Digital twins, those virtual models that mirror physical systems, have become indispensable tools for industries ranging from energy to urban planning. Their ability to integrate live data streams with realistic 3D models has enabled operators to visualize and optimize complex processes in real time, significantly reducing the risks and costs associated with experimenting in the physical world.
Yet until recently, manufacturing firms have largely relied on analytics dashboards and spreadsheets to interpret factory data, missing out on the immersive potential of modern 3D engines. That changed in May 2025, when SAS Institute announced a partnership with Epic Games to combine SAS’s AI-driven analytics with Unreal Engine’s photorealistic rendering and simulation capabilities.
In this article, I’ll explore how Unreal Engine and other video game engines are already reshaping industries beyond gaming, and dive into the SAS/Epic partnership’s technical underpinnings and potential impact.
Transforming Enterprise with Game Engines
Game engines like Unreal Engine have found a wealth of applications outside traditional gaming. The most powerful ones can create photorealistic graphics, dynamic lighting and reflections, fluid and physics simulations. That’s great for gamers looking for an exciting, realistic gaming experience, but it also translates naturally to any scenario that benefits from real-time, high-fidelity visualization.
There are smart-city projects leveraging video game engines to integrate traffic flows, infrastructure resilience, and flood modeling into interactive cityscapes where planners can test road closures or green-space expansions with high granularity.
In renewable energy engineering, prototype platforms built on game engines allow academic and industrial researchers to model and optimize autonomous drone inspections of PV arrays, combining physics-accurate flight dynamics with real-world environmental data.
The SAS and Epic Games Partnership
The stated aim of the SAS–Epic collaboration is to turn manufacturing data into immersive, actionable insights. Unveiled at SAS Innovate 2025, the partnership leverages Unreal Engine’s open, real-time 3D creation capabilities alongside SAS’s AI and analytics suite.
One of the first pilots took place at Georgia-Pacific’s Savannah River Mill, where operators used Epic’s RealityScan mobile app to capture photorealistic scans of plant machinery and infrastructure. These scans were then imported into Unreal Engine to recreate the facility in a virtual environment, complete with accurate geometry, textures, and lighting.
On the analytics side, SAS Viya ingested telemetry from sensors, programmable logic controllers (PLCs), and automated guided vehicles (AGVs), organizing the data into a unified model.
To bridge Unreal Engine and SAS Viya, SAS engineers developed a plugin that streams live telemetry into the 3D scene, allowing virtual machines to behave according to real-world sensor inputs. As a result, users can run “what-if” scenarios, such as testing new automated guided vehicle (AGV) routing strategies, without halting physical production lines..
Early results at Georgia-Pacific indicate potential efficiency gains: simulations of AGV movements helped identify routing bottlenecks and reduce idle time, while adding digital human avatars enabled safety analysts to assess rare near-miss events in context. Though detailed ROI figures remain under wraps, the reduced downtime and optimized workflows could yield double-digit returns on investment.
Implications for Manufacturing Operations
The reportedly successful pilot at Georgia-Pacific has promising implications for the manufacturing industry. Advanced digital twins should simplify monitoring and troubleshooting. Rather than toggling between dashboards, engineers can observe a single 3D model that updates live with sensor data, highlighting anomalies or equipment failures in context.
Central control rooms could display multiple virtual factory lines side by side, enabling remote teams to diagnose and resolve issues from anywhere in the world. You can also stress test new ways of optimizing workflows, cloud deployments, and more.
Digital twins are also extremely effective for operator training. Historical logs of machine faults and human-error incidents can be replayed in the digital twin, providing immersive, scenario-based training for technicians and operators.
‘What if?’ scenario testing should be fairly trivial: teams can experiment with new layout configurations, process adjustments, or emergency-response drills via digital twins. They could quantify the impacts on throughput, safety, and energy use, ensuring OSHA compliance before committing to any changes to physical infrastructure.
Unlocking Future Advancements
Video game engines such as Unity, Amazon Lumberyard (via AWS IoT TwinMaker), and Unreal are increasingly adopted to build digital twins for manufacturing, offering real-time visualization and simulation of complex processes. Looking forward, there’s likely to be a deeper integration of machine learning and, potentially, quantum computing into these platforms to enable hybrid quantum-AI simulations that tackle highly complex supply-chain and process-optimization problems at unprecedented scales.
There’s likely to be an emergence of two-way communication between digital and physical systems, where control commands generated in the virtual twin feed directly back to shop-floor devices to automate fine-tuning and error corrections without human intervention.
Beyond manufacturing, the same video game engines can support digital twins in healthcare, modeling personal treatment, patient flows, and simulating equipment maintenance. Smart-city planning is already benefiting, where video-game-engine-based frameworks can visualize traffic patterns and utility networks across sprawling urban environments.
As these digital twins grow more autonomous, embedding ethical-AI governance frameworks will become essential, ensuring that predictive models operate transparently, respect data privacy, and adhere to regulatory standards throughout their lifecycle.
Challenges remain. There’s standardizing data protocols across different engines and upskilling personnel to navigate game-engine interfaces, to name a few. Ongoing collaboration between traditional industrial vendors and gaming-engine developers hints at a future where digital twins are not only photorealistic but also self-optimizing and deeply integrated into everyday operations.
The SAS Institute and Epic Games partnership could be a strong indicator of where manufacturing digital twins are going, blending gaming-grade visuals with enterprise-scale analytics to create truly immersive, high-fidelity virtual replicas of factory operations. Early results indicate gains in optimization, safety, efficiency, and decision-making.
Game-engine technology could enable manufacturers to not only make advanced analytics more accessible across their organizations but also transform traditionally sterile processes into engaging experiences that drive innovation.
It’s an interesting convergence of IT and OT, and a fascinating application of engines originally designed for gaming purposes. When combined with AI and the future potential of machine learning and quantum advances, we could see digital twins that do more than mirror reality. They will predict, adapt, and even control physical systems, which sounds pretty epic to me.