Is the paranoia over making simulation data secure justified?

[ParabolicTet]

In my opinion, company's spend too much time trying to make simulation data secure. I never understood why. If someone is smart enough to reverse engineer simulation data, then they can just do the simulation themselves!

I think a lot of engineers have inflated egos and think their simulations are some top-secret thing. They think a competitor would make millions if they got their hands on them. In reality, the simulations probably won't make any sense to anyone but the analyst who built them.

Also nowadays there is a strong open-source/community approach to software. Most of these top-secret simulations likely made extensive use of open source software ( Linux, open-source code, etc ). So why feel the urge to spend loads of resources trying to protect that data?

I am curious what others think. If I were CEO of a large company I would just make all engineering simulation "open-source"!

 

[JohnRBaker]

From my experience, for too many years the analyst community was like a priesthood where there were dark secrets and 'holy' incantations. I suspect that it's from this that these feelings that the data is somehow 'sacred' comes from.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[GregLocock]

Your company would be taken to the cleaners by lawyers. Same as with test data. Any failure that occurred in the field that could be linked back to a sim or test result that was ignored or explained away or superseded or interpreted differently would be a minefield.

You are also giving your competitors insight into your methods, targets, performance, and proprietary knowledge.




Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[ParabolicTet]

I think all simulation software comes with big legal disclaimers like this:

" The software is provided in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the
implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. "

Can't the same disclaimers apply to any products designed with these software?

Regarding proprietary knowledge, I wonder how much really provides a competitive advantage? I think most engineers subscribe to the Hollywood vision that engineering is all about "break-through" ideas. The reality is that most engineers are just making incremental changes to time-tested designs. The "break-throughs" are rare.

Also execution is far more important than fancy ideas. As they say, "talk is cheap". There are tons of Chinese companies that copy other designs. But none of these Chinese companies have any clout. No serious customer would ever buy from them since they have no end-end solution. They just provide one-off designs/parts.

 

[drawoh]

ParabolicTet,

This airplane is provided in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Would you get on it?

--
JHG

 

[monkeydog]

Are you talking about Jet Blue?

 

[gravityandinertia]

Gaining accurate material data for simulations is also an expensive endeavor, particularly S-N curves for fatigue with enough samples for a proper statistical variability to be calculated. If a company has spent hundreds of thousands or millions of dollars of the years accurately characterizing their materials and someone steals that, that's a big boost to the company for free. Of course, it's only useful data if they know the suppliers of those materials as even among different suppliers certain material properties vary.

 

[CWB1]

Companies spend from thousands to millions over months to years running testing to correlate a single simulation with reality. Why would they want to save competitors that capital and time investment by giving them a good model? Sure, there's not much proprietary or earth shattering about using ANSYS to run a basic structural or fluids problem, but many of us are years beyond commercially available software's capability.

 

[GregLocock]

A simple example would be the tire models we use in rollover simulations. Each tire test costs about $20000. Why would we give that data to our competitors?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[IRstuff]

Seems to me that simulation "data," by itself is just a bunch of numbers and hash. Without context and understanding of why something was simulated and for what reason, it's pretty much just noise. While a company might be willing to publish such results because it costs them nothing to just dump it on the web, spending calories on a professionally understandable writeup, with input parameters and caveats explained might be significantly cost negative, particularly if nimrods call up asking for clarification without paying for the engineering time.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm