Are we becoming "sloppy" Engineers 12

[PSE]

One of the most (or perhaps the most significant) tools that has been placed at the engineer is the computer. This tool has become increasingly powerful over the years. I wonder though, if this tool is making sloppy engineers out of us. With ever more intricate algorithms, the capability of the engineer to independently assess the results of an analysis I feel is decreasing. At the same time, computer simulations are bringing products to market with fewer design iterations and even fewer prototypes. I thought to float the following questions for discussion.

Are we becoming overly reliant on a device and accompanying software that uses algorithms too complex to analyze, written by individuals who may or may not have any familiarity with the underlying engineering principles? How many bugs are still within such programs that could affect results? In recent memory, I cannot think of a software program or operating system that was bug free.

I haven't marked this post but will visit it regularly (I get enough e-mails). Interested in your general thoughts.

For a humorous close, from some of "Murphy's Laws"

Undetectable errors are infinite in variety in contrast to detectable errors which by definition are limited.

If builders built buildings the way programmers wrote programs, the first woodpecker to come along would destroy civilization.

There is always one more bug.

Never program and drink beer at the same time.

Regards,

 

[zdas04]

I went through Nuclear Power School in the early '70's with a slide rule. A couple of years later I had the opportunity to re-visit Vallejo and saw that all the Nuc students had HP calculators hanging where a slide rule should be. I asked an instructor what the difference was and he said "the calculator lets us make the math more involved, the concepts didn't change at all". He was very serious and I beleived him.

I use MathCad and Excel a lot and they "just allow the math to be more involved". When I use a simulator (either downhole, pipeline, or compressor) I always verify the results against both observed data and hand calculations. With a calibrated model I have some degree of confidence in the extrapolations that are generally the reason for the simulation.

I had an intern a few years ago that was working on a de-bottlenecking project on a gathering system. The system had two major laterals that came together into a trunk. The design he brought me was a full-system loop of both laterals. I asked "why don't you just start the loop at the head of the north lateral, jump down to the south and loop that one?". He said that he model said that was a terrible solution. I modeled it in another program and that one said my way was significantly better. Looking at his model showed that the software had a bias to flow gas from low numbered model-nodes to high numbered nodes - I re-numbered the nodes in his model and it liked my solution. The intern lacked the "engineering judgement" (whatever that is) to question the model. I think he learned a lesson that will carry him through a career.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

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

The Plural of "anecdote" is not "data"

 

[Zapster]

I am concerned about technology in the classrooms, specifically, computers. Educators are looking at technology to ease their responsibility to make kids literate. I am seeing some schools with every student being stationed at their own computer in grade school. I do not equate the ability to use a computer as a mark for how literate a person is. In contrast, a literate person can read a book teach themselves how to operate a computer. All too often, a person who is savvy with the latest Microsoft GUI is questionably literate.

In the public school system, I would prefer that computers were not introduced until high school. I just don’t see the advantage of computers in the lower grades. In fact, I don’t see the advantage of calculators in the lower grades. I took an upper level course in differential equations in the mid 90’s. The instructor was excellent. He stated at the beginning of the semester that a calculator was not required for exams in his class. I took his advice and did not use one for the exams. Surprisingly, as I finished up my first exam, I saw many classmates frantically punching buttons on their calculators without thinking about the algorithm required for a efficient solution. Given a computer with the appropriate software for engineering, these same button pushers can be dangerous.

In conclusion, I believe that computers have their place, however, that place is not in most classrooms. In most engineering classes, there is little benefit in using computers to make students literate in engineering. Obviously, there should be courses specifically for programming and using computers with numerical methods to solve complex problems. In contrast, computers are of limited benefit for most undergraduate work. Undergraduate studies should focus on the basics and the historical methods of calculations. If the graduate is properly educated, using a computer is a no brainer.

 

[CajunCenturion]

Another way to put is perhaps with the following question:

Would this engineer still be an engineer after the power went out, and the battery ran down?

Good Luck
--------------
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[crossframe]

I had a strength of materials professor that posed nearly the same idea. He said he had a nightmare that all the computers and calculators stopped working, and engineering ground to a halt. No one could do logarithims anymore. We all nervously laughed as this hit a little too close to home...

 

[epoisses]

I've had worse nightmares than that... The DCS system went black and the operators had no idea how to shut the refinery down without any process data in front of them...

