Product Digital Twins

We partnered with the technical team of a client in the energy space to establish a repeatable process for optimizing the design of custom turbines. By building and implementing a deep, learning-based surrogate model of the turbine, we enabled the team to test one million additional design iterations in 75 percent of the original time before selecting the most effective version.

A major automotive OEM was observing an increase in portfolio complexity due to a growing number of derivatives and variants. We helped them design and implement the concept of the digital twin, with a virtual replica of each product across the entire lifecycle. Core to this approach was a streamlined data layer, which increased the visibility across the value chain while also reducing the number of interfaces to internal systems. The result was a decline in portfolio complexity, the possibility to flash new functionality onto the cars in the field, and a decrease in data provisioning and usage costs. These changes translated to two percent EBITDA-cost improvements and a safeguarding and enablement of ten percent of revenue.

We helped a major aerospace company define a strategy and execution plan to build a multisystem digital twin that could simulate dynamic testing in various conditions. This resulted in a 12-month reduction in product verification and validation time along with a 15-25 percent decrease in overall time to market.