When I began my career, my first job title was “foundation engineer”. The title was an anachronism, even at the time, but the scope of my work was consistent with the domain of foundation engineer going back to the early 20th century, at least. I was involved in geotechnical subsurface investigation and classification, development of design parameters, structural design of foundation elements, earth-retaining systems and underground structures and construction consideration for these systems. Foundation problems have always been subject to uncertainty due to the inability to specify and fully observe the subsurface conditions. Therefore, foundation engineering has always relied on empiricism and judgment to provide reliable predictions about foundation performance.
Several prominent 20th-century foundation engineers have suggested that foundation engineering is as much, or more, art than science. But in the 21st century, it seems that much of that art has been lost. The uncertainty and risk associated with what was called foundation engineering remain, but the appreciation of empiricism and experience, which underlies foundation engineering as “art” seems to have been forgotten.
A lot of the art to foundation engineering problems seems to have been lost in the fragmentation of these problems to multiple parties on most construction projects. A geotechnical engineering firm, that is often really a materials testing lab, provides sampling and testing, along with a report that will include design parameters and recommendations. Based on this, the structural engineer of record will usually prepare the design of foundations and below-ground structures, along with construction documents. A contractor is responsible for construction, including the selection of means and methods. This may include the design of temporary works and delegated design elements, which may be performed by a consulting engineer or vendor, often through specialty contractors.
Typically, each party has well-defined responsibilities, but limited knowledge or interest of the other parties’ work. As a result, information has to be transmitted linearly, usually without the usual iteration inherent to good design, in a simple, lowest-common-denominator form. This process usually leads to excess conservatism and therefore, increased cost. However, this process is also error-prone because the upstream parties may not understand what the downstream parties need to execute their intent and the downstream parties may misinterpret what they have been given. A rigid process that prevents collaboration and incentivizes excess conservatism will not produce artful design
With the fragmentation of foundation engineering problems, there is often little incentive to take a holistic approach to design. A common consequence of this is a disconnect between the quality and quantity of data on which the design is based, the level of sophistication and complication of the design, and the design intent as manifest in the specifications and the construction process. I have written about facets of this problem here, here and here, among other places.
But I will say something about the approach to design because it has its own detrimental effect on the art of foundation engineering. For perhaps as long as there have been theoretical models in foundation engineering, there have been those cautioning about the limits of those models. Advancement in computing has greatly reduced the effort required to apply increasing complex theoretical models compared with the days of slide rules. Academia seems to increasingly focus engineering school curricula around computational mechanics and analysis, and this attitude rubs off on inexperienced engineers, particularly those with advanced degrees. The greater the sophistication of the model, the more elements, the more nodes, the more significant figures, the better. The importance of the data used as inputs and evaluating the veracity and meaning of the output is often hand waived. It’s seemingly computation for the sake of computation. (I have written about what engineers should know about their models here.)
In the fragmented world of foundation engineering, this results in modeling effort being out of proportion with the availability and reliability of available data. Some geotechnical and geostructural engineers are overly enamored with advanced computation methods and non-linear models and insist on their superiority, even in the absence of quality subsurface data and realistic structural inputs or even problem geometry. Some structural engineers will take an overly simplistic approach, treating soil pressures as a deterministic load, or taking the results of a linear soil spring model too literally or being unwilling to use design parameters that vary from code presumptive values. In either approach, when the numbers are assumed to be correct despite the limitations of the inputs used to produce them, the models become substitutes for experience and judgment, leading to inefficient and unreliable predictions. This is not conducive to the artful solution of problems.
What is needed is a more consistent approach that addresses foundation problems as a whole. The members of the design team need to have an appreciation and understanding of the construction of their work and how it relates to systems designed by other disciplines and specialists. The best consultants already have this. Depth of knowledge is important too, as it provides insight as to why things are the way they are. This allows one to understand how to apply a method and its limitations. The combination of the two provides perspective on a problem that facilitates collaboration and coordination while reducing misinterpretation and misuse of knowledge.
I personally find that reading historical engineering texts and references provide a useful perspective. The history of a particular approach to a problem (like bearing capacity) will illustrate its limitations. In addition, a lot was accomplished before the technology we have today. While some of the historic methods are no longer practical (moment distribution), a lot of heuristics and simplified approaches still have value and embody the experience of generations of practitioners. Most importantly, historical practice was, in a lot of respects, better balanced between art and science. Computation was expensive, so the application of experience and judgment was a practical necessity.
Perhaps the historical practice of foundation engineering as an interdisciplinary field is an artifact of the “master builder” era. It seems we are now all expected to be specialists and can only practice a subset of our chosen discipline. However, foundations are still interdisciplinary problems. While it is no longer the only thing I do, my old title still applies. I am going to keep calling myself a foundation engineer and practicing it as an art.
The information and statements in this document are for information purposes only and do not comprise the professional advice of the author or create a professional relationship between reader and author. The author has no personal involvement in the matter described.