Digital twins vary based on their providers and the use case but, fundamentally, they have the ability to ingest real-world operational and environmental data and aggregate and combine it with other relevant data. The data is modelled and analysed through algorithmic simulations and visualisation to produce insights. Actions from the digital twin to the physical infrastructure or process can be through actuators or via human intervention.
Digital twins offer a wide range of applications for place-based organisations, including strategic planning, operational planning, and active operations management. While most organisations will begin their digital twin journey with a specific project or goal, it’s important to choose technology that is future ready to support a wide range of use cases, from real-time vehicle management to improving the design of future air mobility vehicles and dynamic grid load balancing.
Core capabilities to consider include the ability to ingest and fuse data from different sources, including real-time data, and the scope to scale the digital twin to more use cases and add extra functionality over time. Another key consideration is the ability to run and optimise multiple scenarios.
INFORMED DECISIONS, LIVING TWINS
Cities outside the UK are already extolling the benefits of digital twins. Virtual Singapore, a dynamic 3D city model and collaborative platform, brings together data from various public agencies and other sources and includes static, dynamic, and real-time data and information such as demographics, movement and climate.
According to Singapore’s National Research Foundation: “By leveraging the big data environment and aggregating information from the public and private sector, the potential uses of Virtual Singapore in tackling liveability issues are limitless.”
For example, urban planners can visualise the effects of constructing new buildings or installing green roofs and understand the impact on the temperature and light intensity in the area. They can also overlay heat and noise maps for simulation and modelling of the most comfortable and cool living environments.
Virtual Singapore also helps to prioritise buildings for solar panel installation, based on interlinked data such as the height of buildings, surface of the rooftops and amount of sunlight. Further analysis can allow planners to estimate how much solar energy can be generated on a typical day, as well as the energy and cost savings. By cross-referencing with the historical data collected from neighbouring buildings, this analysis can be validated and seasonally adjusted to reflect an even more accurate and granular projection.
Another pioneering example, due to its scale, is the City of Seoul’s S-Map digital twin which includes street topography, underground infrastructure, subway routes, real-time information on road traffic, and even street views of small marketplaces and alleys.
“It’s the first of its kind in that although there have been 3D maps in Sejong, Busan and Jeonju, they didn’t cover the whole area of the city,” Seoul city government said in a statement last year. “This digital twin of the city will allow us to make simulations online to predict the impact of disasters and accidents, such as wildfires, and predict how new buildings and structures built in the city could interfere with the wind patterns of the area and possibly worsen fine dust pollution.”
Through digital twins, organisations can do more with their data by gleaning actionable insights from it. A key benefit is operational efficiency.
Linda Wade, Co-founder and CEO of digital twin start-up Spinview, says: “As human beings we’re visual and we are built to understand information spatially. Visual recreation is a way of translating all the data from a space into something that is usable and manageable by every employee.
“Quite often, data sits in silos from the top teams and the engineering teams and if you make it available in a format that everyone can use, you can drive efficiencies that previously weren’t possible at scale.”
Transport for London (TfL) is using a digital twin from Spinview as part of its Piccadilly line upgrade which will see the introduction of 94 new trains, more capacity with a more frequent service, and reduced energy consumption.
“A huge amount of the work is actually in readying the infrastructure for those new trains,” says Robert Frith, Head of Engineering for the Piccadilly line upgrade at TfL. “There’s a full end-to-end infrastructure upgrade that is linked to the new trains coming into service.”
This includes considerations related to the size and the shape of the tunnels, as well as about power assets, platforms and more. Differing levels of data for different types of assets throughout the line was causing complexity.
“With the digital twin, we wanted an environment that we can work within that’s consistent for the whole line, and that is able to present the data,” adds Frith.
The project will initially focus on creating a digital replica of the Piccadilly line, which is part of the ‘deep-level’ Tube network with track in some locations as far as 30 metres below ground.
Spinview’s Deep Rail Sensor (DRS) device, attached to a train, will capture the geometry of the space and measures environmental pollutants as well as noise and heat.
The digital twin could be used throughout the lifecycle of project delivery – from surveying, planning, designing, and upgrading, to virtual approvals from TfL’s senior managers. The initiative also aims to help TfL better manage assets and support Mayor Sadiq Khan’s goal of a zero-carbon rail system by 2030.
Importantly, TfL staff will no longer be restricted to physically inspecting and surveying assets below ground during engineering hours from 1am to 5am – they will be able to conduct surveys, assess locations and make decisions at any time via the digital twin.
Wade says the digital twin will also reveal information that may have previously been “hidden from the human eye,” such as faults, and heat and noise hotspots. This data could then be used by TfL to make targeted improvements to the Tube network.
She comments: “By establishing a visually intelligent ecosystem for TfL, Spinview hopes to not only help TfL in tracking its environmental goals, but also to support in facilitating better decision-making, driving forward business efficiencies, saving costs and future-proofing the network in the long-term.”
Visualisation is also becoming an increasingly important aspect in resident engagement – to enable planners and space owners to bring more residents and customers into the conversation about the future development of the places they live.
Tallinn in Estonia, for example, has announced a new hub in the city centre which will use digital twins and virtual reality to foster more input on urban development, transport planning and city greenery, including the ability for citizens to simulate various scenarios.
PEOPLE AT THE CENTRE
With this in mind, people are becoming central to place-based digital twins, in an evolution from their predominantly asset-based roots. This approach will help to ensure that people have great experiences in complex places, and will also inform better infrastructure investment decisions.
