With the completion of Dudley Hall and the new manufacturing technology laboratories, the School is advancing the development of the digital architecture needed to support its manufacturing curricula. As part of this effort, our team has begun the development of a Digital Twin intended to mirror the laboratories, their processes, and the work performed within them.
To date, we have established a working definition of the Digital Twin for this project by identifying the principal specifications, behaviors, and system components that must be represented. We have also developed an initial model-based framework by decomposing Station 3030 into its functional subsystems and mapping those physical elements to corresponding digital representations, thereby creating a foundation for future expansion and integration.
In support of this objective, we have employed commercially available software tools, primarily Visual Components, to develop a high-fidelity digital representation of Station 3030. This representation includes major system elements such as the robot arm, clamps, conveyor components, and sensor locations, with imported geometry positioned to reflect the physical manufacturing cell as accurately as possible.
We have further utilized open-source programming and scripting techniques, in conjunction with industrial communication tools, to connect behavioral definitions to the digital model. Through Python-based workflows and Kepware OPC communication, we have linked PLC data and simulation parameters so that the virtual environment can respond to real system signals.
Additionally, we have captured and evaluated run-time machine data by transmitting robot joint and operational values from the PLC into Visual Components. This effort has enabled us to verify tag mapping, confirm real-time communication between the physical system and the digital twin, and identify as well as correct motion interpretation issues encountered during testing.
At this stage of the project, we have demonstrated an initial real-time update loop between the physical station and the digital twin, confirming that the simulation is capable of reflecting live behavior from Station 3030.
The Smart Learning Factory at Purdue needs a locally hosted digital twin for Station 3030 that can accurately reproduce the physical system and update in real time using live operational data. Without this, students and researchers have limited access to a realistic platform for studying smart manufacturing, process monitoring, and digital factory behavior. This project addresses that gap by creating a high-fidelity digital twin that improves accessibility, supports real-time system interaction, and replaces the earlier cloud-based AWS solution with a faster and more practical local implementation.