product is at a shared resource by a certain time, it will be delivered to the next process by a certain time.
Since the interval is so essential to establishing regularity and predictability at shared resources, it’s important to get it right – and that means making it small.
Defining the Interval
An operation uses the interval to measure the capability of a process, meaning how much it can produce in terms of volume and mix. Typically, an operation defines its interval at the pacemaker, which is the single point at which it schedules the value stream. But an interval can be defined for every machine and process in the same value stream. To illustrate how to establish the interval, it’s helpful to see an example of it in action.
Below is a product family consisting of four products, all of which are made on the same process.
Each product has a specific weekly demand and the production capability for the process that produces these products is four hundred units per day. Here’s one way an operation can produce all the products in the product family and satisfy customer demand.
On Monday, the operation begins producing plastic arms. It continues to produce them on Tuesday, and, since the demand for them is one thousand units per week, some time on Wednesday needs to also be devoted to them. The rest of the production capability on Wednesday is used to make steel arms, which will be finished on Thursday. Then, on Friday, the operation makes aluminum arms and uses the remaining production capability to make titanium arms.
What’s the interval in this case? If the interval is defined as the time it takes to cycle through the mix of parts in a product family, then the interval in the above scenario is one week because it takes one full week to produce all four parts in the product family.
Another way to produce the parts in the product family is below:
In this case, rather than produce the entire weekly demand for plastic arms before moving on to the next product, on Monday, the operation makes some plastic arms, then some steel arms, then some aluminum arms, and then some titanium arms. For the remaining days of the week, it follows the same pattern. In this case, the interval is one day because it takes one full day to produce all four parts in the product family.
So there are two ways to approach production: in the first example, the interval is one week and in the second, it’s one day. Which is better?
Less is More
As production lead time, which is essentially the same thing as the interval, increases, so does variation and waste at all levels of the organization. That means with shorter production lead times, they’ll be less chaos since there will be fewer opportunities for things to change – or go wrong.
For example, using the one-week interval scenario above, what happens if there’s a change in customer demand? Suppose a customer calls on Tuesday and wants titanium arms, which aren’t scheduled to be made until the end of the week. The earliest they could get to the customer is after production on Friday. Or the operation could stop running plastic arms and switch to titanium, but that would cause disruption.
Another potential issue with a longer interval is lot size. If an operation’s producing each part only once per week, it needs to produce a full week’s worth of it to ensure enough inventory. But that means it will be carrying a week’s worth of inventory every time it runs a part, even though it will be a mix of some that was just made and inventory made in the last production cycle that didn’t sell yet that week.
Lastly, quality can also suffer with a longer interval because defects will only be identified once the full run has concluded. Depending on the cause of the defects and when it occurred in the run, there could be a large amount – up to a full week’s worth – of defective parts.
With a one-day interval, on the other hand, the potential for negative consequences is greatly reduced because the operation can be more responsive to changes in customer demand, will have small lot sizes for improved inventory turns, and can catch defects after a maximum of one day’s worth of parts is produced, reducing scrap.
While the smaller interval always yields better operational and business results than the larger interval, remember that “small” is a relative term and will vary widely from industry to industry.
Don’t Forget Changeovers
While the one-day interval scenario above is the better choice, it also means that the operation is doing five times the number of changeovers over the course of the week (25) compared to producing at a one-week interval (5). So if changeover times are preventing the achievement of a smaller interval, an operation needs to get really good at changeovers.
By reducing changeover times through programs like Single Minute Exchange of Die (SMED), the operation will drive smaller intervals, and positive business results.