Irregular seasonal demand is a challenge for many manufacturing operations today striving to implement flow in their production systems. But companies can optimize flow in these high-variety environments by designing mixed model value streams that produce products at the pull of the customer.
Step One: Identify the Right Product Families
Just like establishing flow in companies that must produce hundreds of products through the same value stream, the concepts of mixed model production also apply to environments with seasonal variation in demand since mixed model can respond to the variability as needed. The key is to start by selecting the right product family.
A product family is a group of products that pass through similar downstream processes or equipment and have similar work content. To identify product families, it’s helpful to create a matrix that lists processes in columns and a list of products in rows. The matrix is then sorted so products with similar processes – meaning those that require about 80 percent of the same downstream processes – are grouped together. The families can be further defined by applying a work content criteria. Generally, the total work content of the downstream process steps for each part in the family should be within 30 percent of each other.
Designing Lean Value Streams
Once families are identified, the next step is to design a value stream for each family. The first guideline used to design value streams is to establish takt time. In mixed model production, takt time – which is the rate of customer demand – is calculated at the pacemaker (the point in the flow for a family where work is scheduled that determines the speed at which the value stream will operate) by dividing the effective working time by the total demand for the various models running through the pacemaker.
Next, it’s important to check that there’s enough machine capacity to support the proposed product family and mix within the takt time by reviewing the effective working hours and the cycle time, meaning the time it takes to get a piece out of a process.
It’s also critical to determine the interval, or how often the pacemaker will produce all the parts in the product family. For example, if it takes one week to produce all the parts in a given family, the interval would be one week. The key is to lower the interval as much as possible to increase flexibility and on-time delivery, and reduce changeover times and inventory.
When Customer Demand Varies
After the capacity of the pacemaker has been established, the next step is to understand how to handle variations in volume. When customer demand changes go beyond the existing design limits of the pacemaker, there are different methods that can be employed, including:
- Supermarket. When demand changes, a supermarket – a controlled inventory of items that is used to schedule production at an upstream process through visual control or Kanban – can be used to help smooth it.
- Takt Capability. If demand trends begin to change and exceed what can be produced, companies can level the amount of production for a fixed period of time by setting different takt capabilities or “modes” for the value streams. These pre-established modes are ready to go when demand changes to more than the current mode can produce to meet seasonal trends in customer demand.
- Dynamic Finished-Goods Supermarket. Another tool is to create a dynamic finished-goods supermarket that changes based on demand signals. To make a supermarket dynamic, a system can be put in that alters the supermarket size in line with demand changes. When deciding on a finished-good strategy, a good method is to break demand into lead-time intervals. Then the operation changes the finished goods supermarket based on lead-time intervals. For example, if five lead-time intervals from the present day demand exceeds what can be built, then that will send a signal to increase the supermarket and begin building immediately to the higher level. The finished goods supermarket needs to be dynamic and expand – and shrink – with customer demand.
- Leveling by Intervals. While dynamic supermarkets can help level production as demand changes, to build exactly to demand, which should be the goal, the concept of leveling by intervals can be applied. This is done by looking at what interval can be supported – for example, if an operation cycles through its product mix every six weeks, then the product would be leveled to six weeks and not a fixed calendar time.
For operations with seasonal variation, balancing the interval is a key concept in designing a capable system that can respond to different demand. For example, if demand increases at a certain time every year and everything that’s produced gets sold then, increasing the interval would allow for more productive time. Then in the slow season, the operation can decrease the interval.
By following a step-by-step approach to design mixed model value streams that produce products at the pull of the customer, the end result is a process that can absorb variation and transform a complex environment into a lean environment. The results? Reduced inventory, increased on-time delivery, and improved quality, which will yield a competitive advantage and increase market share.