The AMT Technology Forum returned in 2021 as a virtual conference on IMTS spark. The forum hosted 12 presentations on peer-reviewed research with an emphasis on applied research or implementation of new technologies, including artificial intelligence, standards, additive manufacturing, blockchain, and more.
In his presentation “Manufacturing Digital Twins: Understanding Digital Twins and How Standards Can Enable Them,” Jan de Nijs, LM Fellow for Enterprise Digital Production, Lockheed Martin, started by sharing the formal definition of the digital twin according to the ISO 23247 standard: “A manufacturing digital twin is a fit-for-purpose digital representation of an observable manufacturing element with a means to enable convergence between the element and its digital representation at an appropriate rate of synchronization.”
He went on to explain why he believes it is important to limit the scope of the digital twin (DT) to the manufacturing process and discussed how asking oneself what it is that you want to accomplish is key to the process of defining the DT. He next reviewed several different use cases to illustrate different possible purposes of a DT for rotating tools, the first being real-time location tracking and the second being tool life management. These would have different digital descriptions and different data artifacts.
To further illustrate his points through representative use cases, de Nijs discussed a use case for a fracture critical machined part where the objective is to be able to trace back how the part came into being and what it looks like. In this case, the minimum required artifacts would include the original, as-designed engineering requirements; the original engineering artifacts related to those requirements, such as FEA, simulations, FMEAs, DfX; and the original, as-planned manufacturing and inspection plan. It would also include the artifacts specific to the serial number, such as the material certifications, the as-built geometry (i.e., the as-built QIF results file), the as-processed MTConnect data files, the complete non-destructive inspection artifacts, and any other data elements specific to the serial number. De Nijs contrasted this to the DT for a standard bolt or nut if the purpose was to be able to add to the design. In this case, the minimum required artifacts would only be the original, as-designed engineering requirements.
He then discussed a use case of a DT for a manufacturing process assuming its purpose was being able to prove many years later that the process was executed according to a given plan as well as to be able to visualize the way the part was created using an augmented reality application. The minimum required artifacts in this case would include the original, as-planned manufacturing and inspection plan (i.e., ISO 10303 AP-238 and QIF-inspection plan) and the machinery, fixture, tool, and part CAD data. The serial number-specific artifacts would be the as-processed MTConnect data files for both manufacturing and inspection, and the proof of manual operations, such as a video clip or other data artifacts.
De Nijs concluded the presentation by reminding us that DT-related data artifacts are a deliverable for both internally as well as externally manufactured parts, and they require the same amount and format of data regardless of where the part is made. Because DTs can be customer-driven, the industry needs to embrace standards including ISO 10303 AP242, ISO 10303 AP238, MTConnect, QIF, and the IPC series of standards. This will also enable the long term archival and retrieval of the data indefinitely into the future.