With a decades-long background in using digital replicas in the automotive industry and then crowd simulation tools in rail infrastructure projects, Nicolas Le Glatin went on to bring these together through his start-up OpenSpace. The company is using its passenger flow digital twin technology to bring the customer experience into large-scale architecture, engineering and construction (AEC) projects.
“Cities are a series of experiences that fundamentally deliver the fabric of life or society, but engineers still tend to think about pouring concrete and moving cars and trains,” says Le Glatin.
This approach is important given the ‘passenger first’ mandate of organisations such as Great British Railways, the new public body that will oversee rail transport in Britain from 2023, as well as the wider push to encourage people back to public transport and to choose mass transit over the private car.
The OpenSpace platform analyses historic, real-time and predictive data on people movement, fusing new and existing data sources such as 3D lidar asset scans and passenger flow from computer vision, with train timetable feeds and Wi-Fi analytics as well as other information related to managing the customer experience. At St Pancras, this includes data from escalator company Schindler and contract cleaning firm Churchill.
“Companies like HS1 are really in the business of monetising the movement of people by train but also leveraging the quite remarkable retail opportunities,” says Le Glatin. “A digital twin is a means of enabling them to continuously optimise against a set of objectives.”
Following the development of a people-flow-management digital twin for St Pancras International station in 2020, which was later adapted to help manage social distancing, OpenSpace has been contracted to expand its digital twin infrastructure to all four HS1 stations: Stratford International, Ebbsfleet International, and Ashford International, alongside St Pancras. It has also been selected as part of the HS2 innovation programme and is delivering a ‘living lab’ at Euston station where HS2 is looking at how to design stations around the movement of people with real-time data.
This includes the configuration of buildings, the design and scheduling of services, and other real-time decisions which help to maximise revenue while minimising operating costs. The tool can, for example, guide when an escalator could be switched off to save energy and enable predictive maintenance based on footfall, or inform the facilities cleaning schedule on a weekly or daily basis.
“People are the common denominator between all the contributors of the built nvironment so a digital twin for managing people flow does more than making great decisions; it brings people together,” adds Le Glatin.
“It is astonishing how people are engaged or willing to be part of it and want to share data when you say we are going to optimise the end-to-end journey of their customers by collaborating with other rail operators.”
These collaborative examples point towards an emerging trend – the need for individual digital twins to be interoperable and ‘talk’ to each other to meet the integrated needs of places.
This is something that will continue to be a focus for the Digital Twin Hub which from 1 April will transition from its current home at the Centre for Digital Built Britain to Connected Places Catapult, offering new opportunities for place-based organisations and supporting companies to launch projects focused on digital twins.
Tamar Loach, Technology Initiative Director at Connected Places Catapult, points to a pioneering digital twin example in the Climate Resilience Demonstrator (CReDo), which is a particularly important project in the light of a spate of extreme weather events around the world and warnings from the Intergovernmental Panel on Climate Change that these will get worse.
The initial one-year phase of CReDo focused on proof of concept and showing how data from energy, water and telecoms networks can be connected up in one secure digital twin. It demonstrated the use of synthetic data, legal and data-sharing agreements, and the development of initial multi-level ontologies and an information management framework to enable data to interoperate.
“It’s bringing together different sectors that haven’t historically worked together,” says Loach. “They’re actually working together, sharing data, and therefore they’re able to look at their own systems and the investment needed to reinforce those systems so that they’re robust and resilient. And they’re able to look at that in the context of the systems that are affected and affecting each other.”
GETTING DIGITAL-TWIN READY
While there is a lot of excitement around digital twins, the featured projects also highlight the practical groundwork still to do to unleash their true power in cities and spaces.
Research commissioned by Connected Places Catapult outlined key competencies that place-based organisations need to develop to overcome barriers to adopting digital twins, such as lack of buy-in from the top, skills gaps, and limited data availability and quality.
The report advocates a range of practical steps including identifying and sharing successful digital twin case studies, facilitating partnership building, and growing digital operations competencies.
Loach says: “Finance and advertising, for example, are much more advanced in how they make use of data in a joined-up way. And the opportunity for places to take advantage of data and interoperable digital tools is huge, but there’s quite a long way to go and we can’t underestimate the digital maturity aspect.”
To gain confidence and demonstrate the value of a digital twin, experts urge place-based organisations to avoid the data-lake approach and instead pick a use case, gather specific data, evaluate results and scale from there. The right technology partners, large and small, will also bring technical best practices and skills to help projects succeed.
In their initial phases, digital twins are a good way to enable connected places to increase the return on investment for their existing data, and break down silos within the organisation.
To really get the most from the technology, though, they require data from an ecosystem of providers – increasingly this is a mix of internal departments, public sector organisations, the private sector and, with adequate permissions and privacy in place, citizens themselves.
This is seeing organisations at the early stages of exploring various aspects of the ‘data economy’, starting with driving awareness of the value of sharing data.
“With Schindler and Churchill at the moment, we’re exchanging data for the benefit of both parties and also the benefit of HS1, our client,” says OpenSpace’s Le Glatin. “Where we all are as an industry is understanding how to monetise datasets, and how each dataset contributes to unlocking value [for others]. That, I think, is going to take some time for everybody.”
Projects such as CReDo also highlight the growing need for a ‘shared language’.
To be able to create models of complex scenarios that involve multiple systems, Loach says we can learn from best practices in information management, such as using shared ontologies and striving for semantic precision, as well as principles from software engineering relating to modularity, interoperability and automated testing. Methods from data science around statistical validation and reproducibility of results will also help.
“Combining a multidisciplinary approach to technical development plus putting subject matter experts at the centre of use case creation, and user-centred design at the heart of the front end for digital twins will make them a success,” Loach comments.
The good news is that work to create these building blocks is underway.