How often do your calculations get checked ?
How often do your calculations get checked ?
(OP)
Hi all.
In my 13 years as a structural engineer (8 in buildings & 5 in railcar structures), I can count on one hand the number of times my design calcs. have been checked by a second engineer. I've always worked in small engineering firms that simply didn't have the time or budget for it.
Usually, design drawings have been reviewed for blatant omissions etc, but nothing more.
I'm curious how many of you other engineers out there are in the same boat. Any thoughts?
In my 13 years as a structural engineer (8 in buildings & 5 in railcar structures), I can count on one hand the number of times my design calcs. have been checked by a second engineer. I've always worked in small engineering firms that simply didn't have the time or budget for it.
Usually, design drawings have been reviewed for blatant omissions etc, but nothing more.

I'm curious how many of you other engineers out there are in the same boat. Any thoughts?






RE: How often do your calculations get checked ?
It is a comfort to have another set of eyes on anything.
RE: How often do your calculations get checked ?
Just like Ron, I've worked on jobs where my calculations were checked by senior engineers/peers and on other jobs where my calculations and drawings were not checked at all prior to IFC. I too welcome a second set of eyes looking over my design. No one's perfect all the time!
Regards.
RE: How often do your calculations get checked ?
The value of an unbiased check by a second person is often ignored due to the time and budget constraints. We have a quality system in our company which emphasises the second check. We should realise that in spite of one's experience and knowledge slippages do occur and these checks help to eliminate such errors before the deliverable leaves for the site. It is far better that the errors are identified and rectified inside the design group rather than being pointed out by the customer.
I strongly recommend that in the interest of the project a reasonable provision shall be included in the budget and schedule for this activity.
We try to minimise the time and effort required for this activity by using validated software (either in-house developed or marketed ones)for analysis, design and detailing so that the checking of a less volumenous input ensures the correctness of a larger output.
RE: How often do your calculations get checked ?
Also, SERMC puts out a Calculations Checklist which you can run through yourself, which is very helpful in catching errors.
THey have a Drawing Checklist, also.
RE: How often do your calculations get checked ?
No one wants to check calculations on their own; so it falls onto the project manager since his/her neck is on the line.
So what we usually end up with is a less than thorough review of plans and a crap shot on the calculations. Some members of management seem to believe it's more important to check drawings than calculations. I've always been under the impression that one needs calculations in order to advance to structural drawings.
Sorry to lament but that's the way it is here.
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
I would definitely vote for a chcking system in any engineering activity. One can think of reducing the checking time by innovative methods without defeating the purpose of the system.
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
Andy Machon
RE: How often do your calculations get checked ?
At the smaller firm, we are striving to make sure that quality issues are not shorted. As with larger firms, we have procedures in place to require sign-off of all calculations, reports, documents, designs, etc. The difficulty in a smaller firm is the lack of sufficient number of senior level individuals to affect these processes. Often a single senior individual is responsible for the technical review of several different locations, some of which can be hundreds of miles apart. I suppose Fedex loves us!
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
I have worked for many years now at a large company (ENR Top 5) and can say that checking is paramount to what we do be it engineering or detailing. While I'm glad for that I also realize that it is my responsibility to see that things do get checked. And yes we must add this cost/time in our budgets/schedules.
I have also spent some time for smaller companies and have experience a general lack of concern for checking. In those cases I have always developed another method of looking at the results to determine if they are indeed correct. I have always placed a critical eye on the plans as they are the product by which design firms are evaluated by the contractor/manufacturer.
Lastly, I worked a number of years for a state agency doing structural analysis. This is more to the point of what RSTucker note - not only was checking not a problem it was too much of a problem. Believe it or not we did everything twice. A project was given to two engineers and each went on their merry way to design and sat down at the end to compare answers and methods. Wow! talk about overkill. But I can also say that not much went out the door that was in error or couldn't be defended.
