Improving the layout is often the stated objective on an improvement project. Re-arranging the equipment and the workplaces to minimize transportation will always give you productivity improvements.
If there are only assembly benches and small machineries involved, layout changes can be made on the fly. If major machines (monuments) are involved, such as a panel retrieval system, panel saw
or a finishing line you
want to think twice before moving them.
The schematic in Image 1 shows a portion of a value stream. Material (WIP) is pulled from a buffer to feed the machine or work centre. One or more operators work at the work centre feeding, for example, the machine and placing the WIP on a cart, pallet or gravity conveyor. From here it is staged in the buffer for the next operation. The same cycle then repeats.
In many cases, the transition from the current situation to the above
model would already
result in an improvement.
The improvements would
be the reduction in
transport distances.
Lean manufacturing teaches us to create flow. We would link work centers/machines, so that no stacking down and transport is required. In cellular manufacturing we would
link the machines and
the Image 1 would
become Image 2.
The more processes you can successfully link the higher the impact. The comparison of the two pictures illustrates how much space would be saved. Also, if we eliminate the down stacking, transport and feeding between the work centers and keep the product flowing, it will cut the factory throughput time down to a fraction.
We understand the theory and where we want to end up, so why have we not done the layout change accordingly?
Changing the layout and the manufacturing concept, without doing some homework could spell disaster. There is a reason why we have our layout and why we do it that way.
When you get your employees in the improvement planning (and you should), they will most likely ask for more space. Dropping the idea that they will have, let’s say, only 66 per cent of the current space in future, will probably derail any rational discussion.
We need to dig deeper into the subject and understand what drives the need for space and why we need buffers.
The “JUST IN CASE” bufferLean tells us that more material than what you have in your hand, and what is currently in the machine, is not value-added (the book might even challenge some of the “holding and machine time” as non-value added). Theoretically, if the next piece is always available, we would be fine. There are two main reasons for asking for this just-in-case buffer:
Unstable processes upstreamThe team will tell you the reason they want more buffers is to ensure a steady supply just in case the upstream machine breaks down or the employees in the upstream processes did not come to work that day. To address this, you need to always look at the root causes of these issues.
Is it an unreliable machine or tool that just needs fixing, or is it an old machine with
no redundancy and you
need a new machine or a second machine?
Also, underperforming machines which constantly have micro breakdowns (stopping often for less than a few minutes each time) or machines which need to be slowed down in order to work properly, will interrupt the flow. Often these machines are just limping along and
the problem is often not reported and not realized
by management.
Fit & completeOne of my favourite improvement subjects, is improving fit and complete within the process flow.
For example, when looking along an assembly line/area, how many products are partially completed? How many cabinets are using space
in the buffer, waiting for that one critical part to allow assembly. There
are many possible causes for a part not to arrive
fit & complete.
If the parts are not to spec, meaning wrong size, wrong colour, or wrong drilling they do not fit together and need to be reworked or repaired. Identifying the root causes and initiating corrective actions is the process here.
Parts which were never ordered, got lost, or are just arriving unreasonably late to the marshalling point, have a long list of root causes.
The buffers are there to cover the supply interruptions and to ensure steady supply. Here you have the options of implementing corrective actions and improving the processes or covering up the problem with more and bigger buffers.
It is a sort of an insurance policy. The question is how much insurance you want to buy? The cost is space, lead-time extension, or inventory cost with all their negative side effects. Do you want the insurance to cover you for a few minutes, hours, days or even weeks. Each buffer may have different requirements. The raw material buffer often needs to cover days/weeks, whereas the buffer between hinge drilling and door hanging can be as small as five minutes.
Batch sizeWhen you produce in batches (and most manufacturers still have some type of batching), the batch size also determines the size of the buffer. For example, if you produce cabinet parts in weekly or daily batches, the buffer you produce into must be dimensioned accordingly.
When you produce something in a batch (for example cabinet components), we need to assume that, in the subsequent assembly, the first part of the batch will be assembled with the last part of the batch (Murphy’s law). This means you cannot start assembly until the last part of the batch is completed. This means your buffer size must be big enough to fit at least one batch.
If you want a visual tool, to separate the buffer spaces by batch, then you need to double the buffer space. This method alternates buffer A and B from one batch to the next.
The suggested corrective action here is to reduce your batch size. In my opinion it is immensely difficult to switch from a batch-mode to batch-size-one. As a more continuous improvement proponent, I suggest to aggressively and steadily decrease your batching. Whatever your current batching is – cut it in half! (i.e. weekly to twice a week; daily to twice a day…) All the problems caused by that change need to be solved and the operation needs to be stabilized. Then the next round of batch reduction can be initiated.
Linking processes
Another cause for buffers can be when linking work centers. It is all fine when you are linking machines/processes with the same performance profile.
If three centers all work
at a rhythm of 10 pieces
per minute, then the parts can go from one workplace to the next workplace/machine with practically
no buffer between.
If they run at different speeds the linked system can only run at the speed of the slowest process. This is more complex when you have workplaces with different running speeds, but all averaging to the same speed. If you do not build some elasticity (buffers) into that system, it will always net out to the slowest element. This becomes more evident when trying to balance an assembly line.
When linking three machines/workplaces into one system, and each individual machine interrupts, for example, five per cent of its operating time (3 min/hour) then linking the machines/workplaces will cause up to 3x3min=9min/hour of error/interruption. In complex machine lines mechanical accumulators are built in. In assembly lines you leave some accumulation space to create this elasticity. If uncoordinated, unequal processes tethered too close together, you will have the same result as in a three-legged race.
SummaryThe Lean manufacturing book is right. Create flow, link processes and eliminate inventory (and all the other waste categories)!
The reduced inventory and the reduced space for buffers is the result of many improvements. Improving maintenance, tooling technology, cross training, scheduling, expediting and overall communication within the operations will bring the results. The more you improve the organization the more benefits you can achieve from a layout change. Just restricting the space and mandating that it must work is most-likely not the best implementation strategy.
There are huge benefits to layout improvement. And while it does not come easily, it is a road worth taking.