 

[Tobalcane]

I do have to admit that the latest generation of engineers (past 10 years) are computer dependent. My site would have power outages every once and a while during the hotter summer days and in middle of big snow storms. When it happens, we come to a “slamming in to a brick wall” stop. During those black out moments, while sitting in the glow of emergency lights, I do think wow if engineers were with out computers, we would be in trouble. Not so much in knowledge, but in speed of completing a project.

Every now and then I come across circa 60s and 70s photos of engineers with their draftsmen in a sea of drafting tables and wondered what was the speed back then on completing a project. Just to do engineering change orders from the start of an ECO thru CCB and then to the draftsmen for completion must have been long.


Go Mechanical Engineering
Tobalcane

 

[epoisses]

Guess the speed must have been about the same (after all, the speed is a function of well-informed or ill-informed management decisions) and I think the quality generally wasn't less than today's, so I guess the difference must have been in the cost (corrected for inflation) because of the conservatism that was thrown in as certain things could not be calculated very accurately.

 

[btrueblood]

Hmmm. I worked in a shop that had not yet converted to CAD as my first job out of school. We had the "sweat shop" of drafting tables. Generally, the productivity of the manual drafting was a lot lower (in terms of drawings per man-hour) than nowadays. But, because several drafters would be working on various parts of a machine simultaneously, there was a lot more back-and-forth discussion, involving engineers & drafters. Nowadays, a lot of my work goes unreviewed to the shop floor (this is a much smaller company though), essentially I think there is less time for and therefore less thoroughness in review of drawings/designs than before. Yeah we're quicker, but...
Like the sign in our shop says: "Fast, Cheap, Good. Pick any two".

 

[Tobalcane]

<<< Nowadays, a lot of my work goes unreviewed to the shop floor>>>

This is not a technology problem, but a process engineering problem. If your work is getting to the floor with out even going thru a configuration control board (CCB) for approval, it does not make a difference if the drawing was done in ACAD or on J size sheets. I would suggest that you put in some kind of check and balance into your configuration management. So that when a new or changed drawing is released, every significant person knows about the new design or change.


Go Mechanical Engineering
Tobalcane

 

[BobPE]

btrueblood makes a good point though Tobalcane. Back in the days, communication and respect between engineers and drafters was at a much higher level. Today, it seems that even though the engineer is involved, CAD product makes it by them in many ways to become final work product. You have to remember, a lot of the CAD people are being substituted as engineers now...so technically, work is being reviewed and money is being save by not haveing to bother with those engineers...I have seen this business model applied to engineers all too often, and I believe it is the technology that is aiding this new way of thinking...

Now, I don't know if this is good or bad...but it is definately different that how things used to be done, and change can be a bad thing before it becomes a good thing....


just my thoughts...

BobPE

 

[Tobalcane]

Is it because technology made designing faster so that engineers did not have time to contemplate or communicate…I kinda see that; or the manager skipped the ethics class and did not involve the engineer so that the design can move forward with out snag? I kinda feel that it was not the software that made people not communicate, but the users of the software who decided “not” to communicate. Like the question “does guns kill people or do people kill people”? The gun did not kill the person, it was the user of the gun who killed the person.

Go Mechanical Engineering
Tobalcane

 

[BillPSU]

btrueblood makes a good point collaboration between engineers and draftsmen have helped to meet design specs and smoothed the design process. In today's world collobration must be done so the design can meet not only the design specs but manufacturing, assembly, service and customer's wish lists.

I've lived with many designs which were thrown over the wall at manufacturing. The machine was tooled up and the parts made. The proto machine had many problems and redesign of many parts parts required. Costs were too high. Assembly time was excessive. Servicing the unit was a nightmare. Time to market was not good.

I've also been involved in one project where a new machine was needed to compete against a competitor's new machine. A team of people was called together and told we need a new machine in 6 months. Previous design and build of a new machine was 18 months. The design engineer was the final decision maker on any disputed concepts. Reuse or modification of previous machine parts with their proven designs reduced testing requirements. Manufacturing knew the tooling needs long before the actual designs were released. Assembly and service problems from previous designs were addressed. The end result of all this collaboration was a completed machine in 6 months. Meeting and exceeding expectations on almost every area.

The machine I'm talking about is a digger derrick used by electric utilities with a 45 foot boom height mounted on chassis. It still is one of the standard machines available today.

Collaboration must start at the conceptual phase not at the end of the design process. Many design decision are made very early in the process and changing late in the process becomes time and cost prohibitive.

