How We at Siemens Energy Enable Engineers to do Data Science

Astrid Walle, Tools and Data Analysis, Siemens Energy

A​bstract

In this talk, we will share our innovative approach at Siemens Energy that empowers engineers to tap into the power of data science within the context of classical engineering, with a focus on Computer-Aided Engineering (CAE) and simulation. One of the key challenges in leveraging data science for engineering applications lies in the effective management of simulation data, which necessitates proper tagging, adding meta-data, and ensuring data quality. To address this, we have developed a comprehensive data science platform that facilitates seamless data management and allows users to perform a wide range of data science tasks. Our platform enables engineers to upload, tag, and explore data, as well as carry out advanced data science tasks such as visualization and machine learning. It accommodates users with varying coding expertise, offering both code-based and no-code solutions. This empowers engineers to create their own surrogate models, including auto-tuning and automated selection of the best model. Furthermore, they can develop data processing pipelines that can be shared alongside the uploaded data, which is sustainably stored and can be combined into larger datasets. Our cloud-native solution is provider-agnostic and ensures secure access control through managed user groups. By streamlining data management and providing an accessible platform for data science tasks, we at Siemens Energy are enabling engineers to unlock the full potential of data-driven insights in classical engineering fields.