SRAM transforms cycling with Autodesk on AWS, delivering components that are 20 percent lighter and 2x stronger

Bicycle component manufacturer SRAM was founded on the vision of creating a faster, more fun, and more efficient cycling experience for riders across the globe. Throughout its history, the company has taken advantage of new technologies to improve both its competitive position and the rider experience. Working with Autodesk and Amazon Web Services (AWS), SRAM embraced the concept of generative design, which uses the power of cloud computing to speed time to design and time to market while optimizing performance. By coupling generative design with additive manufacturing (3D printing), SRAM can produce parts that are twice as strong and 20 percent lighter in less time and with fewer resources.

SRAM, a manufacturer of innovative bicycle components for all types of bikes, has one focus: to create the best cycling experience for all. “Our goal is to create the next generation of products that our customers, end users, and competitors never thought possible,” said Dhiraj Madura, Global Director of Industrial Design at SRAM.

To deliver pioneering products and features that make bicycles faster and safer and inspire riders to go farther, SRAM has embraced the use of new materials, advanced manufacturing techniques, and, more recently, the concept of generative design with the help of AWS Advanced Technology Partner Autodesk.

Generative design is a form of artificial intelligence that leverages the power of the cloud and machine learning while accelerating the entire design-to-make process. Designers or engineers input design goals into generative design software, along with parameters such as performance or spatial requirements, materials, manufacturing methods, and cost constraints. The software then explores all the possible permutations of a solution, quickly generating design alternatives and testing and learning from each iteration what works and what doesn’t.

Autodesk is a leader in generative design technology. Its proven Fusion 360 platform integrates computer-aided design (CAD), computer-aided manufacturing (CAM), computer-aided engineering (CAE), and printed circuit board (PCB) software into a single platform that includes generative design tools powered by AWS.

SRAM chose to focus its initial generative design efforts on the bicycle crank arm, a key structural component connecting the pedal to the bike, with the goal of reducing weight without sacrificing strength or performance. Before generative design, engineers and designers would devote significant time to the laborious process of massaging pixels back and forth on a single 3D CAD model in an attempt to hit all the product requirements. “Now, with Autodesk generative design and AWS cloud computing, we’re able to generate far more concepts at the beginning of the project, giving our design team multiple solutions to validate through prototyping and refinement,” said Madura. “Given our tight timelines, this simply wouldn’t be possible with conventional 3D modeling.”

The project began by analyzing the existing crank arm. SRAM’s engineering team ran a virtual simulation known as finite element analysis (FEA) to put the crank arm under static stress. Using FEA, the team was able to simulate the pedal under maximum load, in which a rider is standing on the pedal and landing off a jump. “This is where we were first able to see the benefits of our partnership with AWS,” said Sean Manzanares, Senior Manager, Industry Strategy at Autodesk. “These FEA studies were solved in minutes versus hours when compared to a desktop computer with limited computing cores.”

With this information in hand, the team had a solid baseline understanding of the crank arm and was ready to move to design exploration using the generative design process. By inputting design parameters into Fusion 360, the team was able to use the supercomputing power of AWS to create literally hundreds of different design variations. “With AWS, we were able to solve close to 20,000 calculations. Each of the generative design studies used up to 170 processing cores in the cloud, and we had results within 30 minutes,” said Manzanares. “There’s no way a team of designers and engineers could come up with this volume of iterations from which to pick the best design otherwise.”

Each design variation was evaluated, and the team chose the most promising for further exploration. This involved virtually validating the new designs by running the same FEA static analysis that was originally run on the existing crank arm. The next step was to machine the validated designs, first by simulating the actual machining to avoid any potential issues in the process, and then by actually producing a part that could be installed on a bike for further testing.

While this produced a part optimized for weight and performance, SRAM’s ultimate goal was to push the boundaries of innovation by designing a crank arm that could be produced using 3D printing, a process known as additive manufacturing. Using the same generative design and testing process, SRAM produced a part that was twice as strong and 20 percent lighter than the original.

“Generative design is a tool that allows us to not only optimize performance parameters for the rider, but also shorten the development time to try new ideas, evaluate them, and throw them out or bring them to the prototype phase,” said Will King, Senior Design Engineer at SRAM. “We're an innovation company. As we look to inspire the consumer with our components, we want to continue to add innovation and make the riding experience better. Generative design is helping us get to that finish line quicker.”