Re: Overhead and the Future

[I've snipped parts to what I wanted to comment on. -PB]

Hawco, Paul R. wrote:

> 2) Sometimes yes, but an engineer's role/tasks hasn't substantially
> changed due to the use of software.
>
> PRH: That is my point - the highest level software has significantly
> changed the future of engineering/design. The new Intelligent P&ID
> software allows you to "draft" the P&ID's, but associated with each
> graphical element is 20-50 database attributes that can support the
> ongoing 3D design, maintenance and operations. The most efficient
> process with this software might mean that the process engineer now
> performs all the tasks and custom tailor's the data to his (or her)
> needs - the line, valve, and equipment lists are now reports from the
> P&ID database. A lot of the high end structural software has analysis
> capabilities built in (or easily linked in) - the quickest design cycle
> would have the engineer creating his own model, analyzing it, making
> changes and signing off on it.

2.1) I already know this, this is a relatively simple part of the entire process. Once the database is built everything else follows. As I mentioned (I don't want to be seen as plugging COADE, since they are a site sponsor, but) doing this sort of thing with non-PDS/PDMS is getting easier and easier. I don't need/want the process department to have to exactly buy a component while I am running pipe and placing inline components. That can be specified and entered later.

>
> * Don't fight the software - new tools often require
> a change in work practices
>
> 4) Sometimes good enough is good enough.
>
> PRH: Unless the new tools don't fit with the old methods - crazy thing
> is that changes in technology will make good methods obsolete sooner and
> sooner. The horsepower is there to perform on the fly stress analysis
> now, we just need someone to write the code and then provide training to
> take people to the next level. At one company the lead stress engineer
> had developed a series of pipe support details that could be customized
> fairly easily using a set of word perfect macros - it worked well for
> his group and had been used for years. We implemented a PDS add-on
> called pipe support modeler that allowed his designers to pick supports
> from a catalog for placement in 3D - the supports could now be
> automatically included on the piping iso's and stress runs - the
> software automatically produced a detail sheet for every support
> including materials, dimensions and views from the 3D model. The
> company was lucky to have someone like him who could see the advantage
> of knowing when it was time to make a change.

4.1) "we just need someone to write the code and then provide training to take people to the next level". Do you know what BRL-CAD is?

> * Checking - the power of the software is in
> automation, the checking process must still be taken seriously -
> checking now requires reviewing "log" files.
>
> 5) Agreed. Please explain reviewing log files.
>
> PRH: Most software you use today will often leave "Log Files" after
> certain tasks have been completed - if you plot in AutoCAD, a log file
> is generated on your system that will contain the details of the plot
> "JOB" - the printer, pen table, scale, ... and whether there were any
> problems printing. Most plant design software uses ISOGEN to produce
> the piping iso's - many people do nothing other than review the iso
> drawing file that is generated - to do the task right you should also
> review the iso "log file" - this file contains the warnings and error
> messages for the run - in most cases things are perfectly fine but if
> you don't check it how can you be sure.

5.1) Oh, that's what you meant. It sounds like a complicated procedure but it's not, actually. Isogen's output files are pretty easy to comprehend and debug. Then again, when I first started using computers for CAD (it was a 286 many years ago) I noticed that on booting the machine would light up the 5.25 floppy's LED before providing the C prompt. I wondered about this, thinking that the computer might be looking for something and asked an "expert". "No, that's not how computers work" was the expert's response. That probably explains my disdain for many computer specialists.

>
> * Failure to Exploit the software - takes a while to
> figure out how to make the best use of all the features
>
>
> 7) After my previous bitching about SPR, I learned a bit more about it
> and used some of the more powerful features. It can be very useful, but
> it's still kludgy and user-hostile.
>
>
>
> PRH: It's one of those things where the more time you spend with it the
> better - I agree that it could be much better.
>
>
>
> * Complicated - it is not AutoCAD, don't treat it
> like it is
>
> 8) I monkeyed around with COADE's 3D piping design program and the
> equipment builder (which are based on AutoCAD) and was able to put
> together piping systems and vessels within a few hours.
>
>
>
> PRH: one model, one designer, no problem - they all work - try
> coordinating the data from 500 models created by 50 designers - that's
> what differentiates the men from the boys in the software world.

8.1) Maybe this 500 model created by 50 designers scenario could be cut down to, say, 10 designers. Some of the projects I've worked on had designated designers who basically sat on their work and waited for others to catch up. In the "old days" of manual drafting I was able to handle multiple areas with complete knowledge of what was going on in my areas and matching areas. That was in the "old days", now it's even easier to do so. Coordinating data should not be a problem for a good admin. Maybe you can clarify this.

>
> * Market Share - software has a solid user base, the
> projects modeled are very valuable and clients are likely to want to
> reuse the data
>
> 9) Which software do you refer to here?
>
> PRH: my point here is simply pick one that has a solid user base - as
> you have pointed out - you are often locked out of the game if you don't
> use the right tools - what I also like to tell management is pick a tool
> that will do the biggest job you anticipate in the next 5 years. By
> aiming high you may spend more but you may have time to gain experience
> with it before the big one comes along. If you aim too low, well you
> just might find that when the big job comes you now have to support 2
> different design tools - the one you picked initially and the one you
> need now.
>
>
>
> * Add-ons - more & more 3rd party add-ons becoming
> available to extend capabilities
>
> 10) For which design programs?
>
> PRH: any one you choose to use - consider Caesar II an add-on if you can
> readily export design data to it - my point is to consider what other
> add-ons are out there when you look for a software solution.
>
>
>
> * New Software - next generation software is tightly
> integrated with standard business products - MS Office, email, web
>
> 13) This is overrated.
>
> PRH: Paul - are YOU really saying that email & the web are OVERRATED -
> that surprises me. Actually I agree with you and I disagree with you.
> For most people in the office life without email and the web would
> probably just cause a slight hiccup - for others it is a lifeline to
> very timely data and information. Going back to the smart P&ID's I
> talked about earlier - you can input the data with in the software or
> export it via EXCEL and edit it in a spread sheet format and then import
> it back into the software - very nice especially if you want to do
> efficient project wide changes. It's all about managing data and I
> find that these capabilities are necessities for me.

13.1) Too much reliance on newfangled programs is a bad thing. I've seen transmitted PDF engineering drawings that are illegible (but the supplier has fulfilled his obligation) and other PDF documents that were sent by a whiz kid using a version of Adobe that had features that were invisible to the end user (unless, of course, the end user had clicked and enabled an obscure command within Acrobat Reader). This is good somehow?

>
> * Checking - the power of the software is in
> automating the design process
>
> 14) As long as the modelers are working from proven designs.
>
> PRH: the automation I talk about is primarily the drafting tasks,
> producing ISO's, GA's, pipe support details and loop drawings can now be
> almost 100% automated (at the highest level with a lot of add-ons) -
> this gives the engineers and designers time to ensure that the they have
> a good design. At a minimum you should use a tool that automates your
> piping iso's. From time to time it is very sweet to be able to
> copy\mirror a unit and have to make only slight modifications.

14.1) You are talking about deliverables (or whatever "drawings" are being called today). Isn't that what engineering companies are supposed to provide? I agree that it's sweet to be able to mirror/copy previous designs. Shortly before CAD slowed everone down, I used to do this mirror/copy thing with a large format Xerox vellum copier that had magnification and reducing capability. It was quite effective and had all the time-saving features of CAD without the overhead.

Look, I *like* CAD, I just think that many people are all wrapped-up in the "coolness" and expense of it all. Your "less people" mantra is particularly relevant.

With Best Regards,

Paul Received on Thu Sep 08 15:27:00 2005