Primer: Digital Twins

A deep dive for construction industry leaders.

Feb 03, 2025

This is part of a series of ‘primer’ articles which aim to provide accessible guidance to leaders in the construction industry. Each primer will arm you with the foundational information to understand key trending technologies impacting the construction industry.

These primers are suitable as reference materials when the subject matter arises as part of your role - perhaps in preparing for a meeting. Consider bookmarking them to come back to when needed.

Disclosure: I used Google's Gemini Deep Research as a research tool for this article, assisting me in finding case studies and confirming my understanding of the topic.  While Gemini contributed to the research process, the views and conclusions presented here are solely my own.

blue yellow and red abstract art — Photo by taopaodao on Unsplash

You’ve no doubt heard about Digital Twins. And, like many, you’re confused about what they are. If you’re not then you could be mistaken about your level of understanding!

The term “digital twin” is widely used across the construction industry, in all sorts of contexts. The problem is, when the term “digital twin” is used, there is rarely a common understanding of what is meant. In other words, ‘Digital Twin’ is a buzz term.

The proliferation of the term degrades the power of the technology. A Digital Twin is a powerful asset in its own right.

If you’re a leader in the construction industry and aren’t clear about digital twins read on. In this article we will look at:

Foundational Understanding: What a digital twin is, case studies from across other industries, types of twins, some example software solutions.
Actionable Insights: What are digital twins in construction, implementing a digital twin solution, risk/challenges/opportunities, complementary technologies, future outlook.
Wrap Up: Skip to here if you’re in a real hurry! Key takeaways and recommendations, further reading and sources

Share this post with your colleagues. You’ll be helping them (and also impress them with your tech focus!)

Overview of Digital Twins

Digital Twins exist to enable decision making about ‘things’ by providing data insights from system or condition monitoring, and data analysis.

Put simply, a digital twin is a digital representation of a real asset or process with a two-way relationship between the real and the digital. That relationship part is critical. The twin receives data about the real things, and is able to affect the ‘real’ twin in some way.

To achieve this digital twins use sensors, Internet of Things (IoT), and other methods to monitor and measure at a frequency determined by the end-user need.

Various organisations have had a stab at defining digital twins, here are some examples:

The main misconception are between a digital models, design models, or simulations, and digital twins. A digital twin is a virtual environment, using a precise replica of the entity to monitor conditions and test actions in real time, with real life data, with the aim of reducing cost, error, or harm to the real ‘thing’.

A marriage of assets, sensors, and graphical models

In order to create the bidirectional flow of information between the real and digital twin sensors are used. Sensors come in all shapes and sizes such as temperature, carbon dioxides, fire, smoke, noise, etc… They are used to sense the conditions in the physical twin so that actual conditions can be assessed by the digital twin. The digital twin processes the information from the physical twin and provides feedback to the owner and/or tests and takes actions on the real ‘thing’. Some examples may help:

An occupancy sensor detects that occupation limits are being reached in an event centre. A graphical indicator is shown to the control room, and the sign at the entrance shows an amber alert to security and patrons.
A noise sensor in a fan unit detects a breach of sound pitch limits. The graphical representation shows the fan as amber on the schematic and a work instruction for maintenance is sent through the Computer Aided Facilities Management (CAFM) system.

Types of Digital Twin

A Digital Twin isn’t a single product, more a conglomeration of solutions aimed at solving specific problems or exploiting specific opportunities. They can come in a variety of forms, provided broad categories below: (Source: What Is a Digital Twin? | IBM)

Product/Component/Part Twin: The basic unit of a digital twin, the smallest example of a functioning component.
Asset Twin: When two or more components work together, they form an asset. Asset twins let you study the interaction of those components, creating a wealth of performance data that can be processed and then turned into actionable insights.
System or Unit Twin: enables you to see how different assets come together to form an entire functioning system. System twins provide visibility regarding the interaction of assets and may suggest performance enhancements.
Process Twin: Reveal how systems work together to create an entire production facility. Are those systems all synchronised to operate at peak efficiency, or will delays in one system affect others? Process twins can help determine the precise timing schemes that ultimately influence overall effectiveness.

