Has the engineering process changed? 5

[Blackcountryman]

The 45 years since I started my apprenticeship as an electrical engineer have seen continuious change. Most of the products I worked with then are museum pieces, even the tools have changed from a drawing board & slide rule to CAD and CAE tools.

My question is has the basic process changed, I suggest that should the likes of Brindley, Telford, Stephenson or Brunnel come back they would recognise the same process.

A need is identified, engineering look at and even with new processes are asked to give cost and time estimates. The money men, with no real knowledge say it should cost half the price be a quarter the size and only need 10% of the time.

Engineering sit down with a blank sheet of paper and start from scratch. I admit that I talk about reusing previous designs but I still do a lot of rework.

At the end the project is always percieved as late, over budget and only meets 80% of the requirement.

There are good examples about, the current T5 project at Heathrow I believe is one, some automotive projects etc but why are they so few?

How should we operate?

Discuss, or tell me I'm an old cynic

 

[zdas04]

Today, we can "solve" more complex parts of problems than in the slide-rule days. Back then we used a lot of simplifying assumptions that always had extra conservatism built in. Today we feel like we can model reality a lot closer so we cut down on the safety factors (because we believe that we've reduced total uncertainty).

At the bottom line (at least in pipeline construction) 95% of what we do would be very recognizable to an engineer from the '50's or earlier. The other 5% would probably elicit the comment "why bother? You come up with something between S30 and S40 pipe, and pick S40. I would have calculated just under S40 and picked S40 with a lot less work".

I'm not saying that the models, spreadsheets, and general purpose computational engines don't allow us to generally do the middle 95% a bit faster with fewer opportunities for error, but it hasn't been revolutionary. If I run the AGA fluid-flow calculation on a slide rule or run it in MathCad, I should get the same answer within the accuracy of the assumptions and limitations of the equation. The problem today, just like 50 years ago is understanding those assumptions and limitations--we don't do any better or much worse now than back then.

I can be more certain of a measurment with a digital micrometer than with a ruler, but not necessarily better or faster than an anolog micrometer.

David

"If its on Green-and-White, its right" mantra of computer professionals in the '70's.

 

[Ashereng]

I don't think the basic process has changed. Engineering is still engineeering.

In your 45 years, what has changed is knowledge. We hopefully now know more than we did 40 years ago, collectively as a profession. What people once had to figure out for the first time, we now have the benefit of our predecessors' hard work and discoveries. All the mistakes, tragedies, and so forth have added immensely to our collective understanding of what to do, how to do it, why we do it, and when we do it.

The tools of our trade, I would also put under knowledge. Using a slide rule vs. a computer/calculator is knowledge. The computer knows the answer, but the answer probably still came from the guy with the slide rule.

Our current stature is a result of out standing on the shoulders of those who came before us.

Should Brindley, Telford, Stephenson or Brunnel come back, they would recognise the same process. But, hopefully, they would be heartened by the amount of progress we have made.

 

[davefitz]

I think the process has incorporated some significant changes . The most obvious one is the universal reliance on computer codes for which the engineer has not developed himself- he does not know the limitations of the engineering model and the numerical methods used in constructing the computer program, and takes on faith that results that are output are valid. The results of at least one computer program is built into every engineering product that meets the marketplace today.

45 years ago, the problems were simplified to the extent that they could be solved to 3 place accuracy on a slide rule and interpolated from some curves drawn thru some test data points- every engineering decision could be traced back to a fundamental calculation that could be verified by the responsible engineer. Not so today- the amount of blind faith used seems to be directly proportional to the amount of hi-tech utilized.

 

[Ashereng]

In years gone by, like during my university days, we used a whole buncn of correlations, empiracal data/charts and the like. I didn't develop them myself. In most instances, I did not verify that they were accurate - and would not know how to verify them (other than versus another set of data). I and my peers took it for granted that the author did his due diligence, and that the published data was accurate. The CRC handbook was a bible that we all used - and still used today. I still rely on its data and graphes.

I see that as being similar to relying on computer codes.

 

[DRC1]

At least for for heavy civil it has changed fundamentaly. At one time the structure was the product. Now the process is more important than the product. At one time the designer would devlop a complete and buildable design, generally by himself and a few assitants. Now specifications are thick but generic and no one checks to see if tey are applicable to the work. the drawings are numerous but conceptual, reling on shop and " working" drawings from the contractors or their vendors for final details. Designers have less and less experience with actual detailed design or the details of construction or construction costs. The promise of cadd and internet collaberation has not been realized. Scheduling programs that can ignore the details of construction develop unrealistic finish dates for the owners. The result has been the increase in litigation.
Where once a structure was designed and supervised by an engineer with a vision, we now have camels built by committees charged with building horsesa

 

[GregLocock]

While it is fair to say that many engineers use computer codes that are essentially black boxes, I don't think that is anything new.

