ISE Seminar Series: PhD Student Seminar Presentations

Date: October 15, 2021
Time: 11:00 a.m. - 12:00 p.m.
Location: Virtual event
Category: Seminar

Industrial & Systems Engineering Department

PhD Student Seminar Series

October 15th, 11:00 AM-12:00 PM ET | Zoom


Predicting Scan Quality: A Comparison of Machine Learning Models 

11:00 AM-11:20 AM ET

Seminar by Neda Sayahi, PhD Student
ISE Department, Wayne State University

Abstract: As a relatively new technology in manufacturing metrology, X-Ray computed tomography has recently become more established. However, setting scan parameters in a quick and proper manner is challenging due to high operator dependency and lack of traceability. We argue that machine learning (ML) can accelerate parameter setting process by eliminating the need for manual setting. In this work, the accuracy of four ML methods on predicting scan quality (whether the scan will be feasible or infeasible), given a set of parameters, are compared. The results indicated that multi-layer perceptron predicted the quality of scan with high accuracy and outperformed the other methods.

Bio: Neda Sayahi received her B.S. in Industrial Management and her M.S. in Operations Research, both from University of Tehran, Iran. Neda is currently a PhD student in the Department of Industrial and Systems Engineering at Wayne State University, pursuing a PhD degree in industrial engineering. Her research interests include utilizing machine learning and deep learning methods in additive manufacturing.

For publication citation and impacts, see

Google Scholar: https://www.researchgate.net/profile/Neda-Sayahi


Predictive Multi-Microgrid Generation Maintenance: Formulation and Impact on Operations & Resilience 

11:20 AM-11:40 AM ET

Seminar by Farnaz Fallahi, PhD Student
ISE Department, Wayne State University

Abstract: Industrial sensor data provides significant insights into the failure risks of microgrid generation assets. In traditional applications, these sensor-driven risks are used to generate alerts that initiate maintenance actions without considering their impact on operational aspects. In this talk, we will propose a framework that builds a seamless integration between sensor data and operational & maintenance (O&M) drivers in a multi-microgrid setting and demonstrates the value of this integration for improving multiple aspects of microgrids operations. The framework offers an integrated stochastic optimization model that jointly optimizes O&M. Operational uncertainty from renewables, demand, and market prices are explicitly modeled through scenarios. We use the model structure to develop decomposition-based solution algorithms to ensure computational scalability. Through a comprehensive set of experiments, we manifest the capability of the model in providing significant improvements in terms of reliability, costs, generation availability, & resilience.

Bio: Farnaz Fallahi received the B.S. degree in Pure Mathematics from AmirKabir University of Technology in 2011 and the M.S. degree in Industrial & Systems engineering from Sharif University of Technology in 2015, Tehran, Iran. She is currently pursuing the Ph.D. degree in Industrial & Systems Engineering at Wayne State University, Detroit, MI, USA. Her main research interests include sensor-driven prognostics, stochastic programming and optimization applied to power systems and electricity market.

For publication citation and impacts, see

Google Scholar: https://scholar.google.com/citations?user=TNy_DI8AAAAJ&hl=en&oi=ao


PM 2.5 Forecasting Utilizing Graph Convolutional and LSTM Neural Networks 

11:40 AM-12:00 PM ET

Seminar by Ali Kamali Mohammadzadeh, PhD Student
ISE Department, Wayne State University

Abstract: PM2.5, as inhalable particles with maximum diameters of 2.5 micrometers, are the cause of many serious health problems. Here, a PM2.5 forecasting framework is developed by integrating convolutional and recurrent neural networks. Although it is common to use recurrent neural networks to study the temporal behavior of PM2.5, this is the first work to take advantage of the geo-correlation of monitoring stations. Here, graph convolutional neural networks are implemented to exploit the nested structure of the data, composed of different time series of various meteorological factors over different monitoring stations in Michigan, feeding an LSTM model to improve the forecasting accuracy of PM2.5.

Bio: Ali Kamali Mohammadzadeh received his B.S. in Industrial Management from University of Tehran, Tehran, Iran. He received his first M.S. degree in Operations Research from University of Tehran and his second M.S. in  Data Science and Business Analytics from Wayne State University, MI, USA. Ali is currently a Ph.D. student in the Department of Industrial and Systems Engineering at Wayne State University, pursuing a Ph.D. degree in Industrial Engineering. His research interests include Extended Reality, Human-Robot Interaction, and utilizing machine learning, deep learning, and optimization models in manufacturing and healthcare.

For publication citation and impacts, see

Google Scholar: https://scholar.google.com/citations?hl=en&user=A5ILHr8AAAAJ

October 2021
SU M TU W TH F SA
262728293012
3456789
10111213141516
17181920212223
24252627282930
31123456