The modern day structural engineer is becoming a commodity. Whilst our science has never been more well-defined, information more accessible, and analytical tools more powerful, we often find ourselves creating standard, overly-conservative solutions that pale in comparison to the beauty, elegance, and material efficiency of engineers that have preceded us. The reason for this paradox is two-fold. First, we live in a unique era in which material is cheap and labour is expensive and have subsequently forgotten our roots as designers who balance safety and elegance. Second, we have yet to fully embrace the digital revolution and as a result, we’re fallen victim to it. In this essay, we’ll discuss our current situation and suggest changes to our education to assure our craft remains respected.
The Sin of Being Overly-Constrained:
Clients are demanding faster returns on their investment and we find ourselves consistently being asked to adhere increasingly aggressive project programmes. Technology has unleashed the architect’s boundless imagination, resulting in more difficult projects with constant design changes. At the same time, our palette of structural components is still limited, as traditional construction methods carried out by cheap, low-skilled labour still govern what can be built in most parts of the world.
Figure 1: Authors Opinion – Structural Complexity Compared to Spatial Elegance for Various Projects
Figure 2: Constraints Leading to a Thin Concrete Shell Solution
Figure 3: Typical Constraints in Hong Kong
Figure 4: Morpheus Tower Project Constraints
Figure 1 gives a subjective view comparing elegance and structural complexity of some famous projects. The position of each project on the matrix, as seen Figures 2-4, is largely a product of the many constraints which effectively limit our design space and also impact the elegance of the spaces in which we live. Given the large amount of constraints we already have, it is amazing to observe in modern practise, we impart further constraints upon ourselves in two predominant ways.
First, due to code-compliance and adherence to contractor inabilities, we have allowed ourselves to be constrained to the simple yet conservative, instead of balancing safety and elegance.
Peter Rice accurately described our fundamentally conservative nature when he said “The structure must not fail. It must work, and continue to work under all conditions of load. It may look elegant….that is very satisfying if possible but if not you compromise a little. But it must carry the load” when describing his efforts to balance safety and elegance during the design of the gerberettes on the Pompidou Centre in Paris. In the modern era of cheap building materials, many times we are incentivized to be overly-conservative; avoiding abortive works onsite holds more monetary value than a few extra tons of building materials. In addition to this, design codes are now taught in parallel with theory and represent a “safe zone” for many engineers. Whilst design codes are crucial to maintaining a standard for public safety, they also have had a constraining effect on our designs; we often avoid designs for which the codes do not explicitly provide calculations. Yet, we have taken on these constraints voluntarily because they symbiotically cater to our conservative ethos. Our daily calculus has become “Where can I add conservatism at the expense of material to comply with the code and increase easy of construction to avoid commercial problems?” This is in sharp contrast Peter Rice’s calculus for the gerberettes – an exercise guaranteeing safety whilst increasing elegance by removing material.
Second, we have further constrained ourselves by not fully entering the digital revolution.
Our science is quite well-defined for civil structures. One would think combining technology and theory such as Finite Element Analysis (FEA) would have liberated us. Yet, we only entered the digital world partially – sure, many of our workflows revolve around the use of machines that can employ engineering theory faster than we could otherwise perform them by hand. However, in most cases, we are using the theory via a software which we did not create, leaving the possibility we may not be able to access some of the theory solely because we’ve centered our workflows around software we have no control over. We far too often find ourselves backing away from design situations because we claim our software of choice “cannot mesh that” or “it will take too long to set up” or “it can’t do that”.
Figure 5: Software Limitations Now Are Treated as Engineering Limitations
Yet, despite their shortcomings, these software can complete analytical tasks in a competitively short period of time over hand-methods that have been used by our predecessors. Despite the fact these hand-methods are transparent and appeal to the engineer’s intuition, such heuristic methods require time-intensive, manual iteration. Instead of being proactive and adapting such methods to digital workflows, we abandoned them, not even teaching them at the University level anymore. We have accepted our fate as slaves to which analyses our purchased software allows us to easily perform, effectively reducing the solutions we can provide.
The true sin of our generation of engineers is we’ve devalued our role by losing touch with how to effectively employ our science to create more elegant spaces.
If our only contribution to projects becomes taking a large set of constraints from architects, clients, and local authorities and converting them to a simple, code-compliant, and overly conservative structures suited to be easily input into analytical software and/or a contractor’s unimaginative construction methods, we will find ourselves being white collar technicians, rather than the influential designers. Engineers such as Nervi, Maillart, Isler, Torroja and many more were able to balance serviceability and safety to create a built environment which improved the lives of many with much less at their disposal. We have no excuses.
Breaking Free:
Marcus Vitruvius, the famous Ancient Roman architect, recognized our buildings affect how humans feel, which is why one of his pillars of good design is it must have beauty. He also recognized beauty is best derived from nature. Therefore, if we are to effectively increase the habitability of the spaces we help create, we must be able to design more natural spaces.
Most of our education and design codes have been set up for the straight and uniform, yet nature knows neither of these attributes, therefore, to create more natural spaces will necessarily mean we break free of our self-induced constraints to design structures that are safe and habitable. Our role on the path to elegant, livable structures will rely on us being able to confidently use engineering theory and computational technology to navigate the complexity of such designs.
To realize this will require a key change in education – we must change our education scheme to teach engineers coding languages which, compounded with first-principles, will allow us to take back control of our designs in the digital era. Many commercially available software have advanced computational engines yet overly-basic Graphic User Interfaces (GUI) which effectively limit what can be feasibly modelled. To overcome this, we need to utilize the programmatic interfaces of software to better access their calculation engines and in parallel, foster respect and understanding of FEA by teaching the art of how to decipher FEA results using first principles. Engineers equipped with coding skills will also be able to revive intuitive, first-principle methods of our predecessors in a modern fashion. Analyses such as Cremona diagrams, Yield Line theory, and Strut and Tie can be implemented into a 2017 workflow using easy-to-learn scripting languages such as C# or Grasshopper.
Figure 6: Strut And Tie Tool Developed in Grasshopper by Author
Also, more elegant solutions will require digital means to document our design and communicate with others. We must not allow for the weak link between our refined calculation and other parties to be a handwritten markup on a drawing. Not only does this ubiquitous paradigm have a risk of human error, but also, it slows our ability to convey our thoughts and threatens our ability to communicate with others parties such as sustainability or MEP engineers, contractors, and architects. Having a digital link between our analysis and drawings or 3D deliverables will be key for better communication with these other parties as well as increase our ability to cater to design changes that increase the livability of our structures.
Conclusion:
We have discussed how modern day structural engineers have lost our way, becoming technicians who create standard solutions rather than designers who balance the safety and elegance. If we are to break free of this trend, we must change our education schemes to provide engineers with the proper computational tools in order to handle the increasing complexity of projects. As seen in Figure 7, if we do not do this, we risk being marginalized by a future generation of professionals who will.
Figure 7: School for Young Children in Hong Kong Teaching Coding, Computer Science, and Robotics
The revolution is here; we must decide now if we will lead it or be replaced by it.