Title: Data-Driven Prediction and Adaptive Decision-Making for Sequential Monitoring of Complex Systems

Date: Friday, May 29th, 2026

Time: 1pm-3pm EST

Location: Groseclose 404

Meeting link: https://gatech.zoom.us/j/97030539974

Dongmin Li

Ph.D. Candidate in Industrial Engineering

H. Milton Stewart School of Industrial and Systems Engineering

Georgia Institute of Technology

Committee

Dr. Xiaochen Xian (Advisor), H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology

Dr. Jianjun Shi, H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology

Dr. Roshan Joseph, H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology

Dr. Xiao Liu, H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology

Dr. Hongcheng Liu, Department of Industrial and Systems Engineering, University of Florida

Abstract

Sequential monitoring of complex systems, such as advanced manufacturing and environmental monitoring, is essential for ensuring system reliability and performance. However, challenges arise when evaluating system status from real-time observations and making detection decisions. Resource constraints, such as limited sensors and data storage, restrict observations to a subset of system variables, requiring adaptive sampling decisions for data collection that account for uncertainty in system status evaluation and potential changes. In addition, system variables often exhibit intricate interactions and may evolve over time, requiring the consideration of the interactions and prediction of system evolution for accurate and timely detection.

To address the challenges, this thesis leverages available data and system knowledge, such as spatial correlation and underlying physics, to enhance system status prediction, evaluation, and adaptive decision-making, ensuring effective detection performance. Chapter 2 proposes data-driven sampling strategies for moving vehicle-based sensors (MVSs) to quickly detect abrupt changes in an area. By integrating statistical process control and mathematical optimization, the strategies handle noisy and partial observations and adaptively adjust MVS routes under movement constraints. Chapter 3 further proposes a Bayesian jump model-based pathwise sampling approach that improves detection performance using MVSs by incorporating spatial correlation and quantifying uncertainties. Chapter 4 addresses monitoring design for optimal statistical or economic performance under partial observations. Through theoretical analysis that derives average detection delays as explicit functions of design parameters and system settings, such as the numbers of observed and total variables, a design framework is formulated to determine the optimal parameter setting to meet detection goals while accounting for sensor deployment costs. Chapter 5 leverages underlying physics for accurate prediction, developing a physics-informed machine learning framework to predict droplet evolution in the inkjet printing process. By incorporating the volume of fluid formulation, it seamlessly integrates the governing physics of the velocity and pressure fields with observed sequential images of droplet shapes, thereby achieving accurate future droplet shape prediction, reconstructing the underlying physics fields based on image observations, and supporting analysis of droplet evolution behaviors.