Image source Daria/stock.adobe.com.
|
This is the final article of our six-part special feature covering data management, digital twins, reality capture, remote workflows & collaboration, AI, IoT, and other technology for AEC designers, engineers, and owners focusing on how Lenovo and NVIDIA offer solutions for every phase of the design, construction and operations process. This issue focuses on using digital twins to improve project outcomes. |
Construction projects are inherently complex, with risks ranging from design and safety issues to cost overruns and schedule delays. As the industry adopts more technology-based workflows, digital twins are emerging as essential for predicting and mitigating risks before they escalate.
Digital twins — virtual representations of physical products, processes, and facilities — have enabled AEC teams to build real-life digital models of designs and update the models throughout project lifecycles. During planning and design, digital twins help AEC professionals develop cohesive designs using a single source of truth. During construction, digital twins aid scheduling and job site planning. Post-construction, digital twins guide owners in operations and maintenance (O&M) activities, providing data-driven, proactive O&M approaches.
Benefits of digital twins can be realized early on in project lifecycles, starting with data capture and planning stages.
By incorporating geospatial, infrastructure, and other site data into a unified digital model, project stakeholders can more thoroughly understand the project environment and explore multiple options to find context-sensitive solutions. NVIDIA Omniverse™ provides a robust platform for aggregating project data and enabling developers to build interoperable applications to orchestrate multiple technologies.
During design, digital twins enable project participants to develop designs collaboratively and validate project requirements while designing. Using Omniverse-based simulation techniques, design teams can optimize designs, accelerate design processes, and reduce errors. For example, on a building project, architectural and structural design can be coordinated with mechanical, electrical, and plumbing design, using clash detection and visualization tools to avoid physical conflicts. Key design parameters such as square footage, energy, and lighting can be monitored as designs are developed, avoiding surprises in late stages of the project.
Designers can also manage risks by performing realistic simulations using Omniverse and high-performance Lenovo workstations. NVIDIA Cosmos™, a platform of world foundation models (WFMs) and curation tools, can incorporate artificial intelligence (AI) trained on real-world geospatial and environmental datasets to amplify simulations to various environmental conditions. “By layering AEC project data over AI-generated WFMs, teams can perform much more accurate and early-stage predictive risk analysis of various scenarios such as flooding, load failures, and compliance, even before breaking ground,” said Jon Clark, Solutions Architect for AEC/Product Design and Development at Lenovo.
During construction, digital twins can simulate construction processes for accurate planning and monitoring of schedules, budgets, safety, efficiency, and worker training. Building on information developed earlier in project lifecycles, digital twins help construction contractors and other stakeholders access a single source of truth to help avoid miscommunication and reduce errors. Four-dimensional construction simulations (three geometric dimensions plus time) also called phasing, can help visualize progress and predict issues such as construction staging conflicts.
Digital twins powered by agentic AI can monitor construction progress, helping owners and contractors verify that construction complies with contract documents. The digital twin may be used in conjunction with AI-guided vision systems and robotics — to perform tasks such as weld inspection.
Benefits of digital twins continue after construction is complete. Incorporating valuable data such as equipment specifications, photographic imagery, IoT sensor data, live camera feeds, and O&M records, digital twins help project owners and stakeholders better manage facilities. By accessing this information, owners can improve operational efficiency, simulate and plan facility improvements, and provide safety and security solutions — predicting issues with AI before they occur. An operational digital twin can also be used to customize AI models and simulate degradation of assets such as soil erosion, steel corrosion, and normal wear and tear.
Digital twins help project owners and other stakeholders simulate operations such as thermal airflow. Image source: NVIDIA.
With a wide variety of information and project participants involved, AEC teams face various challenges in implementing digital twins, AI, and related technology. “The biggest challenge is just getting started with digital twins and AI. This requires people with the right skillsets and the right tools,” said Sean Young, Director of AECO, Geospatial, and AI Solutions Industry Marketing at NVIDIA.
From a hardware standpoint, digital twins and related technologies require significant processing power and GPU capabilities, according to Clark. Lenovo and NVIDIA have developed configurations specifically for AEC applications, as described in earlier segments of this article series.
Data management is also a challenge, with project data often fragmented across various technology platforms. To fully leverage digital twin data and perform real-time risk analysis, teams need to integrate and standardize data from a variety of sources.
