Applying Machine Learning in Vehicle Manufacturing

Vehicles are very complex systems, with a huge intersection of mechanical and electrical components and built in software control functions. As a result, there is a huge amount of data being generated from the systems within a car, much more than can be handled by most diagnostic tools. To continue ensuring a high level of quality and operational efficiency, manufacturers are turning to AI. However, there are numerous challenges that make application of approaches that work well in research difficult in a real-life manufacturing environment.

The discussion will focus on some prominent challenges with data quality, dynamic and continuously changing data streams, and scalable deployment of models.

Jean-Christophe Petkovich is the CTO and cofounder of Acerta Analytics Solutions. He is a PhD candidate in computer engineering at the real-time embedded software group at the University of Waterloo and has over 10 years of experience in data analysis, pattern recognition, statistical modelling, and machine learning. Jean-Christophe’s extensive work history includes writing software for safetycritical systems for leaders in the transportation industry, including QNX and Bombardier.

• Host: Sunny Wang, MS2Discovery Institute
• Date and Time: Nov. 29, 2019 at 3 p.m.
• Location: Lazardis Hall, LH2066