Case Studies: How are digital twins used across industries?

Manufacturing

Siemens: Pioneering the Digital Factory
- Production process replica: By creating a virtual replica of the entire production process, in its Amberg Electronic Plant, Siemens can simulate and analyse different scenarios, optimise production lines, and predict potential issues before they occur.
- Equipment Lifespan: Siemens uses digital twins to predict and prevent equipment failures in its manufacturing plants and other operations. By simulating and analysing the behaviour of equipment, Siemens can identify potential issues and optimise maintenance schedules to extend the lifespan of critical equipment.
- Virtual prototypes: Siemens also use digital twins for product design and development. Using virtual prototypes engineers test and refine designs in the virtual environment. This reduces the need for physical prototypes and accelerates time-to-market. This also leads to greater innovation and increases the speed of iteration cycles, resulting in better products.
- Sustainability: By modelling and predicting energy consumption, Siemens can make data-informed decisions that align with their sustainability goals.
GE: Optimised Assed Performance and Sustainability
- Virtual asset replicas: GE monitors the health in real-time of their assets using digital twins. This helps to minimise downtime, predict failures, and optimise maintenance schedules.
- Predicting Jet Engine Failures: GE uses digital twins to monitor the performance of its jet engines, collecting data on various parameters such as temperature, pressure, and vibration. By analysing this data in real-time, GE can identify potential issues before they lead to failures, allowing for proactive maintenance and preventing costly downtime.
Unilever: Production Efficiency and Agility
- Production Processes: Unilever, a multinational consumer goods company, uses digital twins technology in its factories. By creating digital twins of their production processes, they can simulate different scenarios, identify bottlenecks, and optimise production schedules. This has led to significant reduction in production interruptions, improved quality, and increased agility in responding to changing market demands.
  - For example, Detergent Powder Facility, Brazil: the digital twins helped reduce the number of alarms requiring daily attention by up to 90%, minimising process interruptions and production time. This allowed employees to focus on value-adding tasks.

Automotive

Tesla: Revolutionising Vehicle Design and Performance
- Twin Car: Tesla's innovative approach to digital twins sets a new standard in the automotive industry. Every Tesla vehicle has its own digital twin, which collects and processes real-time data from the car, including performance metrics, sensor readings, and operational conditions. This data is then used to continuously improve vehicle performance, predict maintenance needs, and offer over-the-air software updates. This approach allows Tesla to continuously monitor the performance of its vehicles in real-world conditions, identify potential issues, and proactively address them through software updates or maintenance recommendations. This not only improves vehicle performance and reliability but also enhances the customer experience.
- Simulate and Test: Tesla also uses digital twins to simulate and test different designs before manufacturing physical prototypes, significantly reducing development time and costs. This allows Tesla to accelerate innovation and bring new models to market faster.
Ford: Supply Chain Management
- Logistics efficiency: By simulating scenarios, Ford can identify potential bottlenecks and disruptions, ensuring the timely delivery of parts and materials to its production sites. This helps Ford maintain production efficiency and minimise delays.
BMW: Enhanced Factory Planning
- Factory twin: By creating 3D scans of its production sites, BMW can create virtual replicas of its factories, enabling virtual walkthroughs and simulations. This allows employees to collaborate in real-time, optimise workflows, and identify potential issues before they occur. For example, BMW employees can use virtual reality glasses to immerse themselves in the virtual workplace, capturing and improving processes within a digital environment. This allows for more efficient planning and optimisation of factory layouts and workflows.
- Production processes: BMW's digital twin strategy has delivered significant improvements in efficiency and productivity. By simulating and optimising production processes, BMW has reduced planning effort, minimised capital expenditure, and ensured more efficient and stable vehicle launches.