RE: How often do your calculations get checked ?
I appreciate your last statement. That is in fact the ultimate aim. As I have stated in my earlier post, I feel that the damage caused by unchecked, erroneous documents is not just the loss of manhours or money due to the rework. It is also the embarassment to the team and the consequent fall in the morale of its members. In some cases, such errors stay as history and are referred on occasions.Hence, the reviewer shall realise that he is the last check point beyond which the damage can not be prevented.
There is nothing like framing a checking system and religiously following it.
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
To further the point that checking is essential, we allow our graduate engineers to carry out mathematical checks of the calculations (to ensure the numbers add up correctly and the decimal points are in the correct place). This has two benefits:
1) The designer is forced to lay out his /her calcs in a logical manner with good references in the margin so that the graduate can follow the calculation.
2) The graduate can learn from the designer how the structure is actually designed in the real world and pick up design tips etc.
The final check for overall stability, robustness and fitness for purpose is carried out by a senior engineer.
Think of checking as graduate training on the cheap!! It's good value for money all round !!!
Andy Machon
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
VALUE...........
It is unwise to pay too much, but it's unwise to pay too little.
When you pay too much you lose a little money; that is all.
When you pay too little you sometimes lose everything; because the thing you bought was incapable of doing the thing you bought it to do.
The common law of business balance prohibits paying a little and getting a lot.
It can't be done.
If you deal with the lowest bidder, it's well to add something for the risk you run.
And if you do that you will have enough to pay for something better.
John Ruskin (1819-1900)
I think this says it all !!!
Andy Machon
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
Thanks also for giving me a perpetually nervous stomach...
Joking.
GA
RE: How often do your calculations get checked ?
Checking of calcs has varied tremendously depending on where I worked. My first job out of college was in the structural dept of a large paper company that did all its structural work in-house. Many of the engineers at this company had been there a long time, and they were careful and detail oriented. Plans did not go out until they had been carefully reviewed by the structural department head or a project manager. Sometimes the calcs were reviewed also, sometimes not.
Subsequently I worked for several different size consultants, one an A/E, the other an E. No one ever checked my work at these places, besides an occasional, cursory review at the end of the project just before the drawings went out. The resulting comments usually had more to do with what types of practices and details the reviewer preferred, rather than corrections to gross mistakes or problems.
At the A/E, a mechanical engineer sometimes sealed the structural drawings. So much for quality control (not to say the laws governing our profession!).
RE: How often do your calculations get checked ?
The only procedure I've found that is both cost efficient and effective at finding significant errors is to have one of my expert colleagues review the drawings and specs. Most bridges, wharves and building structures can be visually scanned in an afternoon. With a little experiece and lots of concentration, it's amasing how quickly blunders jump off the page. When I think of the major structural collapses in recent years, none of them should have escaped a visual ispection of drawings by an expert.
I liked Ginger's comments, but the following point may not be as absolute as presented. "The common law of business balance prohibits paying a little and getting a lot.
It can't be done."I think it can be done.
A few years back I was reviewing construction bids for a concrete caisson wharf extension. The low bid was less than half the next higher bid and almost 33% of the average of the other bids. One would think this was an error, but it turns out the bidder had a set of slip forms from a recent previous job and could produce the few required cribs at half the cost of other equally qualified constructors. The owner almost dismissed the low bid on the basis of "fundamental error", but ended up accepting and ultimately was happy with the outcome. Just goes to show anything is possible.
Regards,
RE: How often do your calculations get checked ?
Last winter I was asked to seal a set of "revised" bridge plans. First, I need to explain my use of "revised". I was the project engineer for the original design that was completed about one year earlier. After that, I was on a field assignment during which time there were substantial changes in the highway alignment requiring redesign of the bridges; they became longer. [The redesign was left pretty much to a group of junior engineers; then again whatever guidance they received from senior engineers was worthless]. When the time came to submit the drawings, lo and behold, I was directed by the management to seal the plans [I was one of only two people licensed in that state.] I refused to do so because the work was not done under my supervision.
