This post was originally published on TekUK.org.

A digital twin is a virtual replica of the physical world, created using real-time sensors1. A digital twin’s ability to account for changes in its physical counterpart has enabled transformation from manufacturing and retail to transport and maintenance. As sustainability becomes an ever more pervasive concern in every industry, using digital twins to address some of the challenges and realise the opportunities is no exception to this trend. Sustainability is the concept that finds a way to balance the needs of current generations with the ability of future generations to meet their own needs2. Digital twins are being applied to address sustainability on a range of issues such as increasing energy efficiency in buildings, ‘greening’ agriculture, and in the design of smart cities2.  

To frame the urgent, global concern of sustainability, Earth Overshoot Day serves as an annual alarm clock, created to describe the date when humanity will use more natural resources than Earth can regenerate in a single year3,4. The date of each country’s ‘overshoot’ moves ever closer to January 1st each year; hence, we’ve almost exhausted our allocated resources on day one. In the UK, we barely reach May3. This is Earth’s Alarm – a call to arms to protect our planet. As our global ecological footprint continues to grow, we must devise a strategy to mitigate the overshoot. With the UK ranked well in ‘AI readiness’ (not strictly required for digital twins but represents the preparedness of the UK for advanced technologies) this could be the perfect opportunity for the UK to accelerate and scale the adoption of digital twins for sustainability5.  

So why aren’t we using digital twins more? Why aren’t we using them to truly transform global economies and industries to improve sustainability? If digital twins can allow for changes in their physical counterpart, it stands to reason that they could be employed across the whole gamut of challenges in the sustainability crusade. The most significant opportunity for digital twins is enabling the transition to a sustainable economy, with zero waste and the regeneration of nature. This is known as a circular economy6.  

Whilst digital twins have been used in individual (though no doubt still very important) aspects of sustainability, application at the level of a whole economy – and at the scale of a country or a continent has been largely absent. Some might argue ‘it will come’ – that these are just classic barriers to adoption faced by all new tech. However, it is more likely due to three inherent barriers of most designs and most uses of digital twins: 

  1. The Data Problem  
  2. The Scope Problem 
  3. The Feedback Loop Problem 

The Data Problem – there is uncertainty within most challenges of predicting and implementing a circular economy, and uncertainty within the data that is captured. Coupled with the requirement for often unrealistic volumes of sensor data, this makes the current approach to digital twins at this scale impractical.  As new technology such as IoT (Internet of Things) and XR (Extended Reality) emerge, many predict that every facet of existence will be captured – and therefore the required data will come (…eventually). However, most models of current and even generation-after-next digital twins are expected to require an almost unfathomable number of sensors7 . It may be years before this becomes a reality. This links closely to the Scope Problem. There is, in digital twins, an artificial focus on ‘physical’ simulation. This thwarts the recognition that sustainability issues are complex, socio-technical systems. These systems expand beyond physical measurements, to human emotions, economic incentives, legal frameworks, political doctrine, and even to cultures. This requires a new way of modelling and managing digital twins for sustainability. This brings us to the Feedback Loop Problem. The reality is that digital twins don’t currently ‘do’ much at all – humans do. So without the ability to plan, do, check and act, users of digital twins are unable to learn from their assumptions, change the ‘model’, try new things, and adapt to situations,  

So, how do we switch to a system where digital twins are less ‘data hungry’ and can cope with the enormous uncertainty all around? How do we switch to a system where all aspects of the circular economy, at a country level and beyond, are able to be predicted and tackled? How do we build into the design and use of digital twins a new way of making decisions? At Lone Star Analysis, we do this by embracing uncertainty and use Evolved AI to use the data we have – not the data we wish we had. We use Evolved AI to describe and predict any aspect (from the hard physics to soft systems of culture) that matters, We provide solutions that let decision-makers find – and act on- the levers that matter and learn from it when they do.  

The solution to sustainability may be to rethink and reframe Digital Twins. To predict, and to act without the need for extensive data inputs and integrate them into the management systems of businesses, communities, and governments.  

This could go a long way to help us stop pressing the snooze button on sustainability. 

References:  

  1. https://ieeexplore.ieee.org/abstract/document/9103025   
  2. https://onlinelibrary.wiley.com/doi/10.1002/jid.3380010208   
  3. https://www.overshootday.org/   
  4. https://www.pbctoday.co.uk/news/energy-news/sustainable-action-not-fast-enough-sustainable-futures-report-finds/119436/ 
  5. https://www.oxfordinsights.com/  
  6. https://www.europarl.europa.eu/news/en/headlines/economy/20151201STO05603/circular-economy-definition-importance-and-benefits   
  7. https://ellenmacarthurfoundation.org/topics/circular-economy-introduction/overview