Aerospace

Airbus: Optimised Aircraft Performance and Sustainability
- Aircraft Performance: By creating virtual replicas of aircraft, Airbus can simulate every aspect of their performance, from aerodynamics to system operations. This comprehensive approach has reduced rework, improved overall aircraft performance, and enhanced safety.
  - For example, Airbus uses digital twins to assess aerodynamics, structural integrity, and system interactions in its aircraft designs. This allows Airbus to optimise designs and identify potential issues before physical prototypes are built.
- Operational efficiency: Airbus also leverages digital twins to optimise aircraft operational efficiency and sustainability. By analysing data from digital twins, Airbus can optimise fuel consumption, reduce emissions, and improve maintenance practices.

Energy

Optimised Grid Operations
- Digital twins are used to model and simulate the behaviour of power grids. By analysing data from these models, operators can forecast potential blackouts, optimise load distribution, and quickly respond to faults. This proactive approach ensures a more resilient and reliable power grid. This approach allows grid operators to improve grid stability, prevent blackouts, and optimise energy distribution.

Digital Twin Software Solutions

Here are some of the current providers of digital twin software, their target use case(s), and key industries. This is not an exhaustive list:

Actionable Insights

Digital Twins in Construction

Digital twin technology is as relevant to construction as it is to manufacturing, automotive, aerospace, or any other engineering based industry.

Like the case studies shared earlier, the build environment must be managed and maintained, we must perform interventions to improve, repurpose, or remove assets, and there are processes to design and produce products, assets, and systems. As construction slowly and begrudgingly moves into the Information Age demand is growing to utilise the data we generate to power continuous improvement.

The first leap into digital twin technology within construction has been asset owners. Seizing the opportunity for efficiency and optimisation they have turned to digital twins to help them understand their assets and estates.

Simulation and digital models are becoming common place in the design and construction processes. However, there is huge potential in this area to realise the benefit of the technology for end-product optimisation, logistics, and supply chain management.

Use cases in construction are much the same as discussed earlier in other industries, for example:

Understand the performance of parts/assets/systems/processes
Maintenance and operations optimisation and efficiency
Design optimisation
Planning optimisation

Implementing a Digital Twin Solution

Let’s start by being clear.

The objective should not be to get a digital twin, the objective should be to solve a specific problem or exploit a specific opportunity.

What you end up with may be a digital twin.

The below is an industry and domain agnostic process for addressing issues or opportunities:

Review the as-is: “What’s going on here?” is an excellent question to ask when faced with an opportunity or problem. What is actually happening, what needs to change? This requires in-depth analysis to ensure you get to an accurate problem statement which fully addresses root causes. Failure to implement this step will hamper end-results, potentially with severe time and cost penalty. You could go in the wrong direction.
Define the to-be: “What do we want to be happening?” Now you know what is happening you need to look at what the target end-state should be. Without thinking too much about optinioneering at this stage, given the business context, decide what the to-be should look like to address the underlying problem or opportunity. This step ensures that you and your partners/stakeholders are agreed on what good looks like.
Design and Specify: “What are the specific requirements?” Now you need to choose an option and design the to-be in detail. This includes what it is, how it will be achieved, when it will be achieved, who will do it, and how success will be measured. At this point you also specify the standards and requirements. By the end of this step you should have enough detail to understand the cost. Without fully defining the essential parameters you will end up with a ‘fuzzy’ delivery. This will cause delay, increase costs, and result in failure to meet the desired state.
Budget: “What am I willing to spend on this (time/money/resource)?” So many initiatives fail to deliver the value intended because the budget isn’t robust enough to actually deliver the envisioned result. We end up with a compromise. The budget should be based on a clear set of requirements and milestones. If the budget becomes too large then the previous steps 2 & 3 must be repeated until either the budget fits or the initiative is completely rethought or scrapped.
Delivery: “Is the development/deployment going to plan?” Besides actually delivering the solution the biggest risk during delivery is change. Often, even with the clearest of plans, difficulties arise during delivery which challenge the design, specification, and/or budget. Therefore, like any project, you need to establish a rigorous change management process. As change occurs (and it will!) you need to assess the impact on the to-be state of the final product and on the acceptable budget. Accepting too lightly change which impacts the final solution could mean that you miss the mark in addressing the problem or opportunity you started with.
Verify and validate delivery: “Is this what I asked for?” Once you’re in delivery stage you also need to ensure that the requirements are being effectively delivered. Software specification, user interface needs, information exchange, etc… You must put in place a system of checking which holds the delivery of the project rigorously against the specification and requirements. In terms of information exchange in construction, this is where BIM and Information Management is foundational to the world of digital twins.
Ongoing measurement and improvement: “Is this still working for me?” Lastly, you must continually assess and improve the solution once in place. With the best will in the world you can not see into the future when designing and specifying at the outset. Things change over time. New data sources appear which may need integrating into the solution, software becomes outdated and needs upgrading, new security risks appear, and so forth. Change is continuous.

