This article was reprinted from Maxim Millen’s LinkedIn post
The engineering profession is advancing, advancing in the sense that engineers are dealing with more complex situations and are able to work faster due to automated design tools that can produce large design reports.
However, I consider this the dark under belly of engineering. It creates engineers who don’t think, they just do. And it exposes us to one of the biggest risks in engineering: your analysis model misses a fundamental mechanism of the real world situation. While there are other mistakes such as calculation errors, miscommunications and missing deadlines. There is no doubt that missing a failure mechanism in an engineering design or analysis can result in catastrophe.
To reduce the chance of a fundamental error it is important to operate in a way that allows your peers and mentors to review and critique your work. In my blog I previously talked about the importance of conventions within an engineering firm, as conventions provide a uniform structure to everyone’s work and therefore makes it easier for others to understand.
This article takes a more fundamental look at how the engineering profession is advancing and the principles we should use when adopting and developing processes within engineering. The principles are:
- Keep it Simple – avoid unnecessary complexity
- Use thought-provoking processes – avoid design processes that you
don’t need to think
- Communicate efficiently – Avoid long reports
3D modeling software, optimization algorithms and time-series simulation software all have an important role to play in engineering. However, many of the techniques used in the analysis are beyond our ability to communicate and fully understand, therefore we rely on our intuition or judgment when assessing the accuracy of these techniques. This exposes us to the big risk of missing a fundamental mechanism.
It is absolutely critical that all complex/automated design procedures are complimented with simple mechanism-based hand calculations.
This was the crux of my PhD thesis at the University of Canterbury (digitally available at the UC library) where I developed a simple hand calculation procedure to design buildings for soil-foundation-structure interaction. It was essential that the procedure was explicitly mechanism based. Even if it is crude it allows you to assess the magnitude and importance of each mechanism. In contrast, the numerical simulation of a soil-foundation-structure problem is plague by numerical instability issues and high sensitivity to damping considerations, therefore the focus of the engineer is spent on dealing with those issues rather than whether the fundamental deformation mechanisms are being considered.
Thoughtful or thought-provoking methods are an engineer’s best-friend. They encourage you to think about your problem while you are crunching the numbers. The opposite would be black-box automated design software, where you simply input some values and extract the output.
Unfortunately, most software seems to be a black-box because the designers want to hide all of their intellectual property. However, some software applications are quite the opposite, they provide great insight into the problem through clever visualisation or being explicit about what has been calculated.
Also, black-box processes are not constrained to software, some simple hand-calculation methods use abstracted factors that have no physical meaning and although you do all the calculations yourself, you gain no great insight into the problem.
The best example I can think of where a software provides a more thoughtful approach than a hand-calculation procedure is the calculation of bearing capacity. The typical Terzaghi formula uses vague correction factors for foundation shape and ground slope to adjust bearing capacity factors, which also struggle to derive any physical meaning, making the engineer simply lost until the final bearing capacity value is extracted. In contrast the software LimitStateGeo uses clever graphics to clearly demonstrate to the users what calculations are being run and the meaning behind them.
Efficiency is a word so tied to engineering that sometimes we forget to think about it. Of course we are looking for efficient designs that allow quick construction and minimal material. However, we should also apply efficiency to the way we convey technical information. I don’t mean we should write reports faster, in fact I mean we should write them more slowly and carefully. Technical work is read many more times than it is written. Spending a bit more time to write concisely and using images to convey messages is crucial. The same can be said about writing short emails, or using an alternative to emails for in-house communication such as Slack (slack.com). We use it internally at Pensolve. It is extremely simple and removes all the added “Hi John” and “Best regards” sorts of sign off and email signatures, so you can read messages like a conversation.
We as a profession are speeding up. What we chose to speed up and how we do it are important decisions. New processes are always being developed and old processes are becoming obsolete. By making sure your new processes are simple, thoughtful and efficient you are mitigating the risks around making a fundamental design or analysis error. The principles are not new, but perhaps now they are more important than ever.