PhD position in reinforcement learning for optimal fault mitigation and recovery actions in the operation of complex industrial assets

     
Employer
LocationZurich, Zurich region, Switzerland
Discipline
Activity
Posted

Your Mission

ETH Zurich is one of the world’s leading universities specialising in science and technology. It is renowned for its excellent education, its cutting-edge fundamental research and its efforts to put new knowledge and innovations directly into practice. The Chair of Intelligent Maintenance Systems at the Department of Civil, Environmental and Geomatic Engineering focuses on developing algorithms and decision support systems for data-driven intelligent maintenance of industrial assets.

PhD position in reinforcement learning for optimal fault mitigation and recovery actions in the operation of complex industrial assets

Predicting the remaining useful life of a system or a component enables taking preemptive mitigating actions, thereby reducing the downtime and improving the system reliability and availability. Yet, there is a set of different actions that can be taken if the remaining useful life of a system is known, including maintenance actions aiming to prolong the remaining useful life or preemptive replacements of the component. However, a preemptive replacement at the end of the remaining useful life may not be the optimal action in terms of system performance. A more efficient use of the knowledge on the fault evolution in time is to use this information for proactive control of the remaining useful life and the system performance. The project aims at developing reinforcement-learning approaches for optimal fault mitigation and system health management. A collaboration with the Prognostics Center of Excellence (PCoE) at NASA Ames Research Center is foreseen for this project.

We are looking for a highly motivated candidate holding a Master’s Degree in engineering, physics, applied mathematics, control, computer science or related fields with experience in predictive maintenance, machine and deep learning and particularly in reinforcement learning. The successful candidate has strong analytical skills, is proactive, self-driven with strong problem solving abilities and out-of-the-box thinking. Moreover, programming experience, preferably in Python, is expected. Professional command of English (both written and spoken) is mandatory and knowledge in German is beneficial. You enjoy working in an interactive international environment with other doctoral students and post-docs, referring continuously to practical problems and solutions and collaborating with industrial project partners.
Review of applications will continue until the position is filled, with the position to start as soon as possible and the planned project duration being three years.

Contact
 
In your application, please refer to science-jobs.ch
and reference  JobID 43292.