Navisworks Clash Detective Status Reports

[advandes]

Hello, could someone please explain to me how and if Navisworks Manager automatically determines if a perviously identified clash becomes "Resolved"? Is this something Navisworks can do by somehow comparing 2,3,4.... iterations from updated Revit models? Or is  this just simply a manually input selection by the clash detection coordinator? Thank you in advance for setting me staright on this! Aaron

[dgorsman · Accepted solution]

Resolved clashes are determined by the unique IDs of previously clashing objects.  In some cases, like with AutoCAD models, the unique IDs are regenerated each time the model is read so "closing the loop" isn't possible.  I'm not a Revit guy but I *think* unique IDs are imported and preserved.

[advandes]

Thanks for the response. I'm not exactly sure what you're referring to, but it's also not clear to me how to get the navisworks models to identify fixed clashes. Thanks again for your response.

[IanBadcoe]

Hi, NW will mark a clash-result as resolved if: 1) you re-run the test on changed models 2) the clashing items do not clash any more, this can be either: 2a) at least one item has been moved so it no-longer clashes 2b) at least one item has disappeared The only complexity, which dgorsman alluded to is that if items in the model change their identity from one version to the next, then NW has no way of supposing that the new item is supposed to be a new version of the old one. In this case NW will see that the original clash-result has disappeared (because at least one items involved in it doesn't exist) but that a new clash has appeared featuring the "new" item (effectively created by changing the id on the old one...) One thing I would not suggest is manually setting "resolved" status if you plan to re-run the clash-test. This is because NW expects "resolved" to mean really resolved in the model geometry. If you set a clash-result resolved and then re-run, and NW sees that the result is not resolved, then it will re-open your clash (this is covering the case where a previously fixed result comes back with a new version of the model). If you wish to mark a result as having been accepted by a part of your process, then "accepted" or "reviewed" are the statuses to use... I hope this helps. Ian

[advandes]

Ian,

Thank you for your clarification. My past experience with NW is to export a dwfx file from within Revit, but because I have not had proper training on NW Manage I am thinking that in order to do the "managing of clashes" as we have been talking about, there must be a different procedure. If what I am saying makes sense, can you please comment, and let me know what this procedure should be? Thank you!!! Aaron (Advance Design; AdvanDes)

[IanBadcoe]

Hi Aaron, The only problems with exporting a DWFX are: 1) that it may not represent the data quite as perfectly as the original RVT file (just because there has been file conversion) 2) that each newly expected DWFX may have a different identity. This is because the export utility cannot really know the user's intent. "Export" may mean duplicating the original data (keeping the same identities) or it may be intended to export to a completely different situation (in which case new identities are absolutely required) For these reasons we recommend either reading the RVT directly into NW, or else exporting it to an NWC via the NW Revit Exporter plugin (which does pretty much the same thing as NW does internally...) Regards, Ian

[dgorsman]

Exactly.  Preserving those unique ID's lets you run the clash tests over and over as the models are changed, and each clash is auto-marked as "Resolved" when no longer present.  Then you have a history of not only where the problems are now, but in the history of the models as well.  Clash detection is still possible without the unique IDs but it's less of a loop and more doing the tests from the start each time so there isn't any real history aside from any exported clash reports.

Speaking of which, I'll have to remember to submit something for this in time for the next AutoCAD BETA cycle.  If we can get that unique ID handled in the core product it should propagate out to all the verticals as well.

[advandes]

Ok, I think I got it! I created a NWC file from the Revit Add-Ins>External Tools> NW 2015. Opened up the file in NWM and Created a Test clash and found 2 clashes. I then went back into revit and fixed the clashes. I created another NWC file (same name and overwrote the first one). Went back into NWM and could see that there was a warning symbol on my first clash test. I ran the test again...and voila, the two previous clashes came up "automatically" as Resolved...Cool! Thanks to all you guys who helped me out with this... Much appreciated!

[advandes]

I believe the key for this procedure to function correctly, is that the subsequent NW files have to all be named with the same name so that they will read from a Navisworks File Set file that was initially created. Please correct me if I am wrong on this. Thanks, Aaron

[IanBadcoe · Accepted solution]

Hi, Yes, it has to be a new version of the same file. This assumption is built into the way an NWF file works, you have an NWF that references other files: MyBuilding.nwf +---> foundations.dwg,v1 +---> architecture.rvt,v1 +---> plantings.skp,v1 etc... and you set up your clash tests in that. When new versions of files come in, they are expected to be overwriting the original versions: MyBuilding.nwf +---> foundations.dwg,v3 +---> architecture.rvt,v2 +---> plantings.skp,v5 And the NW file update mechanism now understands to do the upgrade, re-convert the file; and updates to any other data, such as the clash-tests, flow from that. There is no way to read in a file with a different name and tell NW that it is actually the as one we already had. If you want to call versions of the same file by different names, that sounds like you want to keep historical files as part of a full version history. A data management system will do that better, and they generally also allow any archived version of a file to be fetched into the same physical location. Ian

[IanBadcoe]

Agh, sorry. I drew little ascii diagrams in that and forgot that the forum would trash them. Hopefully there is still enough for you to get the idea. Ian