When I look up a value in a book (say the Young's modulus of Tantalum), I am treating that book as a black box. When I use a nomogram to size a pipe, that is definitely a black box - I may have checked the nomogram once or twice, it is very unlikely that I can be 100% confident that the shape of each curve is correct.

In every case one relies on the intuition, knowledge and integrity of the engineer. If the answer doesn't 'feel' right, he has to have the ability to check it, or else to massively overdesign, I suppose. That hasn't changed in 200 years.

Another thing that hasn't changed is gut feel - a system designed by an engineer using whatever tools will always be better if he has a good idea of where he wants to go - the proverbial blank sheet of paper is not only intimidating, it is quite likely to mislead you.

Here's a war story: the first succesful solar car was built by Larry perkins and Hans Tholstrup, which they used to cross Australia some time in the 70s or 80s.

Hans then went on to set up the World Solar Challenge. Nobody could afford to build a series of prototypes, so each team sat down and designed a car from what was essentially a blank piece of paper.

One of the most important things in designing a solar car is deciding what speed it will go at, as this controls the trade off between weight, aerodynamics, battery size and panel efficiency.

One of the outcomes of that first race is that almost every serious car achieved the speed it was designed for, and the team that designed for the highest speed, won. In other words, the design was optimised around the initial assumptions the teams made.

The interesting thing is that the technology in the winning car was pretty good, and for no appreciable additional design effort they could have gone 75 kph instead of 60 odd, but they'd optimised around 60 odd. (I'm not bagging them, it was a fine car that pretty much nailed the correct solution).







Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[RDK]

Has the process changed?

Nope it is still the same old process.

Has the technology used in the process changed?

Yes it has changed greatly. It now allows competent engineers to accomplish much more in much shorter time and present the results in a professional manner.

Unfortunately the technology has also allowed incompetent engineers to make many more mistakes, make greater mistakes and to make bigger and worse mistakes in a shorter time while still allowing them to present these mistakes in a professional manner.


Rick Kitson MBA P.Eng

Construction Project Management
From conception to completion

http://www.kitsonengineering.com

 

[ScottyUK]

The industry of today seems to allow many more individuals to hide their ineptitude and lack of talent behind the facade of a corporation, like passengers along for the ride to retirement. Sadly the days of the great visionaries who could almost single-handedly design a bridge or a locomotive have gone. I do not think that the likes of Brunel, the Stephensons, Charles Parsons, William Armstrong, William Stanier, Nigel Gresley would be comfortable with the endless layers of hangers-on and middle management bureaucracy which stifles our profession today. I am however sure that they would recognise our profession and could no doubt contribute to it.

----------------------------------

Your body might be a temple. Mine is an amusement park...

 

[arunmrao]

In our Clark's tables and Sliderule era,approximations and assumptions were the order of the day. Large computations would take weeks and yet there was an element of uncertainity. Today with modern tools(I am illiterate about these) elegant models are made, but skills of the engineers is lowered.Many issues are compartmentalised and there is a lack of coordination between different elements. Also what is disappointing is very often the physical picture is lost or unrealistic conclusions are drawn.

An engineer from the old school approaches the problem pragmatically and with limited tools provides a solution.

I do not mean to disregard the moden tools,but the users and intepreters of the models need better mentoring.

 

[Blackcountryman]

Arunmroe said that many issues are compartmentalised, I think this is one area where things are getting better.

If we look back the original segregation was between Milatary & Civil Engineering (Not the construction version of civils we have now but a wider scope) then we started to divide to many narrow disciplines.

I think we are starting to look at the wider picture again. It is less acceptable in the company I work for to say that is Mechanical or Software and not my resposibility. Yes I will be expected to work with engineers of other disciplines but I have to own the problem not walk away from it.

John

 

[Jordonlaw]

I think the process has changed with the advent of computer simulations. 45 years ago a new or innovative design usually had to be built and tested at great expense before it was verified and then put into production. Now you can simulate new designs in various loading environments and use comparisons against existing and sucessful designs to see how much better they might be before you build them. Adding in design of experiments software allows you to optimise to a solution much quicker than would otherwise be the case.

OK so you still have to tests things to validate new designs but with greater confidence you'll get it right quicker and at a lower cost to all concerned.

Want a good example ? How about stents used in heart surgery. These are now designed using FEA such that they work better in the human body. How do you think you could do that without much trial and error; doesn't bear thinking about really.

