Designing new aerospace engines is one of the longest-term design-and-manufacturing endeavors in industry. Setting off on this multi-year endeavor that will cost millions of dollars is fraught with challenges, and when Rolls-Royce Germany was looking to reduce costs and shorten lead times, it turned to data analytics and AI to come up with answers.
Rolls-Royce Germany embraced AI and data analytics 20 years ago, but it recently supercharged its AI activities: It partnered with Altair to turn the huge amount of data it had collected into actionable, at-your-fingertips information.
Altair and Rolls-Royce late last year signed a pact to address a range of applications. To begin, they will use the terabytes of real-time and historical engineering testing data that has been locked away in silos to address structural analysis and testing—to reduce engine weight and mass while maintaining structural integrity.
Rolls-Royce Deutschland’s Peter Wehle said in prepared remarks that the pact would “enable us to bridge the gap between engineering and data science, and empower our engineers to truly be engineers, focused on extracting the benefits of machine learning (ML) and AI from our data.”
Rolls-Royce will integrate Knowledge Works, Altair’s low-code data transformation and ML modeling platform, Altair CTO Sam Mahalingam said. The drag-and-drop software offers designers freedom and flexibility—especially at the start of the design process when the data can have a real impact on future production and testing stages.
“The software lets the design engineers think creatively while not having to worry about developing reams of code to try out new ideas and concepts,” he said. “The design engineers can try new things and get instant feedback.”
The Altair and Rolls-Royce Germany collaboration will address a wide variety of use cases, including applying data science to the vast amounts of engineering testing data, which can lead to a significantly reduced number of sensors needed. This use case alone has the potential to reduce recurring costs by millions of Euros.
Engineers using the software can connect to a data silo to choose the data they want to utilize, the algorithms they want to employ, or whether they want to use a neural network to train an ML model, he said.
The companies first partnered in 2017—on UltraFan, Rolls-Royce’s next-gen engine platform. Altair developed and employed simulation tools and methods with a focus on weight reduction, fuel efficiency and reduced noise. The new pact is “a natural extension” of that work.
“Altair has always been envisioning the future, how we can help our customers improve outcomes,” Mahalingam said. “We started on the product validation side and then we introduced to the market the concept of simulation-driven design.”
New simulation technology and expertise in high-performance computing followed. And now the company looks toward AI/ML as the next realm in design development. “We have shown many customers how to augment AI for their use cases,” he said, “and how we can provide additional AI models or ML training so they can do design exploration and shorten the whole design response time.
“There are several ways AI will enhance the end product, saving millions of dollars by shrinking design time,” he said. “Shaving 20 percent off the time is tremendous. But the goal is to shrink it by half.”
ALTAIR BY THE NUMBERS
EMPLOYEES: 3,000+ engineers, scientists and creative thinkers
LOCATIONS: HQ in Troy, Mich., with 86 offices in 25 countries
PRODUCTS: 150+ software and cloud solutions in product development, HPC and data analytics
SPECIALTIES: A pioneer in converging simulation, HPC, and AI, Altair enables organizations across broad industry segments to compete more effectively.
CUSTOMERS: 11,000 engineers and scientists, designers and developers use Altair software to solve the world’s toughest problems including: design and simulation targeting waste and weight reduction; increasing product strength, and analyzing massive data sets using HPC, machine learning and AI.
NUTS AND BOLTS: Altair established the simulation-driven design method, which has transformed the product development lifecycle by reducing time-to-market and enabling companies to spend more time inventing.