Hello to the community!

I'm new to the FME and I'm quite impressed with what it can do and also with the very helpful tips and solutions which I read in this forum!

As a newbee, I'm struggling to get the result which I want, but for most of the experts here, it's probably quite simple:

I'm working with polygons which represent the chart coverages of Electrionic Navigational Charts. They are categorized into "usages" with usage 6 having the largest scale (inside harbours) and usage 1 having the smallest scale (covering entire oceans). The set of polygons is currently overlapping each other and my goal is to have a non-overlapping set of polygons - perfectly touching each other's boundaries - which give me the best chart usage at any given point.

I have already created a workflow where I dissolved usages 2 to 6 and then used the Clipper to clip usage 1 with this. Similarly, I dissolved usage 3 to 6 and clipped usage 2 with it - and so forth. Doing it this way seems to get me there but my guess is that the DonutBuilder could do it much quicker? Also, I would like to make use of the AreaGapAndOverlapCleaner to perfect the result.

However, I don't want to mess with the original data too much since it will be important for the subsequent processes (which will use this clipped set of polygons), whether or not a particular feature is indeed covered by that particular usage. This mapping needs to be rather accurate so I only like to get rid of polygons which, after the clipping process, are remains and practically a line only.

The result should also retain the original name of the chart polygon (cell_name) and not dissolved with other polygons of the same usage. I hope you can help me to set up an efficient process that can do the job.

I'll attach a few polygons from all usages so that you have something to work with. Also, I attached a few graphics.

Thanks in advance for your help already! :-)