ContextTo optimally run industrial processes, it is key to use both AI and human operators' knowledge. In the process industry, deciding on the optimal settings requires experience and understanding of the process. The complex interplay between many internal and external factors makes it difficult for human operators to decide in real time what the optimal settings are based on all available data. Machine learning and AI can be used to make predictions about optimal settings, but final decisions and responsibilities remain for a large part with human operators
Project DescriptionIn the current developments towards human-centered Industry 5.0 we are looking for ways to include AI in the daily work of human operators. This project will specifically study how process operators can optimally interact with AI to make better informed decisions on process settings. To do this, we will explore different data-analytic (AI) models that can predict process settings, balancing model explainability and predictive performance. Multiple sources of data will be integrated into the predictive models to predict various aspects of the manufacturing processes. Next to studying human-AI interaction in simulated processes, you will explore how the interaction should be designed for optimal useability.
Job DescriptionAs a PhD candidate, you will be responsible for doing several behavioral studies into human-AI interaction together with the supervising team (dr. Geert van Kollenburg, dr. Lijia Tan, and dr. Rob Basten). Your research will be on an intersection between data science (Machine learning/AI), behavioral sciences and process analytics. Next to direct interactions with industrial partners, you will write your results in papers for academic journals and will visit conferences related to your work.
Academic and Research EnvironmentThe project and the team of professors are part of the Operations, Planning, Accounting, and Control group (
OPAC). OPAC specializes in various aspects of operations research, operations management, logistics, and optimization. This diverse group has many social activities and currently hosts around 50 PhD students. You will work together with other (PhD) researchers in the domain of Data Driven Decision Making.