Case Study: Mazda improves vehicle frame design efficiency with Q-CTRL

Mazda improves frame design with Q-CTRL using 5X less training data

Mazda, a global automotive manufacturer, faced a costly bottleneck in its vehicle design process. The challenge was to optimize car frame shapes for improved fuel efficiency without compromising passenger safety. The traditional method, finite-element modeling, is extremely computationally expensive and limits design innovation. Mazda partnered with Q-CTRL to explore if quantum computing could accelerate these design timelines by minimizing the number of costly calculations required.

Q-CTRL designed a new Quantum-AI model to act as a fast surrogate for the expensive simulations. The solution leveraged quantum support vector machine (QSVM) models, which were executed on real quantum hardware using Q-CTRL's Fire Opal performance-management software. This approach achieved a 5X reduction in the amount of training data required to deliver frame designs with improved performance. The collaboration demonstrated that quantum machine learning can be meaningfully applied to real engineering workflows, providing an advantage in data-limited scenarios.

Mazda