Skip to main content
Premier resource for practicing structural engineers

Importance of Hand Calculations and Rules of Thumb in the Artificial Intelligence Age

By Ciro Cuono, P.E.

The old rules of thumb never looked so important as now.

Structural Engineering, as a profession, developed in approximately the mid-19 century mainly from the emerging technologies of the industrial revolution: iron, steel, I-beam production, development of railroads, etc. These emerging technologies helped spur the development of a “specialist” who emerged from the fields of architecture, mathematics, and construction. As structural engineering fully developed into an established profession, structural engineers created standards, codes, and textbooks to guide and train the members of their profession. In an age of pre-computing, their tools consisted of pencil and paper, theory, practical experience, and rules of thumb. This served them well for many years and required engineers, who were analytical, proficient at arithmetic and higher-level math, and practical, to be capable of understanding a physical problem and developing a conceptual solution that could be drafted and finally validated with mathematical tools. The mid-20 century Spanish structural engineer, Eduardo Torroja epitomized this notion best in his book, “Philosophy of Structures” (1958), with the following quote: “the calculation of stresses can serve only to check and to correct the sizes of structural members as conceived and proposed by the intuition of the designer. The work itself is never born from the calculation”. This quote summarizes a way of approaching engineering problems that has been endorsed by many past and current engineers in our profession. To paraphrase some past professors and mentors: “you must know the answer to the problem first”; “first do your thinking on a piece of paper or trace paper, then draft and calculate”; “you must have a feeling for structures,” etc. These are all different forms of Torroja’s quote, which shows that through study and experience, engineers develop an intuition for structural behavior. Once this intuition has been developed, a variety of solutions can be quickly developed and tested, and the final answer can be validated through a design check. These notions, of course, are familiar to most (particularly older) engineers and have been the basis for good, well-rounded engineers.

Our profession, however, like all professions, is constantly evolving and never static. This evolution sometimes moves at a faster pace but is indeed a constant force. To summarize a few key points that illustrate this: the industrial revolution brought about new materials such as cast iron, wrought iron and finally steel, which, in turn, brought about new industries such as factories and railroads. This caused the more mathematically minded architects to break off to form the structural engineering profession. Parallel to this, the slide rule, which dates to the 1600s, evolved and became the standard tool for engineers until about the 1960s and ‘70s, when it was replaced by handheld calculators. Parallel to this, structural analysis of indeterminate structures was born from truss analysis and elasticity methods and eventually evolved to computational methods such as the slope deflection method of George Maney (1914) and the Moment Distribution Method of Hardy Cross (1930), which had a short learning curve and produced rapid and accurate answers. At the same time, matrix structural analysis was in its infancy and rapidly developed after World War II, mainly in the aeronautical field, and morphed into the finite element method, which is the background engine of all structural analysis done by packaged commercial analysis computer programs. Today rapid computing and advanced graphics have made it possible for even a small office to analyze complicated structures effortlessly. Fifty years ago, what would have taken a whole team of engineers’ days and days to do can now be done in a matter of hours with a smaller team. On the horizon is now the allure of artificial intelligence (AI) and the tantalizing possibility of removing all computational effort from even the most complicated designs, freeing engineers from the dog work of calculations to focus solely on testing various forms and solutions and focusing wholly on solving the problems of humanity’s built world.

As this next technological revolution (AI) bears upon us, some important questions on the training and development of members of our field must be considered. Namely, what is the importance of hand calculations and rules of thumb in the artificial intelligence age? How should we be teaching and training the next generation of engineers? Do we keep the same curriculum from the last 30 to 50 years? Should we be teaching the slope deflection method to engineers who will likely be modeling and designing structures with computing technology that can be run from a phone or watch? Will the development of AI create an environment where the structural and architectural professions merge into one profession? Nina Rappaport’s book, “Support and Resist: Structural Engineers and Design Innovation” explores structural innovations of firms that have been on the cutting edge of structural expression and collaboration – a timely read in a world of emerging AI.

Given the legal, moral, and ethical foundations of engineering and our obligation to public safety, the answer to these questions is we need to ensure that the training and development of the next generation of engineers should be accomplished in a way that leads to confident and competent engineers who can look back at their history with pride and move forward with new tools and methods and build great structures that benefit humankind. The surest way to accomplish this goal is to emphasize hand calculations and rules of thumb so that the emerging generation of structural engineers can develop their structural intuition and “feeling of structures” and have quick, easy, and meaningful tools to check their designs, which will surely be “spit out” by incredible technology. This does not mean that we should expect entry-level engineers to do their designs with slide rules and the slope deflection method but rather that a certain amount of training should be done by simple hand calculation methods and rules of thumb. As an entry level engineer advances in his or her career, they can use these techniques, training, and rules of thumb to check themselves which will eventually become their “sanity” checks when they have grey hair and become the mentors to the next generation.