SP3D Modelling - How long does it take for a support?

[RobertvdM]

Hi all,

So I am doing a bit of a feasibility study on different Plant Design types of software. Does anyone have an experience how long the modelling of a cable tray support (typical) and then the placing of that would take?

I would say there's three parameters:
1. Creating the typical (x minutes)
2. Placing the typical in the main environment (y minutes / support)
3. Fixing any clashes or redesigning the typical supports (z minutes)

Anyone have an experience here or can point me in the right direction?

Much appreciated!

Thanks,

Robert

 

[3DDave]

In most CAD, inept users will cost more time than anything else. Good users will automate any typical task to the point it takes only a few seconds and won't require fixing clashes.

The difference in productivity between the two is good users are about 100 to 1000 times faster at the sub-tasks. They still lose a good amount of time getting information from other people so the throughput might only be 10-50 times greater.

I suppose some software will have cable tray design to the point that the software will autoplace it and all the supports. For example, I saw a company that wrote their own to automate mobile crane design - it placed all the gussets, produced all the assembly drawings, created nested cutting paths within a few minutes based on a few initial parameters.

 

[RobertvdM]

Hi Dave,

Thanks for that. Yes I agree with you, there's obviously a huge difference between novices and experienced 3D Modellers. But supposing you're an intermediate user, and you have the parts in there (eg angle profile, gussets, whatever) but not the autoplacement, any wild guess you can take in how many minutes it would take for those three parameters?


Thanks,

Robert

 

[3DDave]

You'll have to start with the workers you have access to and be on the lookout for the faster ones. There's no average person until one is looking at employing thousands to tens of thousands.

I'd recommend "The Mythical Man Month" to get an idea of how to understand the problems in trying to put numbers to performance.

Also, I wasn't talking about novice modelers. I've dealt with incompetent ones that had been working for a decade at the task and never cared to learn to do better or were incapable of ever doing better.