Research

My research Interests: 

• Integration of domain knowledge into the data-analytics/machine-learning models

• Data-driven Modeling, Design, Control, and Analysis on various Engineering Processes and Systems by leveraging the domain knowledge.

• Monitoring, Diagnostics, Prognostics, and Decision-making (Prescription) on Engineering systems

Methodologies of interest: Statistical Analysis, Bayesian Modeling, Bayesian Optimization, Gaussian Processes, Reinforcement Learning, Physics-Informed Machine Learning

Below are my previous and ongoing research works:

Tasks: 

• Quality Control (Process Monitoring [J3, C3], Prognosis [J12], Diagnosis [J2, J12], Preventive Maintenance [J9]), 

• Process/Design Optimization [J5, J6, J8, C5], 

• Statistical Calibration/Inference [J2, J4], 

• PDE Heterogeneous Parameter Estimation (Inverse Problem) [J11]

Approaches: 

• Probabilistic Modeling [J3, C3, J9], 

• Statistical/Bayesian Modeling [J2, J4, J12], 

• Bayesian optimization [J5, J6, C5], 

• Gaussian process modeling [J4, J5, J6, J12, C5], 

• Survival Model [J12], 

• Uncertainty Quantification [J2, J4, J5, J6, J12],

• Reinforcement Learning [J9],

• Physics-Informed Machine Learning [J11]

Applications:

• Advanced Manufacturing Process: e.g., Nano-scale Inkjet Printing, GFET Nanosensor Manufacturing & Design

• Manufacturing System: e.g., Autobody assembly

• Service System: e.g., 

• Clinical Imaging (Elastography)/Bone biomechanics

• Monitoring: Probability Modeling & Quality Control of Additive Nanosensor Manufacturing Process

  • Goal: Establishing a stochastic geometry model and a statistical quality control method using the image data

  • Methodologies: Stochastic geometry modeling, Stochastic process, physical property analysis

• [J3] J. Lee, S. Zhou, and J. Chen, "Statistical modeling and analysis of k-layer coverage of two-dimensional materials in inkjet printing processes," Technometrics, vol. 63, no. 3, pp. 410–420, 2021.

• [C3] J. Lee, C. Wang, S. Zhou, and J. Chen, "Spatial distribution quantification and control of ink flakes in reduced graphene oxide FET inkjet printing," Procedia Manufacturing, vol. 34, pp. 19–25, 2019.  47th SME North American Manufacturing Research Conference, NAMRC 47, Pennsylvania, USA.

• Data-driven Robust Design Optimiation of Advanced Engineering System or Nanosensors

  • Goal: Data-driven design method minimizing the variance while satisfying mean outputs leveraging prediction & uncertainty

  • Methodologies: Bayesian optimization, Gaussian process modeling

• [J6] J. Lee, S. Zhou, and J. Chen, "Robust Parameter Design on Dual Stochastic Response Models with Constrained Bayesian Optimization,", IEEE Transactions on Automation Science and Engineering, 2023, doi:10.1109/TASE.2023.3251973

Other Bayesian optimization (Process/System/Design optimization) papers:

• [J5] C. Huang, J. Lee, Y. Zhang, S. Zhou, and J. Tang, "Mixed-input Bayesian Optimization Method for Structural Damage Diagnosis," IEEE Transactions on Reliability, vol 72, no. 2, pp. 678-691, 2023. doi:10.1109/TR.2022.3179602

• [C4] J. Lee, S. Zhou, and J. Chen, "Sequential Robust Parameter Design With Sample Size Selection", ASME 2022 17th International Manufacturing Science and Engineering Conference, Indiana, USA, doi:10.1115/MSEC2022-85690

• Nonparametric Statistical Functional Data Modeling for Nanosensors & Statistical Calibration with Uncertainty Quantification

  • Goal: Uncertainty quantification, inference of underlying physical variable based on functional data

  • Methodologies: Functional data modeling, Gaussian Process modeling, Bayesian inference

  • 
    

• [J4] J. Lee, C. Wang, X. Sui, S. Zhou, and J. Chen, "Landmark-embedded Gaussian process with applications for functional data modeling," IISE Transactions, vol. 54, no. 11, pp. 1033-1046, 2022.

