SyncTwin Leverages OpenUSD to Transform Microsoft Files into Factory Digital Twins

SyncTwin GmbH, a software company specializing in optimizing production, intralogistics, and assembly, is on a mission to unlock industrial digital twins for small and medium-sized businesses (SMBs), according to the NVIDIA Technical Blog.

Innovative Use of OpenUSD and Omniverse

SyncTwin has developed an application on the NVIDIA Omniverse platform utilizing Universal Scene Description (OpenUSD) to enable customers to create digital twins of their facilities from non-3D data formats such as Microsoft PowerPoint, Excel, and PDF. Integrating generative AI, the app allows users to create and modify digital twins through simple text prompts.

The application also employs the NVIDIA cuOpt accelerated optimization engine, significantly enhancing the efficiency of factory and warehouse operations. This integration allows manufacturing SMBs to simulate and evaluate operational scenarios, optimizing processes and resources, resulting in a projected 30% savings in facilities planning and process efficiency.

Bringing Manufacturing Domains Together

Factories are composed of interconnected and interdependent domains. Traditionally, the lack of integration between these domains and data sources has led to resource wastage. OpenUSD enables the integration of data between these domains in a digital representation of a factory, facilitating a comprehensive view of factory operations.

Using OpenUSD, SyncTwin customers can combine visual data, production processes, product catalogs, production orders, scheduling, resources, and production environments. For example, floorplans from PowerPoint slides can be converted into digital twins using bitmapping, and warehouse inventory data from Excel spreadsheets can be accurately reflected in the digital twin environment.

Michael Wagner, co-founder and CEO of SyncTwin, stated, “OpenUSD is the common language to bring all these different factory domains into a digital twin. It can be instrumental in dismantling data silos and enhancing collaborative efficiency across different factory domains.”

Optimizing Intralogistics with cuOpt

SyncTwin uses NVIDIA cuOpt, a GPU-accelerated microservice for logistics and route optimization, to optimize intralogistics for its customers. This integration enables near-real-time dynamic decision-making, improving human-machine interactions for automated storage and retrieval systems. It also allows for dynamic routing of robots for efficient truck loading and unloading.

Supercharging Factory Planning with AI

SyncTwin is integrating generative AI to accelerate workflows and enable planning teams to interact with digital twins in innovative ways. AI-enabled digital twins support planners with operational insights and recommendations, helping them identify optimal material locations and potential improvements based on historical data.

AI facilitates complex analyses and informed decision-making regarding resourcing, machine placement, and routing. Additionally, AI can help create a knowledge base for training, troubleshooting, and decision-making as experienced workers retire.

Developing SyncTwin on Omniverse

SyncTwin has been delivering innovative solutions for factory planning for years, but the development of their app on the Omniverse platform has unlocked new potentials. By leveraging OpenUSD, SyncTwin enables various factory planning tools to contribute to a digital twin environment seamlessly, breaking down data silos.

The Omniverse platform allows customers to simulate numerous scenarios and test various possibilities, providing valuable insights and solutions directly within the ecosystem. SyncTwin’s approach of structuring digital twin environments with OpenUSD components like Layers, References, and VariantSets has proven crucial in their development process.

Michael Wagner highlighted the importance of understanding the OpenUSD approach, stating, “When you begin to organize your data pipelines and digital twin structure holistically, the possibilities are endless.”