After much wailing and gnashing of teeth, the project manager finally agreed to speak with the state board about my concerns. The Board's descision was that I could seal the plans provided that I personally check the calculations and plans. With that resolved I agreed to seal the plans. Initially the PM refused to give me the calculations because "it's to much work to send you a copy". His position as well as that of our branch manager was that if the plans are checked that's good enough. Obviously, they didn't understand that if the calculations are wrong, the drawings are wrong.
Well, back to square one. Finally, they agred to give me the calculatons. In the meantime, they asked me to look at the drawings, which were an abomination! Things just didn't look right.
When I got to the calculations, my fears were realized: Not one page was ever checked. Everything was wrong; among other things:The abutments were underdesigned. The bearing design was wrong. The reactions shown on the plans did not match the ones from the computer output. Actually, there was no design for the bearings someone just put the old calculations in the book. The girder splice design was wrong. [Someone needlessly added wind bracing on the plans and decided to shift the splice location 10' but never redesigned the splice. The wind bracing was unnecessary and the way it was detailed, it wouldn't do any good anyway. This should be no surprise - there weren't any calculations for the bracing.] The girders were significantly overdesigned.
Bottom line: The office had to eat the cost of four people to start from scratch; it took about five months. Luckily, there was a partial delay by the client due to funding issues and management was able to stall for more time. Meanwhile, we looked pretty foolish issue substitution sheets during the bidding process.
RE: How often do your calculations get checked ?
However, I am glad to know how you handled the situation and beleive that you acted in everyway as I would had I been in such a position. You are a good engineer and if they would have postured by hanging your paycheck over your head, you could have easily found employment elsewhere. Best of all, you showed the junior engineers that work itself is not good work or work worthy of being sealed and that drawings by virtue of having ink on the sheets are not worthy of construction. You showed them that arrogance (consultant is right, client is wrong - what does the client know anyway) has no place in our profession and that when principles are on the line those engineers have a place to turn to as well.
Thanks!
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
Since I was born with one eye the armed forces would have nothing to do with me, so I found a job in structural analysis in the aircraft industry. I worked throught the war on at least some part of practically everything that flew, since we subcontrcted large sections. Everything was checked.
After that I worked with a small structural consultant, designing relatively large projects. Nothing was checked.
I later opened my own office. Then I went with a large international A/E.Then I went to another large A/E. Then I quit the race and opened my own office again.
So what has been my experience? Generally speaking, it depends. If the product produces an extremely refined product such as going to the moon, everything is checked again and again. If it is the joists under a frame office building then if the doer is at all experrienced, it is not checked. And if the design is based on data that is not precise, like an office live load, I do not feel worried about that so much as if it is based on the gravitational pull of Mars as it goes by.
So an engineer's judgement which is supposed to be professional, is very important.
Evank, Structural Engineer
RE: How often do your calculations get checked ?
Salutations to one of the very senior members of the forum from a relatively younger generation.
As you have rightly stated at the end, an engineer's professional judgement is equivalent to a number of repeated checkings. But I prsonally feel that this professional judgement is the result of carrying out a number of such checking and counterchecking through one's career. I am sure you will agree with me that going through this experience during your early years of your career has given you that professional confidence.
Right or wrong, the present day younger generation is not having such an opportunity. The exposure to a variety of 'black box' type computer software has made many of them less alert in terms of checking. There seems to be a tendency of 'taking for granted' the results from a software. the attainment of Professional judging capability is retarded by the lack of manual working and understanding the basis of analysis and design. Many of them are very good at operating the software but they know neither to justify their results nor to doubt them. If they get an undesirable result, they do not know which parameter to change to correct it. In short, the 'Physical feel' of the calculations is missing.
There may be a few exceptions to the above but I feel the majority fit into this pattern.
