Your production line may have different workloads for different product variants. This unevenness causes waste and overburden. In this series of posts I will look at ways to address this unevenness. The first post was an introduction to the topic. This second post will look at ways to simply eliminate the problem – although this may not be feasible for many cases. In the next posts I will look at adjusting the capacity and finally at adjusting the product sequence through Mixed Model Sequencing.
As I said in my previous post, one option to resolve the issue with unbalanced workloads is to simply make the problem go away. While this sounds sweet, it is often unfortunately not an option. Nevertheless, you should check if it is possible, as it may make your life in production easier and your company more efficient.
Adjust the Product
One possibility is to adjust the product. For example, if you are attaching wheels to the car, some wheels may have four nuts and other five or even more. Obviously, the more nuts a wheel has, the longer it takes, and you have a product-dependent variable workload.
One possibility is to change the product to remove these differences and make the workload more consistent. If you can do that, try to move to the faster/better/cheaper options and away from the expensive ones.
Unfortunately, the product design is often out of your hands, and the designers will claim that you can sell so many more cars if they have an extra lug nut. Depending on the customer this may even be true. If your company offers only two-door vehicles, you miss out on a large market segment. The same if you do not have sunroofs. Hence, usually our options are limited here, but maybe you are one of the few lucky ones that can twiddle with the design to improve manufacturing.
Adjust the Process
If you cannot change the product, try to adjust the process. Can you make tools, jigs, gizmos, or other devices that simplify the different workloads so that the actual time is the same? Of course, here you should pretty much always go for a faster solution. There is no point in intentionally making the process slower for one product variant!
Let’s stay with the while example. If you cannot change the number of nuts, can you simplify the tightening of the nuts? Many car companies use an automated tire assembly device as shown here in a photo from Suzuki. While this device in the photo is designed for four lug nuts, you probably can design a similar device that can handle three, four, five, or even more nuts automatically at the same time. In this case you have eliminated the uneven work content.
This example with the lug nuts is only an example, and probably won’t fit your situation. But do think a bit if there are ways you could reduce the unevenness of the workload through better tools. It may not always work. For example, if you install a sunroof, there will be additional work no matter what you do. Yet even here you may have options.
For example, you could build a fully automatic sunroof installer. True, the work content is still fluctuating, and the machine may have idle times if there is no sunroof to be installed.But it is a machine and not a human worker, and as long as the machine can satisfy the line takt, I don’t care much about waiting times of machines. While not completely irrelevant, it is usually so far down my list of priorities that it is nothing to worry about.
The bigger issue is the cost-benefit analysis. Creating a fully automatic sunroof installer may not even be feasible. At least it is quite expensive, and the cost benefit of having less fluctuating workloads for the human workers may not be worth the cost of an expensive and potentially troublesome machine. Your choice.
Push it Elsewhere
Sometimes it is possible to move the problem workload out of the main line and towards a secondary production that supplies the main line. This can reduce the problems at the main line, although it may introduce problems at the secondary system. However, sometimes the secondary system can be adapted easier to handle the issue. Even if it is still a problem, the secondary line may be shorter, fewer workers may be affected, and the problem may be overall smaller.
Separate Production Lines
Another option for changing the process is more radical, and for most situations useless. You could split the production line into two independent lines. One line makes only product A, while the other line makes product B. Each line could then be optimized and balanced for its own product, with no worries about the workload of the other product.
For example, instead of having a joint assembly line for two-door and four-door cars, you make one line exclusively for two doors and one line exclusively for four doors as shown below. This, however, is only a theoretical example, as it makes no sense for the car industry to split these lines. You will lose flexibility, you will lose efficiency, your machine expenses will go up, and so on. In most cases it makes no sense to split a line.
But maybe, just maybe, you have a situation where this is feasible, where you have only one or two product variants or product families, and inexpensive machines. For example, if it is a purely manual assembly, you could have a long line with ten people or two shorter line with five people, each line making only one product (family).
As you can see, there are some options on how to eliminate the problem of uneven workload due to different products – or at least push it on a machine where we don’t really care, at least not too much. Yet they may be difficult in many cases. It is quite likely that you read through this post looking for a solution, only to find that none of these options will work for you. But don’t despair. In my next few posts on the topic of product-dependent workload, I will look at adjusting the capacity, before finally getting to product sequencing to handle workload differences. This last one is usually feasible, but it is also more of an art than a science. Until then, go out, see if you can reduce unevenness in the workload (or elsewhere), and organize your industry!