I identified process conditions affecting the production of defective products in injection molding and die-casting manufacturing.

During my time as a postdoctoral researcher in my laboratory at Kyung Hee University, I participated in several projects in the area of Smart Manufacturing. I present the major research achievements below.

Project Synopsis
Title:Rule-based explanations based on ensemble machine learning for detecting sink mark defects in the injection moulding process
Highlights:
  • Ensemble learning models can predict the quality in the injection molding process.
  • Rule-based explanations are developed for interpreting the ensemble models.
  • The method generates decision rules and visualizes them with PDP and ICE plots.
Period: August 2020- July 2021
Outputs:
Skills
&
Technologies:
  • Ensemble learning
  • Explainable Machine Learning
  • Rule-based classifiers
  • Injection Molding Process
  • Product quality prediction
  • Academic writing
Project Synopsis
Title:Rule-based visualization of faulty process conditions in the die-casting manufacturing
Highlights:
  • Ensemble learning models can predict the quality in the die-casting process.
  • Rule lattices are developed for interpreting the ensemble models.
  • The method generates a hierarchical lattice that visually presents the relationships among process conditions, rules and predicted classes.
Period: August 2020- August 2022
Outputs:
  • Academic article in revision in the Journal of Intelligent Manufacturing
Skills
&
Technologies:
  • Explainable Machine Learning
  • Rule-based classifiers
  • Formal Concept Analysis
  • Die-casting Process
  • Product quality prediction
  • Academic writing