A Visit to the Fendt Cabin Plant in Asbach-Bäumenheim—Part 2

Tractor FendtThis is my second post on the excellent Fendt Cabin Plant in Asbach-Bäumenheim, which impressed me a lot with their efficiency and general organization. In my last post, I looked at the plant and the line in general. In this post I will go deeper into the topic of efficiency, line balancing, and shop floor management. My last post will then look into their good digital dashboard as well as their blue collar–driven continuous improvement process. Read on!

Efficiency

As always, I tried to estimate the percentage of value-added time (i.e., what percent of the time the worker at the assembly line adds value, and what percentage of the time was wasted with walking, waiting, searching, transporting, and so on). The results positively surprised me. According to my estimation, their total value-adding percentage was 56%. While this would put the plant in the middle of all my observations, they had this for a takt time of 9 minutes (540 seconds). The longer the takt time, the more difficult it is to become efficient, and for this takt time, their efficiency sticks out quite a bit. They were also more efficient than their main plant for the tractor assembly in Marktoberdorf. Below is the plot of percentage value-adding time versus takt time, with the Fendt plant in Asbach-Bäumenheim highlighted.

They do have an andon system where they can stop the line in case of problems. If the andon line is pulled, the team leader tries to fix the issue together with the operators. If they cannot fix it, they run the cabin through the line without additional work, or remove the cabin from the line completely. A full line stop is exceedingly rare.

Line Balancing

Line BalancingThe line was very well balanced, and the employees did not waste much time in assembling the cabins. For next year, a re-balancing is planned to adjust the capacity. If the capacity has to increase, workers are added to the main assembly line and selected tasks are pushed to sub-assemblies or suppliers. If the capacity has to decrease, it goes the other way round; people are removed from the main line and tasks are also added from sub-assemblies to the main line. Naturally, this requires a careful balancing of the line.

Shop Floor Meeting

Team Meeting On Shop Floor
Shop floor meeting (not Fendt)

Like any good plant, they had a daily shop floor meeting for each hierarchy level, with higher-ups participating in (selected) lower-level meetings. Like with the Toyota dashboard, the first topic after taking attendance is safety (good in assembly, 800 days without accident at the time of visit), followed by the delivery performance, the buffer size to Marktoberdorf (they surely don’t want to turn off their tractor plant), the number of cabins taken off the assembly line due to problems (rare), general problems, rework (at the time of the visit, the right-first-time performance, i.e. cabins without the need for rework, was a pretty good 94%, above their target of 90%), and finally improvement potentials (see next section).

They used their digital dashboards for these shop floor meetings, and I liked these dashboards quite a lot too. Many dashboards in industry disappoint, but this one looked quite useful. More on this in the next post.

A side note on safety: Every Monday morning between 10:30 and 11:30, the entire plant is doing a safety walk, looking for issues that could be a risk to the employees. That is quite a commitment to safety, and I have not seen something like this in other plants.

Other Observations

VMI Screws Boxes Overview
Screws Boxes (not Fendt)

Overall, the plant also looked clean and well organized. The shop floor also was rather quiet. Material was of course supplied on demand using kanban. For smaller plastic boxes with C-parts, the box contained an RFID chip, and simply dropping the empty box in a RFID enabled container re-ordered the material (the system was dropLOG by Keller&Kalmbach). A lot of this material was using modern digital technology to determine the quantity (e.g., of screws) and to request a resupply if needed.

The group structure was usually ten to twelve operators for one team leader (and hence a bit larger than at Toyota with four to five operators for one team leader).

Most of the assembly was a manual process, with lifting aids only for heavier parts (e.g., the roof), but still controlled by the operator. The only automated process was the gluing of the windows, as a robot has a much more steady hand than an operator for applying the glue.

While the operator is responsible for his quality, there were also some quality gates and a final quality control at the end of the line. Most quality issues are scratches on the product.

In the sub-assembly for the fuse box, they used projection onto the board for the fuses to indicate where a fuse is needed and where not. Previously, they just installed fuses in all slots, regardless whether they were needed or not, but this took a lot of time and consumed a lot of fuses. The new projection makes it easy for the operator to see where a fuse needs to be added and where not.

Fendt LogoIn my next and last post on this series on the Fendt Cabin Plant in Asbach-Bäumenheim, I will dive deeper into their excellent digital dashboards, as well as their shop floor–driven continuous improvement process. Now, go out, look for waste, unevenness, and overburden on your shop floor to improve your efficiency, and organize your industry!

PS: Many thanks to Tarik Kadrispahić from Targer Targer Engineering & Consulting for inviting me to come along on his excellent tour of Fendt and other plants!


Discover more from AllAboutLean.com

Subscribe to get the latest posts sent to your email.

1 thought on “A Visit to the Fendt Cabin Plant in Asbach-Bäumenheim—Part 2”

  1. Thank you for these great blogs and your books. Your work, knowledge, insights, and sharing are much appreciated and enjoyed.
    -Duane

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Cookie Consent with Real Cookie Banner