I would suggest that it is best to have a person of your caliber and experience at the exit point to see everything is OK but still have the systematic checking procedure which not only ensures the quality but also make the people down the ladder to become like you in the coming years.
RE: How often do your calculations get checked ?
RE: How often do your calculations get checked ?
In many companies senior engineers have also been guilty of dumping the number crunching aspect onto graduates who are more adept at everyday computer skills but have little feel for the overall design of the whole or the elements. It's easy to hit a default button and end up with an E value for steel when it should have been for aluminium. There is also still the problem that commercial organisations produce charts listing stresses, deflections etc with no supporting background as to the derivation of these, all of course produced by some magic software that is beyond reproach.Ever tried phoning some of these companies and asking just how these results were produced?
Fair do's, computer software has revolutionised everything but as far as checking what is produced, we still need to keep the maxim "rubbish in, rubbish out" uppermost in our minds when overviewing a design.
All design needs to be checked; no matter how long one has been doing the job a fresh look can often see the wood from the trees - something that is overlooked if working in isolation.
RE: How often do your calculations get checked ?
To the subject, My cacluations usually are a mix of hand calcs, printouts from computer programs (with a lot of hand notations on them; I usually hand-compute the first iteration of a design calc, then verify it on the computer, then crank out however many variations with the computer as I judge necessary) and photocopies of whatever pages from whatever documents (e.g., company literature or textbook tables, etc.). All the pages are numbered and I include a table of contents at the beginning, so it's easy to follow... Where I work "checking" isn't what's done, an "Independent Technical Review" is conducted once I'm finished and that's there where actual checking is done. Do I like that? Eh, maybe, maybe not. But to answer the thread, my calcs are checked 100% of the time.
A design calc package I hand in for review starts off with a one-page narrative/memo then a series of "draft final" sketches - usually CAD work (I do it, we have no draftspersons) that doesn't include borders, logos, etc. I assume that the person doing the review will immediately spot errors and/or omisions just by looking at the sketches (hopefully I haven't made any) - we usually have extremely senior design engineers doing the reviews, an awesome benefit.
RE: How often do your calculations get checked ?
We have had some extremely wise comments. They all have merit and none give a black or white answer!!
The comment about the blind use of the computer is one we should keep in mind. There was a failure of a gymnasium roof in Connecticut about 30 years ago. It was a dommed long span that had been designed by the then new computer thing. It failed early in the morning after a snow storm. It was discovered that the conmputer had faithfully designed it just fine, but had not included the effect of one sided snow load. And the snow had accumulated behind a parapet and not the other.
But I admit I am looking at this from a different perspective than others. I still depend on the judgement of designers if they have had sufficient experience to merit trust. So I guess it is TRUST that I look for, and not the nuts and bolts of checking arithmatic or the printouts from a computer, wonderful as they are.
Thanks for your comments. It has been very informative.
RE: How often do your calculations get checked ?
No, I do not recall Sprague... in those years. Most of us were unemployed, and I worked for a month or so at the Federal Land Bank in Wichita. An exciting job of drafting onto maps the locations of Land Grants. I could not stand that pressure so went to American Bridge in Trenton. No better place, but WW II changed all that.
EvanK
Structural Engineer
RE: How often do your calculations get checked ?
Individual competence and diligence are necessities in engineering. As some have said thus far, our younger engineers don't always have the opportunity to learn the importance of these necessities. In the interest of expedience, we often allow younger engineers to "sink or swim". This is not an appropriate posture. Engineering is not a "Risk Management" proposition. It is a scientific methodology. The "Risk Management" approach causes the problem bridgebuster described (which he handled nicely and correctly in my opinion). We have to continue to apply a scientific methodology even in the face of inaccuracies (structural loads, assumptions, etc.) and absurdities (management philosophy), simply because we compound our errors when we don't.
