For many legacy platforms, there is a long history of paper-based designs and specifications or source code which can be spread among a myriad of suppliers. This makes any changes and improvements to those platforms onerous and time-consuming.
We are addressing this challenge through model-based systems engineering, sometimes called ‘Digital Twinning’.
“Model-Based Systems Engineering is really the next generation of how we will design, test and improve our products throughout their lifecycle. Providing a single point of digital truth has benefits across the supply chain.” Mandy Savage, Head of Engineering, Products & Training Services.
Taking performance data from the real world, we are digitising it to create a model of a platform or product in the virtual world. We can then use this ‘digital twin’ to experiment and validate performance using real-world data. The results of these tests can then be fed back into real systems to test the impact of changes, from designs to in-service support.
So far, we have trialled this approach using our proven Archerfish® mine-neutraliser. This is a complex product with actuators, cameras, sonar and a warhead, where changes to one aspect can have significant impacts elsewhere. For example, the effects of an increase in payload can now be tested and modelled on the digital twin to measure resultant implications on range and handling.
With increasing model fidelity and computing power, this approach enables the Royal Navy and our own engineers to consider new scenarios or different capability requirements, with reduced testing times and associated cost reductions. We also expect it to bring down the development time for the first variant and any subsequent enhancements.
Model-Based Systems Engineering is really the next generation of how we will design, test and improve our products throughout their lifecycle. Providing a single point of digital truth has benefits across the supply chain.” Mandy Savage, Head of Engineering, Products & Training Services.
Executable parts of the model can also be shared with different parties in the supply chain in a ‘black box style’ file, which protects intellectual property and maintains security. In the case of Archerfish, for example, new scenarios can be modelled by the battery provider without giving away sensitive data on the overall requirement or capability. As artificial intelligence and machine learning improves, so we may also see the model recommend improvements using data from performance in the real world and comparing it to that held in the digital environment.
We are currently in the second year of our pilot programme and throughout 2020 expect to demonstrate auto-code generated from the model.