 

[GregLocock]

Well, I don't want sound like a salesman, but one of the biggest improvements I've seen has come about BECAUSE of CAD.

Each week (I think) the solid models that the CAD guys have been working on are rounded up and loaded into a viewing system. This means that anyone with access to a web browser can pull up the latest solid model of the car, or whichever parts they are interested in, and using an overhead projector can use it in meetings. Using net meeting the guys in the plant can participate at the same time, and so can the supplier, all they need is remote access to our intranet, or to go to one of our offices for a meeting.

OK, it isn't as good as getting everyone together around a drafting table, but it is the only way of dealing with suppliers from around the world.



Cheers

Greg Locock

 

[zdas04]

Greg,
That is a great example of technology being used as a communication tool to yield better results than were possible just a few years ago. All of the "stakeholders" have the ability to participate in the design decisions. The end result has to be better than we ever could have done with a T-Square.

This thread has been about the other end of the spectrum where technology is used instead of communication as opposed to acting to enhance communication. The fault isn't the technology (it is the same technology in both cases), it is with an organizational model makes people act as though the technology could "think". Any time someone allows a computer program to replace a decision (as opposed to supporting the decision process) you get another case study on how not to engineer.

David

 

[Comcokid]

I believe that indirectly, the computer and the push for ever increasing productivity has made most engineering more "sloppy". My argument may apply more to engineers that work within an "industry exempt" company structure rather than those engineering disciplines requiring professional licensing.

25 years ago, an electrical engineer was specialized in RF, Analog, Digital, Systems, etc with only a limited amount expertise in other areas.

Today, A company may advertise for an "Analog Engineer", but listed in the experience requirements (in addition to analog design) will be something like:

"Experience required : FPGA programming, verilog, C+, Visual Basic, LabView, evaluation testing, proficiency in PowerPoint and Microsoft Project. Familiarity with FMEA, ISO2000, EMC testing or experience with products for such-and-such-market a plus."

Now Lets Face It: You just can't be a expert at everything. I know I am constantly required to solve problems, make decisions, and use software tools I am not completely familiar with. On occasions, I have even had to solve problems in engineering areas of: mechanical, materials, manufacturing and patent law.

Engineers working in some areas are being slowly forced to know everything at the expense of being specialized, and that results in sloppy engineering.

 

[ewh]

I agree with Comcokid. Like the saying goes, "jack of all trades, master of none."

 

[Tobalcane]

I see software as tools for your profession. Just like tools for a carpenter. You can not expect a carpenter to use one or two tools to build a house. The carpenter has a host of different tools that he uses to build a quality house. Now granted that person is not an expert at hammering, he/she may not know how to forge or create a hammer, but he/she knows how to swing it. He/she may not be a top expert consultant on saws, but he/she knows how to cut wood with it. The same goes for engineers. For Mechanical engineers, they may not be experts at Pro E or A-CAD, or Pro-Mechanica or Ansys, or MathCad or Excel, or TAS or FlowTherm but he/she knows how to wield the software to their advantage. If you had two engineers in a room and one can only work with pen and paper and the other can work with pen and paper plus use the soft previous stated, which one would you hire? Well the defense companies are hiring the latter. Being sloppy is not making the effort to improve your software skills so when you are called upon to tackle a task and then you are not prepared to do it.

Go Mechanical Engineering
Tobalcane

 

[CajunCenturion]

An interesting analogy Tobalcane, but can a carpenter does the job without any tools?

Can engineer do the job without a computer? I sure as hell hope so.

Good Luck
--------------
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[Comcokid]

Unlike a carpenter’s tools, software is in a constant state of change, and every software hammer is designed so different that between some there is no similarity. To make full use of the features in some tools may require YEARS of experience.

Some of the higher-end software packages have not only a new set of features with each release, but a new set of bugs as well. One electronics package (that I will not name) recently made such a complete overhaul to their user interface that it took me several months to get to the same level of proficiency as before - and I had used it for 12 years. The project I had during this period is very sloppy in terms of the document quality - and would be a little lower cost had I not been relearning the tool.

Not all packages have the same issue. In AutoCAD, I can type into Release 14 the same line commands that it used in the old V2.13 DOS days (cir 1984), and it still works.

 

[HgTX]

AutoCAD is the only software I've met that put on a Windows overlay without removing any old functionality.

Otherwise I'd still be using WordPerfect.

Hg