I'm a great proponent of progress and computerization, but I still check my calculations (after a cooling period!) and those of the engineers I mentor. I also welcome the check by others...even though I have a lot of experience, I'm not infallible. Those of us who remember slide rules also remember the need for "order of magnitude" checking. This mental process is necessary and good. I try to instill it in those I mentor, even though we obviously don't use the slide rule. I do this through sometimes subtle (sometimes not so subtle) questioning of their approach and how they came to realize their analysis or design. I have soiled the back of many a napkin in this process!
RE: How often do your calculations get checked ?
You design it and sign off on the plans and someone dies. There's no question about it... it's your fault ( 100% ). Not checking can easily be considered as negligent or incompetent.
I work for a large government agency doing bridge designs. We also review bridge design plans developed by private firms. It is very obvious to us here that a LOT of private firms don't do very good checking on their designs even though their contract states that they are supposed to do all designing and checking ( they are paid quite well for their services... I see no excuse for them shortchanging the process ). The number of blatant erros we find in their plans is awful.
Dan
www.dtware.com
RE: How often do your calculations get checked ?
I used to work for a large government agency doing the same thing many, many years ago. I've worked in consulting for a longer time now as well. But one thing that gripes me is exactly what you stated in you last post. Errors are far too common and the industry seems to be very complacent about it. Others, including the company I work for have developed Quality procedures that are meant to curb this but in my opinion they are sometimes too cumbersome and extravagant to do any good. I have always relied on the two sets of eyes check on everything before it goes out - and that includes preliminary drawings.
Another things that gripes me is the way some companies don't produce plans like those provided by the government agency - doing so has many benefits not the least is easier review and coordination with other plan sets and contractor familiarity. Inevitably, I see someone who wants to re-invent the wheel just to be different, just to stand out. It drives me crazy!
RE: How often do your calculations get checked ?
In my earlier commercial consulting days little was formally checked, but I used to do my own reality checks. Manual load take downs were checked (reactions totalled, applied loads totalled - were they reasonably matched?). Larger more critical members, I would recheck, with a different method if I could.
As someone earlier mentioned even senior engineers can make silly little mistakes, and as one of my profs used to say (his Law 199) no one mistake usually causes a collapse, it takes an accumulation of mistakes. An early boss tought me a simple rule - does it look right? Many mistakes can be spotted by looking at something with a detached eye, and the absurdity can jump right out at you. Experienced designers (technicians) and draughtsmen can often spot these goofs.
In the calc checking that we currently have to do we have two options. Literal detail calculation checks, or alternate approach. I generally hate the detail calc check, particularly with some individuals whose calcs while maybe correct are extremely difficult to follow, so I tend to favour trying to come at a problem from a different angle and see if I arrive at a similar solution whenever I can.
On one project with an intermediate engineer whose calculations were always difficult to follow, I was getting quite frustrated trying to follow/check his numbers, so I decided to sit back and look at the problem a little differently, and the flaw that I finally found in his calculations was not in the number crunching itself, but in an assumption of flange continuity that I was not convinced was realistic (vessel lifting beams were cantilevered off eiher side of a deeper saddle extension, so that the top, compression flange was discontinuous, with only a partially stiffened web plate connecting them). In lieu of doing a detailed finite element analysis to make sure that we would not get local plate buckling somewhere in the beam/saddle combination causing us to drop the 150 ton vessel/platform assembly during the critical installation lift, I decided to add some temporary lift bracing between the upper and lower top flanges to transmit these forces via truss action. Although the other end was more lightly loaded and did not present a "failure" problem in the number crunching, higher than anticipated deflection during a lift to weigh the module caused us to decide to add the flange braces at that end too. This was partially exacerbated by the fact that the vendor had given us a bad c-of-g and so we actually had more load on the other support than we had expected. But basically the check here had been sit back and look at the overall concept, not get tangled up in all the detailed numbers.
RE: How often do your calculations get checked ?