Category Archives: design

Tacit hurdle to digital twins

Tacit knowledge is traditionally defined as knowledge that is not explicit or that is difficult to express or transfer from someone else. This description of what it is not makes the definition itself tacit knowledge which is not very helpful. Management guides resolve this by giving examples, such as aesthetic sense, or innovation and leadership skills which are elusive skills that are hard to explain [see ‘Innovation out of chaos‘ on June 29th 2016 and ‘Clueless on leadership style‘ on June 14th, 2017]. In engineering, there are a series of skills that are hard to explain or teach, including creative problem-solving [see ‘Learning problem-solving skills‘ on October 24th, 2018], artful design [see ‘Skilled in ingenuity‘ on August 19th, 2015] and elegant modelling [see ‘Credibility is in the eye of the beholder‘ on April 20th, 2016]. In a university course we attempt to lay the foundations for this tacit engineering knowledge; however, much of it is gained in work through experience and becomes regarded by organisations as part of their intellectual assets – the core of their competitiveness and source of their sustainable technology advantage. In our work on integrated nuclear digital environments, from which digital twins can be spawned, we would like to capture both explicit and tacit knowledge about complex systems throughout their life cycle which will extend beyond the working lives of their designers, builders and operators. One of the potential advantages of digital twins is as a knowledge management system by duplicating the life of the physical system and thus allowing its safer and cheaper operation in the long-term as well as its eventual decommissioning. However, besides the very nature of tacit knowledge that makes its capture difficult, we are finding that its perceived value as an intellectual asset renders stakeholders reluctant to discuss it with us; never mind consider how it might be preserved as part of a digital twin. Research has shown that tacit knowledge sharing is influenced by environmental factors including national culture, leadership characteristics and social networks [Cai et al, 2020]. I suspect that all of these factors were present in the heyday of the UK civil nuclear power industry when it worked together to construct advanced and complex systems; however, it has not built a power station since 1995 and, at the moment, new power stations are cancelled more often than built, which has almost certainly depressed all of these factors. So, perhaps we should not be surprised by the difficulties encountered in establishing an integrated nuclear digital environment despite its importance for the future of the industry.

Reference: Cai, Y., Song, Y., Xiao, X. and Shi, W., 2020. The Effect of Social Capital on Tacit Knowledge-Sharing Intention: The Mediating Role of Employee Vigor. SAGE Open, 10(3), p.2158244020945722.

Alleviating industrial uncertainty

2 Replies

Want to know how to assess the quality of predictions of structural deformation from a computational model and how to diagnose the causes of differences between measurements and predictions? The MOTIVATE project has the answers; that might seem like an over-assertive claim but read on and make your own judgment. Eighteen months ago, I reported on a new method for quantifying the uncertainty present in measurements of deformation made in an industrial environment [see ‘Industrial uncertainty’ on December 12^th, 2018] that we were trialling on a 1 m square panel of an aircraft fuselage. Recently, we have used the measurement uncertainty we found to make judgments about the quality of predictions from computer models of the panel under compressive loading. The top graphic shows the outside surface of the panel (left) with a speckle pattern to allow measurements of its deformation using digital image correlation (DIC) [see ‘256 shades of grey‘ on January 22, 2014 for a brief explanation of DIC]; and the inside surface (right) with stringers and ribs. The bottom graphic shows our results for two load cases: a 50 kN compression (top row) and a 50 kN compression and 1 degree of torsion (bottom row). The left column shows the out-of-plane deformation measured using a stereoscopic DIC system and the middle row shows the corresponding predictions from a computational model using finite element analysis [see ‘Did cubism inspire engineering analysis?’ on January 25^th, 2017]. We have described these deformation fields in a reduced form using feature vectors by applying image decomposition [see ‘Recognizing strain’ on October 28^th, 2015 for a brief explanation of image decomposition]. The elements of the feature vectors are known as shape descriptors and corresponding pairs of them, from the measurements and predictions, are plotted in the graphs on the right in the bottom graphic for each load case. If the predictions were in perfect agreement with measurements then all of the points on these graphs would lie on the line equality [y=x] which is the solid line on each graph. However, perfect agreement is unobtainable because there will always be uncertainty present; so, the question arises, how much deviation from the solid line is acceptable? One answer is that the deviation should be less than the uncertainty present in the measurements that we evaluated with our new method and is shown by the dashed lines. Hence, when all of the points fall inside the dashed lines then the predictions are at least as good as the measurements. If some points lie outside of the dashed lines, then we can look at the form of the corresponding shape descriptors to start diagnosing why we have significant differences between our model and experiment. The forms of these outlying shape descriptors are shown as insets on the plots. However, busy, or non-technical decision-makers are often not interested in this level of detailed analysis and instead just want to know how good the predictions are. To answer this question, we have implemented a validation metric (VM) that we developed [see ‘Million to one’ on November 21^st, 2018] which allows us to state the probability that the predictions and measurements are from the same population given the known uncertainty in the measurements – these probabilities are shown in the black boxes superimposed on the graphs.

These novel methods create a toolbox for alleviating uncertainty about predictions of structural behaviour in industrial contexts. Please get in touch if you want more information in order to test these tools yourself.

The MOTIVATE project has received funding from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 754660 and the Swiss State Secretariat for Education, Research and Innovation under contract number 17.00064.

The opinions expressed in this blog post reflect only the author’s view and the Clean Sky 2 Joint Undertaking is not responsible for any use that may be made of the information it contains.

Reinforcement ensures long-term structural integrity

When seeing nothing is a success

2 Replies

In November I went to Zurich twice: once for the workshop that I wrote about last week [see ‘Fake facts and untrustworthy predictions’ on December 4^th, 2019]; and, a second time for a progress meeting of the DIMES project [see ‘Finding DIMES’ on February 6^th, 2019]. The progress meeting went well. The project is on schedule and within budget. So, everyone is happy and you are wondering why I am writing about it. It was what our team was doing around the progress meeting that was exciting. A few months ago, Airbus delivered a section of an A320 wing to the labs of EMPA who are our project partner in Switzerland, and the team at EMPA has been rigging the wing section for a simple bending test so that we can use it to test the integrated measurement system which we are developing in the DIMES project [see ‘Joining the dots’ on July 10^th, 2019]. Before and after the meeting, partners from EMPA, Dantec Dynamics GmbH, Strain Solutions Ltd and my group at the University of Liverpool were installing our prototype systems to monitor the condition of the wing when we apply bending loads to it. There is some pre-existing damage in the wing that we hope will propagate during the test allowing us to track it with our prototype systems using visible and infra-red spectrum cameras as well as electrical and optical sensors. The data that we collect during the test will allow us to develop our data processing algorithms and, if necessary, refine the system design. The final stage of the DIMES project will involve installing a series of our systems in a complete wing undergoing a structural test in the new Airbus Wing Integration Centre (AWIC) in Filton, near Bristol in the UK. The schedule is ambitious because we will need to install the sensors for our systems in the wing in the first quarter of next year, probably before we have finished all of the tests in EMPA. However, the test in Bristol probably will not start until the middle of 2020, by which time we will have refined our algorithm for data processing and be ready for the deluge of data that we are likely to receive from the test at Airbus. The difference between the two wing tests besides the level of maturity of our measurement system, is that no damage should be detected in the wing at Airbus whereas there will be detectable damage in the wing section in EMPA. So, a positive result will be a success at EMPA but a negative result, i.e. no damage detected, will be a success at Airbus.

The DIMES project has received funding from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 820951.