Addressing societal challenges by engaging 100% of society’s intellectual capital

1 Reply

Decorative photograph of author's research groupToday is International Women in Engineering Day (INWED).  I have written previously about the lack of progress in achieving gender equality in the engineering profession in most Western countries (see ‘Reflecting on the lack of women in engineering’ on March 16th 2022) and the seismic shift in attitudes required to increase the number of women in engineering at all levels (see ‘A big question for engineers’ on June 8th, 2016).  I can see signs of change locally.  My research group has hovered around an equal number of men and women for some years.  In the School of Engineering in Liverpool four women have been promoted to be professors in the last four years – though at a reception following an inaugural lecture given by one of the pair of women who were the first female professors in the School, I was gently admonished by a senior female colleague in another school about why it had taken so long.  Of course, she is right.  Progress is very slow and we need to do better.

Simone de Beauvoir wrote in her book, The Second Sex first published in 1949: ‘Representation of the world, like the world itself, is the work of men; they describe it from their own point of view, which they confuse with absolute truth.’  More than seventy years later, this still appears to be true, at least in the engineering profession who are responsible for the technology everyone uses everyday.  About twenty years ago, the then President of the US National Academy of Engineering, Bill Wulf said, ‘As a consequence of a lack of diversity [in engineering] we pay an opportunity cost, a cost in designs not thought of, in solutions not produced’.  However, diversity on its own is not enough, we have to be inclusive and treat everyone equally – as we would like to be treated ourselves.  If we do not engage women in the engineering enterprise then we ignore 50% of society’s intellectual capital and we cannot hope to solve the challenges facing society, in part because we will be confused about the truth.

Thank you to my two guest editors who reviewed this post for me.

Photo: Author’s research group in 2022.

 

Fairy lights and decomposing multi-dimensional datasets

Leave a reply

A time-lapsed series of photographs showing the sun during the day at North Cape in NorwayMany years ago, I had a poster that I bought when I visited North Cape in Norway where in summer the sun never sets.  The poster was a time-series of 24 photographs taken at hourly intervals showing the height of the sun in the sky during a summer day at North Cape, similar to the thumbnail.  We can plot the height of the sun as a function of time of day with time on the horizontal axis and height on the vertical axis to obtain a graph that would be a sine wave, part of which is apparent in the thumbnail.  However, the brightness of the sun also appears to vary during the day and so we could also conceive of a graph where the intensity of a line of symbols represented the height of the sun in the sky.  Like a string of fairy lights in which we can control the brightness of each one individually  – we would have a one-dimensional plot instead of a two-dimensional one.  If we had a flat surface covered with an array of lights – a chessboard with a fairy light in each square – then we could represent three-dimensional data, for instance the distribution of elevation over a field using the intensity of the lights – just as some maps use the intensity of a colour to illustrate elevation.  We can take this concept a couple of stages further to plot four-dimensional data in three-dimensional space, for instance, we could build a three-dimensional stack of transparent cubes each containing a fairy light to plot the variation in moisture content in the soil at depths beneath as well as across the field.  The location of the fairy lights would correspond to the location beneath the ground and their intensity the moisture content.  I chose this example because we recently used data on soil moisture in a river basin in China in our research (see ‘From strain measurements to assessing El Nino events’ on March 17th 2021).  We can carry on adding variables and, for example if the data were available, consider the change in moisture content with time and three-dimensional location beneath the ground – that’s five-dimensional data.  We could change the intensity of the fairy lights with time to show the variation of moisture content with time.  My brain struggles to conceive how to represent six-dimensional data though mathematically it is simple to continue adding dimensions.  It is also challenging to compare datasets with so many variables or dimensions so part of our research has been focussed on elegant methods of making comparisons.  We have been able to reduce maps of data – the chessboard of fairy lights – to a feature vector (a short string of numbers) for some time now [see ‘Recognizing strain’ on October 28th, 2015 and ‘Nudging discoveries along the innovation path’ on October 19th, 2022]; however, very recently we have extended this capability to volumes of data – the stack of transparent cubes with fairy lights in them.  The feature vector is slightly longer but can be used track changes in condition, for instance, in a composite component using computer tomography (CT) data or to validate simulations of stress or possibly fluid flow [see ‘Reliable predictions of non-Newtonian flows of sludge’ on March 29th, 2023].  There is no reason why we cannot extend it further to six or more dimensional data but it is challenging to find an engineering application, at least at the moment.

