2

I was looking to merge some features to improve the performance of some of my layers on my ArcGIS Server. I have a few very large polyline layers for our fiber network that are exported out of a CAD based program (Mapcom M4) every week and into a GeoDatabase. I did notice that some of the features have the same attributes and fiber ID's, but are separated into up to between 4-15 different lines. When I tested this out and manually dissolved the featueres into one based on fiber ID it drastically improved performance. This would be a significant expense of time to do every week, though, and I was looking to make it part of my Weekly FME Workbench. I just wasn't sure what transformer would make the most sense to use to merge all the features with the same Fiber ID before writing them to the Geodatabase.

  • FeatureMerger or Aggregator are the two I'd look at to start with. Which transformers have you looked at? – Midavalo Jan 20 '16 at 1:22
2

Aggregator will merge geometries based on a specified attribute. If you aren't concerned with any other attributes you can tell it not to keep other attributes, otherwise it can concatenate/sum/average those too.

enter image description here

  • OK I am trying out Aggregator now. It looks like what I needed! thanks – GravitaZ Jan 20 '16 at 19:26
  • Hmm, According to the workbench it seemed to work, but when looking at the actual tables it doesn't look like it did. I think I am using an older version, but does it look like I have the proper settings? I wanted to aggregate the ID's but keep the other attributes. i.imgur.com/6Fj8oQr.png i.imgur.com/xYKvJSB.png – GravitaZ Jan 21 '16 at 5:28
  • do the attributes differ across the different features (of the same ID)? You could possibly pull the attributes into your new dataset using a different transformer – Midavalo Jan 21 '16 at 7:22
  • Thank you! The Aggregator worked. I just had to set the Accumulate Attributes to "Yes" and Concatenate all the attributes besides ID. Thanks! – GravitaZ Jan 25 '16 at 21:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.