Manufacturing Process Optimization

Computer experiments, surrogate modeling, process optimization, and control for small sample-size manufacturing system.

  • Methodologies: sparse learning guided by physics, Gaussian process, linear regression.
  • Applications: fuselage assembly, ship assembly, multistage manufacturing process

Contents

1. Projects

1.1. Automatic Data Synchronization and Identification of Crucial Phases and Variables in the Ceramic Firing Process

Students: Yukun Xie

research/optimization/ceramics_firing.jpg

Ceramics Firing Process.

Motivation

  • The firing process for ceramic tiles is a MMP process with multiple sensors to monitor the temperature, pressure and equipment parameters. The data contains rich information​

  • The streaming data collected from sensors are separated and not related to the quality characteristics of each product

Objective

  • Synchronize the continuous sensor data into product-oriented data and to conduct relationship mining in MMPs​

Challenges

  • Time delay​
  • Missing sensors and multiple sensors at some stages​

Proposal

  • Automatic data synchronization to get the product-oriented mixed data​
  • Model the relationship between the mixed data and quality characteristics ​