“… VW Group has about 10,000 engineers, but only a few hundred programmers. You need people…, who can work in cloud computing and who are proficient with artificial intelligence”  ​

Martin Hofmann
Chief Information Officer, VW Group

loreal and monolith ai

Use cases

Automotive Wind Tunnel Testing​

Customers use Monolith to create a Digital Twin of the product to be tested and can manipulate the Digital Twin’s properties while in the wind tunnel​. By importing data from the sensors, test rig, and airflow, the customer can accurately predict readings from sensors that have failed, the aerodynamic performance of the product, and the impact of air contamination on the test​. The resulting reduction in wind tunnel time saves tens of thousands of pounds per day.​
Learn more

Automotive crash testing

The Monolith platform enabled the customer’s engineering team to quickly import and process sensor data, then transform it and create machine learning models. ​Models created from a small initial number of tests and tests from previous vehicles enabled the engineers to predict the outcome of the regulatory tests. This enabled them to adjust the safety aspects of the design of new vehicles to satisfy regulatory tests much faster, with fewer resources and lower cost.​
Speak to us

Vehicle Dynamics

Using Monolith’s intelligent exploration tools, the data can be sorted and easily visualized within the platform to inform engineers. Machine Learning models can be trained on certain maneuvers, to predict the forces on the vehicle during maneuvers that were not performed, thereby reducing total testing time. This allows for a test campaign to be optimized, reducing the number of tests performed to characterize the vehicle behavior.
Learn more

Engine calibration

Our solution was to learn from previous tests and perform new virtual tests in the Monolith platform to quickly validate or invalidate options to reduce fuel consumption. In this use case, we looked at the effect of three design changes on the fuel consumption: 1) earlier combustion, 2) lower charging loss, and 3) use of an intercooler.

Learn more

Key Automotive industry challenges

1.

Wind tunnel tests are expensive, and almost every test is rushed towards the end of the shift. Self-learning models offer high-value insights into wind tunnel measurements, thereby improving test planning and execution. Test engineers can draw complex data dependencies between multiple design variables, without being constrained by the limits of traditional physics-based methods.

2.

Track tests are a fun day out, but preparation for track days can be a headache. Failing sensors, loss of data, noisy data, and poor weather conditions can all jeopardize a successful track test outcome. Self-learning models can significantly enhance the value of track tests. For example, a transient track test is often sufficient to produce results that are otherwise obtained from a steady-state. Test less, learn more.

3.

Crash tests are one of the most expensive destructive tests that every auto OEM must perform, as each requires building a new prototype car. Using self-learning models has enabled our customers to find a direct correlation between different crash scenarios and parameters linked to achieving a 5-star Euro NCAP rating.

Jota Sport cuts car setup time by 50% with Monolith

Since teaming up with Monolith AI, Jota engineers can better understand and predict the aerodynamics of their cars by building self-learning models.
As a result, they have reduced the number of simulations and tests by 50%, cut car time-to-setup in half, and achieved a 66% reduction in overall costs.