To address the various challenges and leverage digital twin technology, AEC teams need to continually assess resources for performance capabilities, as well as other issues such as compatibility and security. “Many firms use hardware today that cannot support the demands of digital twins, AI simulation, and real-time rendering,” noted Clark.
For compatibility, Lenovo workstations are ISV-certified for AEC software such as Autodesk Revit, Navisworks, and Bentley products, aiding integration of various CAD and BIM technologies that feed into digital twins. Lenovo P-Series ThinkStation and ThinkPad mobile workstations, powered by NVIDIA RTX™ GPUs, support demanding digital twin workflows.
Omniverse provides the common platform via the Universal Scene Descriptor (OpenUSD) format applicable to multiple technologies. Cosmos WFMs enable realistic world state generation as videos with physical details, to guide robotic systems and autonomous vehicles.
Lenovo workstations running Omniverse and Cosmos can run securely on-premises or in hybrid-cloud environments. Lenovo also provides a comprehensive range of cyber-security services, ranging from security assessments through managed detection and response and cyber resiliency. NVIDIA helps organizations deliver robust cybersecurity solutions with AI and accelerated computing, enhancing threat detection, boosting security operational efficiency, and protecting sensitive data and intellectual property.
Lenovo P-Series ThinkStation and ThinkPad workstations, both powered by NVIDIA RTX GPUs, fulfill the needs for digital twins. Image source: Lenovo.
AEC firms are continually pushing the envelope of BIM and digital twin technology to improve project outcomes. Bouygues Construction, an international construction firm with more than 35,600 employees, has leveraged technology to improve efficiency and workflows on building and infrastructure projects. Building on established BIM workflows, the firm is exploring how Omniverse and AI can push the boundaries of multi-discipline coordination, visualization, work phasing, safety, and quantity extraction.
“Technologies are evolving rapidly, and Bouygues Construction aims to leverage them to address the challenges of the construction industry,” said Jérôme Loywick, R&D-Innovation Project Manager at Bouygues Construction. “With Omniverse, we can create virtual construction sites that allow us to find better solutions and improve safety for everyone.” Loywick shared insight on the firm’s use of technology at NVIDIA’s GTC Conference in March 2025.
NVIDIA Omniverse enables integration of data from multiple sources into a cohesive environment. Image source: Bouygues Construction. Click image to enlarge.
In pursuit of safer work sites, Bouygues Construction has built 3D scenes of construction projects by integrating assets from various software tools into a cohesive environment. Scenes can include BIM, IoT site data, computer vision, GIS, and other data. Project participants can then use AI modeling to simulate potentially hazardous work conditions, such as mechanical lifting and work-at-height operations.
With limited data available to model specific work configurations, the firm has used Omniverse Replicator to develop synthetic data for simulating those configurations. The approach generates physically accurate 3D data for training AI models and simulating various work scenarios. “We can simulate dangerous situations in a virtual construction site and combine this with real-world data to train AI models,” said Loywick.
The use of synthetic data to simulate work zones represents a promising opportunity to better identify and manage major risks on construction sites, according to Loywick. In addition to the simulations, Omniverse has aided visualization of site equipment and creation of immersive experiences for project participants to test project operations virtually before conducting them physically.
NVIDIA Omniverse enables integration of data from multiple sources into a cohesive environment. Image source: Bouygues Construction. Click image to enlarge.
With risk management a key concern throughout the AEC industry, digital twins, along with AI and other technologies, offer industry professionals new and expanding opportunities to improve risk management. The past decade has produced myriad changes in AEC, and the next decade promises even more.
“AI will transform every aspect of AEC, from planning to operations,” said NVIDIA’s Young. “I’m most excited about
physics-informed designs and robotic construction. Physics-informed designs will augment human creativity with science, to augment the built environment with both form and function. Robotic construction will accelerate the construction process, improve quality, and reduce risk of human safety and errors.”
Equipped with the necessary technological resources, AEC professionals can harness the new technology in innovative ways, ultimately improving project outcomes.
↓↓↓Download the Whole Series ↓↓↓
***
This article was sponsored by Lenovo and NVIDIA.
Searching for more information about Architecture, Infrastructure, and Construction?
Click here!