  • This paper was featured in the ISE Magazine (October 2022 issue)

IoT-enabled Smart and Connected System

• Diagnosis on Faulty Operation leveraging domain knowledge IoT-enabled Smart and Connected Manufacturing System

  • Goal: Identifying faulty operations with excessive variation in a multistage manufacturing system; establishing a sparse-variance statistical model.

  • Methodologies: Random-effects model, hierarchical Bayesian prior modeling

• [J2] J. Lee, J. Son, S. Zhou, and Y. Chen, "Variation source identification in manufacturing processes using Bayesian approach with sparse variance components prior," IEEE Transactions on Automation Science and Engineering,  vol. 17,  no. 3,  pp. 1469–1485, 2020.

• Preventive Maintenance on multi-component system (IoT-enabled Smart and Connected Production System)

  • Goal: Establishing a reinforcement learning method combined with a stochastic model to find the optimal preventive maintenance policy in a multi-components production system.

  • Methodologies: Deep Reinforcement learning, deep neural network, stochastic process modeling

• This work is done with a Ph.D. student, T. Srikitrungruang and Collaborator, Dr. S. Jahani

• [J9] J. Lee, T. Srikitrungruang, and S. Jahani, "Domain-Informed Reinforcement Learning for Multi-Component Preventive Maintenance Planning under Economic Dependence", under revision

• Physics-Informed Machine Learning to estimate heterogeneous stiffness (elastic moduli) of bones based on the bone deformation data

  • Methodologies: Physics-Informed Neural Network, deep neural network

  • PDE Equation: Linear Elasticity with heterogeneous elastic moduli

• This work is done with a Ph.D. students, T. Srikitrungruang, S Aghaee, an undergraduate student, M. Lemon, and  Collaborator, Dr. Yuxiao Zhou.

• [J11] T. Srikitrungruang, M. Lemon, S. Aghaee Dabanghan Fard, J. Lee, Y. Zhou,  "Robust Physics-Informed Neural Network Approach for Estimating Heterogeneous Elastic Properties from Noisy Displacement Data", https://doi.org/10.48550/arXiv.2506.14036, under review.

• Prognostics with diagnostics capability : Real-time failure time and types prediction and uncertainty quantification

  • Methodologies: Bayesian Hierarchical Modeling, Convolutional Gaussian Process, Variational Inference, Proportional Cox Hazard model

• This work is done with a Ph.D. student, S. Aghaee

• [J12] S. Aghaee Dabanghan Fard, A. Deep, M. Kim, J. Lee, "Bayesian Joint Model of Multi-Sensor and Failure Event Data for Multi-Mode Failure Prediction'', https://doi.org/10.48550/arXiv.2506.17036, under review.

• Collaborated Research: Fully Inkjet-Printed, 2D Materials-Based Field-Effect Transistor for Water Sensing

  • Detail: Printing uniform channels by the (nano) inkjet printer is extremely challenging due to the well-known coffee-ring effects. The quality of the inkjet printed pattern hinges on many factors, including various material properties and temperature, which are often difficult to obtain. In this work, we proposed a method to achieve a uniform channel covered by an average of a single layer of graphene. By leveraging the fact that multiple printing passes can smooth out the coffee ring effects on the boundary, we formulate an optimization problem and achieve a uniform single layer coverage on the target region.

• [J8] X. Sui, S. Rangnekar, J. Lee, S. Liu, J. Downing, L. Chaney, X. Yan, H. Jang, H.Pu, S. Zhou, M. Hersam, and J. Chen, “Fully Inkjet-Printed, 2D Materials-Based Field-Effect Transistor for Water Sensing,” Advanced Materials Technologies, 2023, doi:10.1002/admt.202301288