 

[RDK]

The technology has only helped the process not fundamentally changed anything.

We make some assumptions then test these assumptions in models. We just have the ability to have computer models rather than doing all the theoretical modeling by pen and paper and we also have the ability to do calculations more accurately and more ability to have the virtual models refined before moving on to physical models and then the actual product.


Skill, knowledge and innovation are still required just the drudge work has been made a lot easier in the process. The skills may be different, the amount of knowledge may be greater but the innovation factor still is important.






Rick Kitson MBA P.Eng

Construction Project Management
From conception to completion

http://www.kitsonengineering.com

 

[DrillerNic]

I agree- I don't think the process has changed, merely the tools. And these tools have allowed us to become more efficient- reducing the 'fudge factors' and so on it equations for example. This can have some unfortunate effects if the original design conditions turn out to be optimistic- there's less fat in the design to allow for errors!

I'd also disgree that we don't have great individuals like Brunel etc in engineering today. They simply aren't generally working in the same areas of engineering as Brunel etc but in newer areas where the rules haven't been codified like sofeware or computers (if Brunel was alive today I doubt he'd be working in civil engineering but in IT)! Or they are 'mavericks' challenging exisitng technology: Trevor Bayliss with the clockwork radio and James Dyson and his vacuum cleaner for example.

 

[ziggi]

When I started out in engineering back in the ol' days of 2003 . My mentor was actually asked to come out of retirement to manage the engineering office, he had done so much in his life that an entire wall of his office and it was a big office, was covered in certificates, awards, degrees etc. The most important thing I learnt from him was, "When in doubt go with the more conservative number and when in a hurry go with the more conservative number".

I am quite puzzled when young engineers who are more or less my age take ages to get the exact answer and eventually end up coming to the same conclusion I did, except that it took me far less time and a bit of conservatism.

 

[HgTX]

Eventually they learn that the whole "exact answer" thing is overrated.

Easy for me to say; I have a job that can be done with a 4-banger calculator. Possible answers are "Looks good", "Doesn't look good", and "Better ask the guys that have the real numbers".

'Course, the guys who have the real numbers vary as to whether they are willing to eyeball something and say "Looks good", or spend a week crunching numbers to confirm what my gut already told me (but that I felt ethically compelled to confirm with them). And to be honest, the more of a number-cruncher the EOR is, the more likely I am to decide to trust my own gut after all.

Regarding the Great Engineers who saw a whole project through from start to finish--these days the "greats" wouldn't want to be bothered with it. The big idea people don't want to be general contractors, hiring the cranes, etc.--and why should they? How does that make them better big idea people?

For example, last I knew, the city of Dallas was commissioning a series of bridges from Santiago Calatrava. Ol' Santiago makes a heckuva bridge, but somehow I doubt we'll see him hanging around in a hardhat for the full duration of the project. Or doing all the detail drawings himself. Does that make him less of an engineer/architect?

Hg

Eng-Tips policies: faq731-376

 

[ScottyUK]

Does that make him less of an engineer/architect?

Compared to the likes of Brunel (Royal Albert Bridge), John Roebling (Brooklyn Bridge), or John Fowler & Benjamin Baker (Forth Rail Bridge), yes, Calatrava is a lesser engineer, although perhaps a greater artist.

I defend that opinion because I think the creations of the engineers of yesteryear, with the tools and materials at their disposal at the time, were far greater challenges than those of today. Calatrava's talent lies in combining purpose with graceful elegance, and his work has a fragile beauty. The Forth Rail Bridge has an appeal all of its own but fragile beauty is hardly the term to describe it. Like the Royal Albert Bridge it has tremendous character and a sense of permanence lacking in modern structures. Perhaps I am sentimental for a bygone age?


----------------------------------

Your body might be a temple. Mine is an amusement park...

 

[HgTX]

You're getting awfully subjective. Does "sense of permanence" necessarily trump "fragile beauty" in terms of pushing the edge of convention, and of what technology and scientific knowledge allow us to do? Besides, my point wasn't really about tremendous character vs. fragile beauty but an earlier point in the thread about ownership of a project from start to finish. How does taking on what are now considered general contractor duties make one more of a Great Engineer?

Hg

Eng-Tips policies: faq731-376

 

[Ashereng]

ScottyUK,

I would vote for you being sentimental for a bygone era.

Nothing wrong with that.

I just think you are sentimental.

 

[macmet]

I have been watching this thread progress and I keep wondering. What's wrong with the engineering process changing? Is that not what is otherwise known as evolving?