Engineering in 2076?

Leave a reply

Today is National Engineering Day which is organised annually by the Royal Academy of Engineering.  The Academy is celebrating its 50th anniversary this year not by looking back, but by looking fifty years ahead.  Recently, as a fellow of the Academy, I was asked to participate in a survey about what engineering jobs of the future might really look like.  This is a challenging question, because it is very difficult to predict what technology we might have developed by 2076.  I am not old enough to have been an engineer in 1976 when the Academy was founded, but I graduated the following decade becoming an engineer by chance rather than design [see ‘Reasons I became an engineer: #1’ on April 4th, 2023].  In the mid-80s, computational modelling of engineering structures involved writing hundreds of lines of computer code, waiting overnight for the computer to process them, and trawling through pages of numerical output the following day.  Now, we can create and process hugely more complex models of systems, rather than single components, almost instantaneously and the results can be viewed in an immersive virtual reality.  I cannot imagine how this aspect of engineering activity will change in the next decade with the emergence of AI, never mind the next fifty years.  However, the role of the engineer has not really changed in the last fifty years: engineers make things happen [see ‘Making things happen’ on September 26th, 2018].  Engineering is about using human ingenuity to produce something useful; preferably something that improves the quality of life [see ‘Engineering is all about ingenuity’ on September 14th, 2016].  I think this will not change over the next fifty years, nor will the skills required to be an engineer: creativity, complex problem-solving, communication and teamwork.  These skills are highly transferable and hence will remain useful regardless of new technologies that appear in the next fifty years.  My evidence for making this confident prediction is that I acquired these skills in the context of naval engineering but today I use them in fields as diverse as additive manufacturing [see ‘Star sequence minimises distortion’ on July 2nd, 2025], digital twins [see ‘Mapping the road between digital twins and NDE’ on September 2nd, 2026] and microbiology [see ‘Real-time label-free tracking of bacteriophages interacting with bacteria’ on March 3rd, 2026] using technology that had barely been dreamt of fifty years ago.  Isabella Loaiza and Roberto Rigobón at MIT proposed something similar in their EPOCH framework of human capabilities that AI is unlikely to replace.  EPOCH stands for Empathy, Presence, Opinion, Creativity and Hope.

Reference:

Loaiza I & Rigobón R, The EPOCH of AI: Human-machine Complementarities at Work, (November 21, 2024). MIT Sloan Research Paper No. 7236-24, Available at SSRN: https://ssrn.com/abstract=5028371 or http://dx.doi.org/10.2139/ssrn.5028371

All arms and legs

Leave a reply

Imagine building a humanoid robot with head, arms and legs from different organisations that will not share the technical details of the products, probably because they are worried about their intellectual property being stolen.  The likely result would be a robot that moves in slow uncoordinated jerky steps, unlike those seen at the recent World Humanoid Robot Games in China where a humanoid robot beat Usain Bolt’s 100m record.  We have encountered this problem in attempting to build an industrial metaverse consisting of a connected series of digital twins [see ‘Digital twins that thrive in the real-world’ on June 9th, 2021] that represent different parts of a complex engineering system, e.g., a power station.  The organisations that have developed the digital twins or computer simulations are reluctant to share technical information because they want to protect their intellectual property.  Our solution has been to develop an apex software architecture that seamlessly and securely connects computer models of all of the parts of the system with high-speed two-way exchange of information – the equivalent of the central nervous system in the humanoid robot.  Our architecture reduced the processing times for simulations of a nuclear reactor sixteen-fold compared to current approaches and required one-third of the number of analysts to perform the simulations.  The architecture is readily adaptable too; you can plug n’ play with different computational models – the equivalent to changing a leg on a humanoid robot using a snap-fit connector.  Read our 2022 paper if you would like to know more or get in touch if you would like to use our apex architecture.

Image: Figure 9b from Bowman et al, 2022, showing simulated power distribution and control rod position overlayed on a model of a PWR reactor.

Reference: Bowman D, Dwyer L, Levers A, Patterson EA, Purdie S, Vikhorev K. A unified approach to digital twin architecture—proof-of-concept activity in the nuclear sector. IEEE Access, 10:44691-709, 2022

Mapping the road between digital twins and NDE

1 Reply

RoadmapMichael Grieves, who probably first formulated the concept of digital twins in 2002, stated that all information about a physical entity should be present in its cyber or digital twin with data flow from the physical entity to the digital twin, and information flow from the digital twin to the physical entity throughout its lifecycle [1] [see ‘Digital twins that thrive in the real-world’ on June 9th, 2021].  The realisation of data flows between cyber-physical twins has been slow and there are relatively few cases where this definition has been fully implemented.  Non-destructive evaluation (NDE) is a well-established set of techniques used to assess the properties, integrity and condition of engineering components and structures that could play a significant role in supplying data to digital twins.  At the moment NDE is commonly used in quality control to check for manufacturing defects, in integrity assessments to investigate damage in safety critical components during maintenance inspections, and for determining the remaining life expectancy of engineering parts.  The integration of NDE in digital engineering has been slow; so, in a recent study, involving representatives from a wide range of industries, we have defined in a roadmap [2] the potential opportunity and the route to its realisation.  This work represents a different approach to research for my group in that we have explored the need or the ‘pull’ from industry rather than delivering innovative technology which can result in a ‘push’ to industry [see for example, ‘Our last DIMES’ on September 22nd, 2021].  My regular readers might also detect the influence of social science, which also surfaced in our recent work on knowledge management [see ‘Opportunities lost in knowledge management using digital technology’ on October 25th, 2023 and ‘Evolutionary model of knowledge management’ on March 6th 2024].  We have used the same methods, based on semi-structured interviews and thematic analysis, across all of these studies.

Image: roadmap [© University of Liverpool] which should be read in conjunction with [2].

References:

  1. Grieves, M., Vickers, J. (2017). Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems. In: Kahlen, J., Flumerfelt, S., Alves, A. (eds) Transdisciplinary Perspectives on Complex Systems. Springer, Cham.
  2. Middleton CA, Nguyen T & Patterson EA, NDE4.0 and digital twins in industry: current user experiences and future development, J. Nondestructive Evaluation, 45:107, 2026.

Iron Nanochains

Leave a reply

Caustics formed from iron nanoparticles coalescing into chains aligned with the lines of the magnetic field from an adjacent magnet

Many people will be familiar with an iron filings and magnet experiment often performed in high school physics classes.  Iron filings on a sheet of paper with a magnet underneath, when the paper is gently tapped, align with the lines of the magnetic field sometimes forming ‘chains’.  In our research group, we have performed a version of this experiment at the nanoscale using iron nanoparticles smaller than 200nm in diameter floating in deionised water on a microscope slide.  We brought a magnet into close proximity with the slide and the nanoparticles formed chains aligned along the lines of magnetic field which we visualised using caustics [see ‘Holes in fluids’ on October 22, 2014].  When we removed the magnet, the nanoparticles resumed their Brownian motion.  Not a ground-breaking experiment but it’s exciting to be able to demonstrate that the phenomenon occurs at the nanoscale and that we can observe it.