Challenges, Risks, and Opportunities

The table below looks at some of the general risks and opportunities for digital twin implementation. You’ll recognise some of these from the implementation steps above. Though the below tackle issues and resolutions in more detail.

These are not a thorough review, and are certainly not complete or exhaustive. The table below should be tailored to your specific actions and context of your company or project:

Complementary Technologies

Digital twins don't exist in isolation. They rely on and integrate with a range of other technologies to achieve their full potential. Here's a breakdown of key complementary technologies:

Reality Capture: Technologies like laser scanning (LiDAR), photogrammetry (using drones or cameras), and computer vision are essential for capturing the "as-is" state of physical assets and environments. This data forms the foundation for creating accurate and up-to-date digital twins, especially when sensor data is unavailable or insufficient. Example: Point cloud surveys for creating building models, drone inspections for infrastructure.
Internet of Things (IoT): IoT devices and sensors provide the real-time data streams that are crucial for dynamic digital twins. These sensors collect data on various parameters (temperature, pressure, location, performance, etc.), enabling the digital twin to reflect the current state of the physical entity and enabling predictive maintenance and real-time control. Example: Sensors monitoring equipment performance in a factory.
Artificial Intelligence (AI) and Machine Learning (ML): AI/ML algorithms process the vast amounts of data generated by digital twins, identifying patterns, anomalies, and insights that would be difficult for humans to detect. This enables predictive analytics, automated decision-making, and optimisation of processes. Example: AI predicting equipment failures based on sensor data, ML optimizing energy consumption in a building.
Building Information Modelling (BIM): For built assets (buildings, infrastructure), BIM is often a foundational technology. BIM provides a structured, data-rich representation of the asset's design and construction, which can be used as the basis for creating a digital twin. BIM standards like ISO 19650 and the UK BIM Framework are crucial for ensuring data quality and interoperability. Example: Using a BIM model to create a digital twin for facility management.
Product Lifecycle Management (PLM): In manufacturing, PLM systems manage all aspects of a product's lifecycle, from design and engineering to manufacturing and service. Integrating PLM data with digital twins enables manufacturers to track product performance in the field, optimise designs, and improve product development processes. Example: Using PLM data to create a digital twin of a product for virtual testing and simulation.
Computer-Aided Facilities Management (CAFM): CAFM systems help manage building operations and maintenance. Integrating CAFM data with digital twins enables real-time monitoring of building systems (HVAC, lighting, security), automated maintenance scheduling, and optimised space utilisation. Example: Using a digital twin to optimise energy consumption in a building managed by a CAFM system.
Simulation Software: Simulation tools are used to create virtual models of physical systems and processes. These models can be integrated with digital twins to simulate different scenarios, predict future behaviour, and optimize performance. Example: Using computational fluid dynamics (CFD) to simulate airflow in a building's HVAC system.
Data Analytics Platforms: Digital twins generate vast amounts of data. Data analytics platforms are used to process, analyse, and visualize this data, providing insights into asset performance, operational efficiency, and other key metrics. Example: Using a data analytics platform to identify trends in equipment performance data.
Data Integration and Pipeline Tools: Connecting the various systems that feed data to a digital twin requires robust data integration and pipeline tools. These tools ensure that data is collected, transformed, and delivered to the digital twin in a timely and reliable manner. Example: Using ETL (Extract, Transform, Load) tools to integrate data from different sources.
Cloud Computing: Cloud platforms provide the infrastructure, scalability, and flexibility needed to host and manage digital twin solutions. Cloud-based digital twins can be accessed from anywhere, enabling collaboration and remote monitoring. Example: Hosting a digital twin platform on a cloud service like AWS or Azure.
Augmented Reality (AR) and Virtual Reality (VR): AR/VR technologies can be used to visualise and interact with digital twins in a more intuitive and immersive way. AR overlays digital information onto the real world, while VR creates a fully immersive virtual environment. Example: Using AR to visualise building information on a construction site, using VR to train operators on a virtual twin of a complex machine.

