Tag: Databooster

We would like to invite you to the next Smart Maintenance Expert Group meeting. In this meeting we will host Dr. Gabriel Michau from Stadler Services, who will tell us about

Full Steam Ahead: Unveiling the Future of Railway Rolling Stock Maintenance.

The maintenance regime of rolling stock creates by nature a competition between the reliability of the asset and the availability. Traditional maintenance regimes, mixing preventive and corrective maintenance strategies, look for a cost effective optimum between these metrics and years of optimisations have left little room for further improvements using the traditional optimisation strategies.

To achieve even more efficient maintenance strategies, dynamically scheduled maintenance, based on the real condition of the assets and of its components needs to be introduced, the so-called “Condition Based Maintenance” but this requires the development of three new types of capabilities: collecting and managing the right data, processing this data to assess the health of the monitored components and the ability to change and adapt the maintenance schemes seamlessly. Each aspect brings its own challenges, and many of these challenges are similar to the ones faced by the industry in general. In this presentation, we will draw upon exemplary projects carried out within Stadler Service’s fleet management to delve into the specific challenges encountered and the strategies employed to address them.

Dr. Gabriel Michau leads the New Maintenance Technologies team at Stadler Service AG, where he focuses on pioneering innovative maintenance strategies driven by data insights and cutting-edge technologies within the workshop. Prior to this role, he served as a Senior Scientist at ETH Zurich and at the ZHAW, specializing in the application of Deep Learning to Time Series for unsupervised fault detection and diagnostics in complex industrial systems, with a particular focus on very high frequency data (up to 100s of MHz). There, he led research initiatives aimed at advancing intelligent maintenance practices in collaboration with esteemed industrial partners, including SBB-CFF, Airbus, Stadler AG, GE, Oerlikon-Metco, Bystronic and more.

In 2016, he earned a dual PhD in Physics (Signal Processing over Networks) and Transport Engineering through a joint program between the Queensland University of Technology in Brisbane and ENS de Lyon in France. His research focused on inferring traffic flows in cities using Bluetooth and traffic counts data, and he developed an innovative traffic representation tool with graph theory and advanced convex optimization algorithms. His work earned him the first place in both the National and International Abertis Award in 2016.

The talk will be followed by an active discussion about successes and challenges in implementing and scaling condition-based maintenance in real operational systems.