Structville

A hub for civil engineering related designs, analyses, discussion, information, and knowledge...

Trending Posts:

Search This Blog

Sunday, April 22, 2018

Do Commercial Softwares Make Engineers Better?


In the 1950s, a method of analysis involving discretization of a continuous domain into a set of discrete sub-domains called elements started in the academia. The Finite Element Analysis method is a numerical solution to problems in engineering and mathematics, and has been applied extensively in structural engineering.

In 1965, NASA issued a request for a proposal for the development of a structural analysis software tool.  The result of this was NASTRAN (NASA Structural Analysis), which implemented the available FEM technology to solve structural problems (https://www.esrd.com/simulation-technology/brief-history-of-fea/). In the 1970's commercialisation of finite element analysis began in full force, and gave birth to finite element modelling. Apart from FEA, design packages also sprang up within that period. For instance, DECIDE (DEsk-top Computers in DEsign) package was written by A.W. Beeby and H.P.J. Taylor in the mid 1970s, when they were both in the Cement and Concrete Association. The program was covering all aspects of  design in accordance with CP110, from structural analysis to bar curtailment.




Also around 1980, OASYS software was developed by Ove Arup Partnership for Hewlett-Packard 9845S system. The program was produced to include a large number of various aspects of structural anaysis and design, and other aspects like surveying, drainage, road works, thermal behviour of sections etc. The CP110 design package of the program contained programs for analysis and design of continuous beams, rectangular and irregular columns, flat slabs, foundation, etc. While all these were going on, in 1982, Autodesk made Autocad commercially available for drafting and computer aided designs.

From that moment till now, numerous computer packages have been developed to assist engineers in analysis, design, simulation, and drafting. We have ANSYS, Staad Pro, Etabs, SAP, Safe, STRAP, Tekla, Orion, and so many packages too numerous to mention. There are also many computer programs written with packages like FORTRAN, MS Excel VBA, MATLAB etc, which serve the purpose of helping the engineer solve a specific problem.

These packages have made life easier for design engineers in the office by considerably lowering the man hours spent on analysis, design, and drawing. This means that working drawings, costing, and presentations can be done faster prior to kicking off of constructions. On the other hand, it has made it easier for corrections and modifications to be made to a structure, with their effects appropriately accounted for.





But the question still remains, 'Do all these programs and packages make engineers better?'

In my own opinion, the answer is relative. An engineer is an individual who has been trained in an accredited institution, with appropriate years of experience and professional licence to practice his profession. An engineer is expected to improve on the job, by following up with relevant technological advancements in the field, attending seminars and conferences, and networking with his colleagues. For instance, a new construction procedure might demand a new method of reinforcement detailing and it expected that a modern engineer keeps up with all these developments.

Back to the issue of softwares, if an engineer is proficient in the use fifteen softwares, does that make him better than an engineer who can use just two softwares? The first answer that will come to mind is probably 'NO'. This is because the knowledge of softwares do not make an engineer. What makes an engineer is the person, his knowledge and background on engineering, and his years of experience. To very sound engineers, the knowledge of softwares is not an end itself, but an alternative means to an end. These packages are here to help engineers work faster and more efficiently, and that is why they are designed to request our input before they can give us output. 

I was involved in a project here in Lagos Nigeria, and when I looked at the structural drawing, it was a complete disaster and waste of resources. The residential building was to go four storeys high, and no span was longer than 7m. At the ground floor, a column section of 900mm x 300mm was provided with 10Y25mm (Asprov = 4910 mm2) as reinforcement. At the first floor of the same column, 28Y25mm (Asprov = 13748 mm2) was provided!! This drawing has been approved and construction has commenced already. I cried in my heart for engineering. If the person who designed this had passed through the same training as we did, he would have asked questions. But a certain software gave him the result, and he submitted it just like that without probably reviewing the output. Even if there were unbalanced moments due to eccentricity or structural arrangement that is causing heavy bending on the column, it his duty to scrutinize the arrangement. But he is a registered engineer who can use a software, yet he has shown that he has no solid foundation of engineering as far as that output was concerned.





