How a digital twin can deliver immeasurable benefits without immeasurable costs
Author: Brian Mackay, programme delivery consultant at Costain
The benefits of having a digital twin for just one infrastructure asset are many: This single source of information can reduce the number of site visits, enhancing safety and minimising downtime, as well as minimise the risk of delays and escalating costs to capital projects, to name a few. The benefits of a digital twin at a national scale would be immeasurable. As part of pioneering leading-edge digital technology solutions that will be fundamental to how we shape our future infrastructure and ways of working, Costain is involved in the National Infrastructure Commission’s journey to create a National Digital Twin1. Affectionately becoming known as the ‘BritTwin’, it aims to catalyse innovation and deliver greater customer benefit whilst reducing costs through new insights from the data we already have. However, in the infrastructure sector we are far from starting with a blank sheet of paper.
With the exception of a handful of major infrastructure projects in the UK and those constructed over the past decade in line with BIM principles, we have an infrastructure asset base which is primarily designed and constructed in isolation of the infrastructure companies’ business systems and with records that may have been completed in a digital form (if you are lucky) but transferred to the business in a traditional analogue workflow. Our infrastructure industries still operate Victorian assets with Victorian records. We do have the tools and capability to digitise our records right now and it is easy to be enticed into creating deeply detailed digital twins of new assets and a national digital twin at that. Again, all of this is great, but the money and time this will take is potentially as immeasurable as the benefits it will bring.
To help put a more finite figure on the costs, we need to be clear on how much history is incorporated and to what level of detail the digital twin is built. To do this, the first critical step is to be sure what we are asking the digital twin to do. Defining the level of detail for a digital twin of our historic infrastructure ahead of the twin’s purpose will not only lead to undue effort and cost but could also lead to misunderstandings between asset owners and their stakeholders as well as contractual disputes about how far to take its development. If the investment in a digital twin does not achieve the goal of delivering greater customer benefits and reducing costs, then the business case for a digital twin will not stack up.
So how do you clarify the purpose and define the level of detail for a digital twin?
As part of helping our water client optimise their asset, Costain is involved in a proof of concept project which is looking to build the foundation of a digital twin for a wastewater treatment plant site. We had recently constructed a new installation on the site and had comprehensive records built to modern digital information (BIM) standards and extensive use of supply chain integrated 3D modelling. This new installation made up around 25% of the assets on the entire site. We set out to see if it was possible to create the first generation of a digital twin of the whole site through which we could visualise and access information. Our deep understanding of our client’s operations as well as our experience of designing, building and maintaining the physical assets meant we could easily construct a business case that investing in a digital twin now would remove 40% inefficiency. This would be in future workflow attributed to the manual exercise of engineers searching for historic information about old assets before they can begin designing modifications, additions or replacements. The benefits of having a digital twin that is a component level virtual replica, as it is in the manufacturing space, are extremely enticing. As well as the benefits cited above, further benefits include being able to run impact scenarios of weather events on our networks, feeding real time information back into optimised designs, operations oversight and control and predictive maintenance of systems and assets. We could quite easily run away with spending money on a detailed, reverse engineered digital design of old tanks, pipes and pump sets. However, that may not deliver any further benefit to what we need the digital twin to tell us at this point in time and in the immediate future.
We are well underway with the project and the learnings so far have tested and challenged the level of detail we have chosen to adopt in creating the realistic representation of the physical asset. Due to the legacy of existing assets that would require scanning and digital modelling, a deep domain knowledge of the industry and its assets has been critical in defining a level of detail that was suitable for the purpose.
It was also critical to adopt a systems thinking approach that would enable us to consider how the asset would work together with other assets and the wider network to achieve the overall outcomes. Systems thinking can help ensure that the right problem is being solved, as well as being solved in the right way and has enabled us to define the purpose of the digital twin for now and thus the level of investment required for its current operator to maximise efficiency. It wasn’t worth developing a fully detailed digital twin when balancing cost to benefit to the immediate problem that needed solving.
Our view was that to unlock the potential of a digital twin you had to start with an open platform that could capture and host a visual representation of the physical asset to a level of detail that could be built upon and in the future. Systems thinking can help us to see what future problems that digital twin may need to solve and therefore justify the cost as this stage. Alternatively, it can help us decide to wait until there are more financially viable technologies available. Much like how the US military Global Positioning System (GPS) development was created for one purpose. The data it generates enabled the foundation of other technologies such as satellite navigation and creation of companies like Uber.
We therefore must approach the development of our utility infrastructure digital twin with a vision that enables us to redefine the level of detail in the future as the value for doing so can be realised. This will then act as a launch pad for all kinds of possibilities we cannot even begin to conceive in our current state. To be able to define the level of detail and how much history to incorporate now, it is critical to have deep domain knowledge, take a systems thinking approach and fundamentally, be clear about what we are asking it to do.
Found this article interesting? You might also like to read The relationship between 'systems thinking' and 'systems engineering' and When the going gets tough... by Jeremy Dick.