Future Outlook

The future of digital twins is bright in the construction industry. There is vast untapped potential in the space. This can be shown by the example in other industries and what we have so far implemented in ours. Technology continues to develop at a rapid pace, and as it does you can expect to see:

Increased adoption: More enterprises will adopt digital twins as the technology matures and becomes more accessible. This will be driven by the increasing availability of affordable sensors, cloud computing platforms, and accessible software tools.
Enhanced capabilities: Digital twins will become more sophisticated, incorporating advanced technologies like AI and machine learning - building an ‘intelligence layer’ - to provide even greater insights and value. This will enable more accurate predictions, more efficient optimisation, and more autonomous decision-making.
Wider applications: Digital twins will be applied to a wider range of use cases, including design development, supply chain management, business development, and sustainability initiatives. This will drive innovation, improve efficiency, and create new business opportunities across various sectors.
Share this post with colleagues to help increase their understanding. Plus it’ll make you look super smart…
Share

Key Takeaways & Recommendations

A digital twin is a tool, not a destination: ensuring that you define the problem or opportunity, design the end-state, and identify coherent actions are the critical pieces to ensure success. Starting with simply desiring a digital twin is a sure way to stunt potential value and benefits.

Enhanced Design and Planning: Digital twins enable the creation of highly detailed virtual models of buildings and infrastructure, facilitating better collaboration, design optimisation, planning, and coordination. This can lead to significant cost and time savings during the design and planning phases.

Improved Construction Processes: Digital twins can be used to simulate and optimise construction processes, identify potential bottlenecks, and improve resource allocation. This can lead to increased productivity, reduced construction time, and improved safety on construction sites.

Optimised Operations and Maintenance: Currently, the primary use case is to monitor the performance of buildings and infrastructure in real-time, predict maintenance needs, and optimise energy consumption. This can lead to reduced operating costs, improved building performance, and increased sustainability.

Challenges and Risk: Whilst the opportunities are attractive these must be balanced with the key risks and challenges. Digital twins can be expensive and difficult to implement. Therefore, planning and development should be done with care to ensure that design, specification, and delivery are managed to exploit well defined opportunities or mitigate well defined risks.

Start with Pilot Projects: Begin by implementing digital twins on smaller projects to gain experience and demonstrate value before scaling up to larger, more complex projects.

Focus on Data Quality: Ensure that the data used to create and maintain digital twins is accurate, reliable, and accessible. Invest in data management systems and processes to ensure data quality. Implementation of Building Information Modelling (BIM) on construction projects can help to create and manage data for digital twins, ensuring data quality.

And that’s it. I’d love to discuss this with you. Whether you’ve implemented, used, or are just curious about Digital Twins give me a shout on LinkedIn, X/Twitter, via message here at Substack or in the comments below:

BLDGcertainty

Resurrecting Notre Dame: A Digital Cathedral

Buyer beware: Procuring Information with Confidence

BIM Bankruptcy

Compelling Digital Investment

Discussion about this post