In terms of job opportunity, knowledge of a certain software can be an advantage over another candidate. Yes, this is because a firm might be users of a particular software, and they might prefer to higher someone who already has knowledge of that software. But on the other hand, it would still be very disastrous if the desired candidate does not know the flexural behaviour of cantilevers, or where hogging moment would occur in a continuous beam. This calls for consideration on the part of employers.

In summary, commercial softwares are helpers. We are supposed to do the thinking and design, while computer programs help us carry out the lengthy calculations. When they produce an inconsistent result, we are supposed to ask why, and review our inputs. These softwares do not make better engineers, but they have made engineering as a profession better. That is why Structville remains committed to original and sound knowledge from the first principles. The current curriculum of our universities is sufficient to produce engineers who can handle simple designs after graduation, and lecturers should be willing to 'teach' design, and not 'lecture it'. 

From my experience in the field so far, the pattern of teaching received contributes immensely to the problem most fresh graduates encounter in carrying out design independently. They were lectured and not taught, and as a result, they spend the rest of their professional life relying on softwares. Structural design at undergraduate level should not be an issue of, "take this handout and study the examples." It should be taught passionately from scratch.

I will ever be grateful to Prof. C. H. Aginam of the Department of Civil Engineering, Nnamdi Azikiwe University Awka. In the first semester of our 3rd year, he loaded us with knowledge on deflection of elastic systems, analysis of continuous beams (Clapeyron's theorem of 3 moment, slope deflection method, force method), Maxwell's theorem, Betti's law, Vereschagin's rule, Castigliano's theorem, analysis of portal frames etc. When we found ourselves mid-way in 2nd semester, we were already designing three storey building using pen and scientific calculator from the roof beam to the foundation. I still remember the way he shouts and sweats, writing on the board, and cleaning over and over again, inspecting our notes, making sure every one downloaded and printed a copy of BS 8110-1:1997, ensuring that everyone purchased at least one design textbook of his/her choice. He is a phenomenal teacher.

Benefiting from his wealth of knowledge, I did all my designs with pen and calculator until I went for my Industrial Training in 4th year. It was during my industrial training that I had my first laptop, and learnt how to use AUTOCAD, and a few other softwares. My point is this, it is the raw knowledge of engineering and ingenuity that makes a good engineer; the softwares are here to make our work faster and eliminate human errors from length computations. However, the presence of these softwares have made engineering as a discipline better and more efficient. Let us keep getting it right. 


4 comments:

  1. Nice post brother. Having the basic of engineering taught in the school should be a guide for your software usage. I always tell graduate students who come to my office for National Service that the software's are good but the ultimate is knowing the manual procedure. If you are good with a design software and you cannot explain your parameters or understand the outcome of analysis then you are at risk with engineering.

    ReplyDelete
  2. Thank you Ubani. This is a really insightful post and your assertion that the software does not make a better engineer is very true. Like your design "teacher" I was privileged to work with a structural engineer who had his education in the UK, after my year II in the polytechnic. I was already bored with AutoCAD at the time and was hoping to learn how to use many of his numerous softwares (safe, etab, orion etc), but i was disappointed.

    Fortunately a residential job had just came in at the time, and my boss asked me to prepare a general arrangement from the architectural drawing. With some corrections and gk & qk parameters determined, I began analysis of structural elements. When we were done, he asked me to model the analysis on Excel by programming using Qbasic & formulaes, this was to serve as a worksheet for all our subsequent works (with little modifications ofcourse).

    Then he took the earlier GA and design parameters and modelled it on SAFE and when he told me the SAFE result wasn't economical and has too much unnecessaries on the detail plan, I asked him how? Then he began to take the design moments from my earlier analysis and I saw a remarkable difference. Ever since, I have appreciated hand analysis more, especially writing it on excel sheet to help me understand how my structure is behaving. And I think engineers should do this more often. It helps them appreciate even the softwares even better.

    ReplyDelete
  3. Hello sir,
    I wanted to know your method of designing three storey RCC buildings with pen and scientific calculator and without using analysis software.
    If you have published other blog articles relating to this topic, please share the links.
    If you have not yet, please kindly give me some pointers on this and better yet write one. It would be very helpful.
    This was a very enlightening post. Thank you!

    ReplyDelete