Photo by PCmarja2006 on Flickr

Reasons I became an engineer: #4

Leave a reply

Images from the optical microscope showing the tracks of bacteria interacting with a surfaceThis is the last in a series of posts reflecting on my steps towards becoming an engineer.  At the end of the previous post, I described how I moved to Canada becoming a biomedical engineer in the Medical School at the University of Calgary.  It was a brief period of my career, because shortly after I started, I was encouraged to apply for a lectureship in mechanical engineering at my alma mater which I did successfully.  So, I returned to the University of Sheffield and started my career as an academic engineer.  I continued to work in biomedical engineering, focussing initially on cardiac mechanics [see ‘Tears in the heart’ on July 20th, 2022], then on osseointegrated prostheses [see ‘Turning the screw in dentistry’ on September 9th, 2020] and, more recently, on computational biology [see ‘Hierarchical modelling in engineering and biology’ on March 14th, 2018] and cellular dynamics [see ‘Label-free real-time tracking of individual bacterium’ on January 25th, 2023].  However, the dominant application area of my research has been aerospace engineering informed by, if not also influenced by, my experiences in the Royal Navy, including flying a jet trainer aircraft shortly before leaving.  In the last decade, I have been introduced to nuclear reactor engineering, both fission and fusion, and have used them as vehicles for developing research in digital engineering [see ‘Thought leadership in fusion engineering’ on October 9th, 2019].  This biographical series of posts has described my evolution as an engineer – it was not an ambition I ever had nor did anyone push me towards engineering but I have found that my way of thinking about problems is well-suited to engineering, or perhaps engineering has taught me a way of thinking.

Image: Figure 4 – Tracks (yellow lines) of the sections (purple circles) of four E. coli bacteria experiencing: (a) random diffusion above the surface; (b) rotary attachment; (c) lateral attachment; (d) static attachment. The dynamics of the four bacteria was monitored for approximately 20 s. The length of the scale bars is 5 μm. From Scientific Reports, 12:18146, 2022.

How many engineers do you need when the lights go out?

Leave a reply
An exemplar adverse outcome pathway for microplastics in aquatic species

From Galloway & Lewis, 2016

One to change the lightbulb and five to perform a Fault Tree Analysis (FTA).  A fault tree is a diagram that illustrates the relationship between failures at component and system levels.  Engineers use them to understand the mechanisms or logic that lead from component malfunctions to system breakdowns and to identify components that are critical to system reliability.  They are useful in optimizing designs, demonstrating compliance with safety requirements and as diagnostic tools when things go wrong.  There are some simple examples of fault trees for ‘no light in room’ and ‘missing the bus’ amongst others available from Visual Paradigm Online.  All of these examples illustrate qualitative relationships but we can also establish quantitative relationships using the rate of occurrence of each initiating event to arrive at a probability of failure (PoF) for the system.  There is an example for an indicator light in an automobile in a 2016 paper by Nabarun Das and William Taylor (see figure 2 in the paper).  An equivalent in biology are Adverse Outcome Pathways (AOPs) that identify the relationship between a molecular initiating event and a toxic effect through a series of key events.  For instance, microplastics causing altered gene expression and oxidative damage leading to altered fatty acid metabolism, stress response and altered cellular division resulting ultimately in population decline in aquatic species as shown in the graphic from a paper by Tamara Galloway and Ceri Lewis also published in 2016. Most AOPs are qualitative; however, quantitative Adverse Outcome Pathways (qAOPs) are starting to be developed as tools for quantitative risk assessment of chemicals.  Biologists and engineers are not using the same words, actually they are using entirely different vocabularies; nevertheless they are talking about the same methodologies.  An AOP network and an FTA are essentially the same concept and a probabilistic fault tree analysis is a quantitative adverse outcome pathway.  However, it seems unlikely that either biologists or engineers will adopt the language used by the other so they will be reliant on a few foolhardy interlocutors prepared to cross the discipline boundaries and highlight the opportunities for cross-fertilization of ideas and solutions.

Sources

Das N, Taylor W. Quantified fault tree techniques for calculating hardware fault metrics according to ISO 26262. In2016 IEEE Symposium on Product Compliance Engineering (ISPCE), pp. 1-8. IEEE, 2016. Also available at https://incompliancemag.com/article/quantified-fault-tree-techniques-for-calculating-hardware-fault-metrics-according-to-iso-26262/

Galloway TS, Lewis CN. Marine microplastics spell big problems for future generations. Proceedings of the national academy of sciences. 113(9):2331-3, 2016.