Are you fascinated by predictive maintenance approaches for smart industry? Are you eager to work on developing AI and data-driven models for maintenance decision support? Are you willing to dive into the data challenges in predictive maintenance? Are you fascinated by combining theory and real-world implementation?
ContextMaintenance service is crucial in ensuring minimal downtime, maximal productivity, and reliability of complex systems at a minimum ownership cost. Data-driven predictive maintenance (PdM) is a promising approach where system owners monitor the parameters of their machines using sensors and inspections, estimate and predict health conditions, and make optimal repair plans. However, a big challenge to implement PdM in practice is related to data. For example, which data needs to be monitored? How can sensor data and expert inspections be combined? How can models be trained from limited/biased/unlabeled data? How can models be adjusted to diverse environmental/operational conditions?
Project DescriptionThis project focuses on developing innovative methodologies for data-driven predictive maintenance, combining inspection strategies, health indicator models, and maintenance planning. These methodologies will address data challenges, such as combining inspection results and sensor data, training with incomplete data, and transferring models to different domain data. When implemented in systems, these will support real-time decisions of human operators for optimal performance in dynamic environments.
Job DescriptionYou will design and lead the research project with the guidance of the supervisory team (dr.
Juseong Lee, dr.
Claudia Fecarotti, and dr.
Alp Akçay ). You will conduct innovative research at the intersection between data science, industrial engineering, and reliability engineering. You will first focus on methodological approaches (improving and developing machine learning algorithms) and then move on to application-driven approaches (implementing algorithms in operational environments). You will publish the results in international journals and conferences to communicate with academia and other societal stakeholders.
Academic and Research EnvironmentThe project, the supervisors, and eventually also the PhD students are embedded in TU/e's Operations, Planning, Accounting, and Control group (OPAC)).
OPAC uses methods from operations research and operations management on a wide variety of problems, and currently hosts around 50 PhD students from various backgrounds. You will be able to collaborate with other PhD researchers in the domain of Data Driven Decision Making.