1. Production in a Matrix as the Concept of the Future
The factory of the future is facing great challenges: will individualization continue to increase and make batch size 1 necessary in all product areas? Or will there be more uniform hardware products again and individualization be exclusively on the software side? Which products will still be available for sale in 3-5 years and what will have to be replaced by other products – possibly by force?
Whoever builds or expands a factory today would be wise to build in the necessary flexibility. Are rigid assembly lines, as has often been the case in the past, the right answer? I am firmly convinced that there will still be products in the future that can only be manufactured with maximum efficiency on an assembly line. However, I am equally convinced that some production processes will be subject to such strong fluctuations in variants and volumes that flexible cell production will become indispensable. The special thing about this is that assembly takes place in flexible cells, which are arranged in a matrix. The products move through the cells in an optimal sequence for the specific product order.
Each product takes its individual path through the factory. The cell duration time is calculated individually by the algorithm.
If the digital twin for assembly and production logistics is comprehensively set up, flexible cell production allows virtually any product to be assembled, even in batch size 1. Assembly and logistics are then optimally orchestrated by a control algorithm so that each worker is always assigned the corresponding product orders. The worker plays a decisive role here, as he or she has much more responsibility, which ensures exciting and varied jobs. An above-average worker qualification is essential for this process. If robots take over the manual activities of humans in the future, this qualification is easier to transfer. In any case, the use of human-robot collaborations and the more flexible use of complex and expensive automation technology offers further potential for optimization in flexible cell production. For example, a special application can be used at various points in the assembly priority graph by approaching the cell equipped in this way several times by the same product order. In the classic line, this application would probably not be automated because the investment would be necessary at several stations.
2. Concrete Study Results and Practical Experience
Together with the Boston Consulting Group, we have demonstrated for the first time with real data what flexible cell production ("flexCell") actually achieves in a study  that has attracted worldwide attention. We optimized a final automotive assembly line with 200 vehicles per shift both in the assembly line and in flexCell. The results are surprising: with 19% fewer workers we can fulfill exactly the same production program in flexCell as in the assembly line! This is mainly due to the significantly reduced cycle time sequence losses, which at some point cannot be further optimized due to the high process time spread in the line. Even the higher logistical effort (many parts are delivered to the cell picked in the goods basket), a higher space requirement, increased lead time and a higher WIP (work in progress) cannot reduce the advantage for the flexCell in the simulated product mix.
The supply of components, tools and the products themselves is mainly carried out by means of automatic guided vehicles (AGVs)!
3. Effects on the Factory
There are also classic factory planning tasks in flexible cell production. However, experienced factory planners should rethink some points when they approach flexCell planning:
- Simulation and layout planning in iterative loops produce an increasingly detailed picture.
- It is essential to include buffer areas in the layout, as the products between the cells may have to wait some time to achieve an overall optimum.
- There are no fixed cycle times and fixed station sequences in the flexCell approach. Flexibility is therefore also required in factory planning and control.
On the other hand, many changes can be made simply by (de)activating or converting individual cells, where much more effort would be required in the classic line:
- Increase / reduction of the number of pieces
- Shortened planning horizon / demand fluctuations (hourly, daily, ...)
- fewer / more assembly / manufacturing steps
- Changes in processes / manufacturing methods
- Integration of an additional product / discontinuation of a product
In brownfield projects, the flexCell approach offers enormous potential and design options for the factory planner and optimizer. Here, it is particularly exciting to see which costs are no longer incurred due to reuse in the existing line, which may not be able to be reused in flexible cell production. It is possible that a much higher price will then have to be paid for the greater flexibility.
GThe situation is quite different in greenfield projects, where flexible cell production places much lower demands on the factory floor (a flat floor is sufficient, lower height is sufficient, etc.). In addition, potential expansion space for additional cells can easily be provided on a greenfield site, which can then be realized by adding an annex at a later date. Publication on the subject of "Flexible cell production": : https://www.bcg.com/de-de/publications/2018/flexible-cell-manufacturing-revolutionize-carmaking.aspx
You may also be interested in:
About the Software
The ipolog software creates an integrated model of your assembly, production and intralogistics. Effects become visible and potential can be developed so that collaborative planning is possible across departmental boundaries.