Oklahoma State UniversityAhmad SalehiyanIndustrial Engineer

Welcome to my world

Hi, I'm Ahmad

Ph.D. researcher in Industrial Engineering and Management at Oklahoma State University. I build machine learning and optimization methods for decision-making under uncertainty, with applications in health data analytics and engineering systems.

Ph.D. Researcher — Oklahoma State University
Deep Reinforcement Learning
Machine Learning
Stochastic Optimization
Explore My Approach View PublicationsContact MeDownload CV

Best Skill On

Python
GAMS
Power BI
Excel
Ahmad Salehiyan - Ph.D. Researcher

About

Data-Driven Engineering Profile

I am a Ph.D. researcher in Industrial Engineering and Management at Oklahoma State University (expected Summer 2027), with concurrent graduate training in Applied Statistics. My research combines machine learning, deep reinforcement learning, and stochastic optimization for data-driven decision-making under uncertainty. My published work includes an analysis of comorbidity and mortality patterns in the MIMIC-IV electronic health records dataset (179,524 adult patients) and transformer-based deep learning models in two peer-reviewed journals.

Research interests: health data analytics and electronic health records; predictive modeling and risk stratification; deep learning and deep reinforcement learning; sequential decision-making under uncertainty (MDP/POMDP); stochastic optimization.

Experience

Professional Timeline

Aug 2023 – Present

Graduate Research Assistant

Oklahoma State University

Research in machine learning, deep reinforcement learning, and stochastic optimization for decision-making under uncertainty. Advisor: Dr. Akash Deep.

  • Co-authored the MIMIC-IV comorbidity and mortality study published in the IISE Annual Conference proceedings (2025).
  • Developed POMDP-based multi-sensor maintenance policies and hierarchical DDQN warehouse picker routing.

2025 – Present

Graduate Teaching Assistant / Instructor

Oklahoma State University

Courses: ENGR 1412 Introductory Engineering Computer Programming (Fall 2026), Python Programming (Fall 2025), Engineering Economics (Summer 2025), Python for Data Analysis (Spring 2025).

  • Fall 2026: weekly Excel assignments, grading, and help sessions for ENGR 1412 lab sections.
  • Taught Python Programming, Engineering Economics, and Python for Data Analysis in 2025.

Oct 2018 – Apr 2020

Research Assistant

K. N. Toosi University of Technology

Engineered time-domain, frequency-domain, and wavelet-based features from sensor signals for early fault detection in industrial machinery.

  • Used feature-importance analysis to identify critical signals.
  • Designed statistical thresholds for fault diagnosis.

Apr 2019 – Oct 2019

Industrial Engineer

Karin Crane Company

Coordinated process improvements and documented operational workflows across production teams; supported resource planning, scheduling, and quality control.

  • Documented operational workflows and coordinated process improvements.
  • Supported resource planning, scheduling, and quality control.

Education Quality

Academic Foundation

Ph.D., Industrial Engineering and Management

Oklahoma State University

Expected Summer 2027

Dissertation (proposed): “Sequential Decision-Making Under Uncertainty in Smart Engineering Systems.”

Advisor: Dr. Akash Deep

M.S., Applied Statistics

Oklahoma State University

Expected 2026

Experimental design, statistical programming (R/SAS), data analytics, and statistical inference.

M.S., Industrial Engineering

K. N. Toosi University of Technology, Tehran

2019–2022

Thesis: “Predictive Maintenance of Advanced Industrial Machines Using AI Techniques.”

Advisor: Dr. Abdollah Aghaie

B.S., Industrial Engineering

Islamic Azad University, Qazvin

2014–2019

Professional Skills

Software, Programming, and Methods

Analytics programming

Python (NumPy, SciPy, pandas, PyTorch, Gym)
R
SAS
Julia
GAMS
Pyomo

Databases and large-scale data

SQL and PostgreSQL (relational databases)
Apache Spark / PySpark (coursework level)
High-performance computing cluster
GPU model training

Machine learning and deep learning

Transformer architectures (the model family underlying large language models)
Generative adversarial networks
Autoencoders
Deep Q-networks (DQN, DDQN)
k-means clustering
Association rule mining (Apriori)
t-SNE
Anomaly detection
Time-series feature extraction

Health data

Electronic health records (MIMIC-IV)
Coded diagnosis data
Comorbidity and mortality analysis

Decision-making and optimization

MDP
POMDP
Monte Carlo simulation
Mixed-integer and linear programming
Benders and Lagrangian decomposition
Value and policy iteration

Research tools and existing tools

Git/GitHub
LaTeX
Simulation environments
Experiment logging
JavaScript
HTML/CSS
Scikit-Learn
Power BI
Tableau
Primavera P6
Excel
Anaconda
jQuery

Professional Certifications

Formal Training and Certification Track

ISO 9001:2015 — Quality Management Systems

ISO 22000 — Food Safety Management Systems

Advanced Microsoft Excel (48-hour)

Microsoft Project 2016 (24-hour)

Production Planning & Inventory Control (24-hour)

ICDL (98-hour)

Publications

Research Publications

Peer-reviewed journal articles

  • A. Salehiyan, N. Serrano, F. Hernando-Gallego, D. Martín, J. V. Álvarez-Bravo. "A Novel Evolutionary Optimized Transformer-Deep Reinforcement Learning Framework for False Data Injection Detection in Industry 4.0 Smart Water Infrastructures." Computers, Materials & Continua, 87(2), 68, 2026.
  • A. Salehiyan, P. S. Moghaddam, M. Kaveh. "An Optimized Transformer–GAN–AE for Intrusion Detection in Edge and IIoT Systems: Experimental Insights from WUSTL-IIoT-2021, EdgeIIoTset, and TON_IoT Datasets." Future Internet, 17(7), 279, 2025.

Conference proceedings

  • P. Vaghfi Mohebbi, A. Salehiyan, A. Deep. "Disease Cluster Analysis in Electronic Health Records: Insights into Mortality and Comorbidity Patterns." Proceedings of the IISE Annual Conference & Expo, pp. 1–6, 2025.

Preprints and manuscripts under review

  • A. Salehiyan, J. Lee, M. Kim, A. Deep. "Deviation-Informed Maintenance Planning: Cost Implications of Departures from Sensor-Based Optimal Policy." Submitted to INFORMS Journal on Data Science, 2025.
  • A. Salehiyan, J. Lee, M. Kim, A. Deep. "A Scalable POMDP-based Framework for Real-Time Maintenance Planning using Multiple Sensor Signals." TechRxiv preprint, 2024.
  • A. Salehiyan, M. Masae, P. Vichitkunakorn, A. Deep. "A Hierarchical Deep Q-Learning Approach for Picker Routing in Two-Block Warehouses." SSRN preprint No. 7052096.
Full list and citations on Google Scholar

Presentations

  • IISE Annual Conference & Expo, Atlanta, GA, 2025 (Talk): Disease Cluster Analysis in Electronic Health Records
  • INFORMS Annual Meeting, Atlanta, GA, 2025 (Talk): Multi-Agent Framework for Scalable Fleet Maintenance Planning
  • Reliability and Maintainability Symposium (RAMS), Miramar Beach, FL, 2025 (Talk)
  • INFORMS Annual Meeting, Seattle, WA, Oct 2024 (Talk)
  • OSU Student Research Symposium, Stillwater, OK, 2024 (Poster, 3rd Place)

Insights

Technical Articles

Machine Learning for Industrial Reliability: From Theory to Deployment

A practical guide to using supervised, unsupervised, and reinforcement learning in reliability-centric operations without overpromising black-box AI outcomes.

Read Article

Integer Programming for High-Stakes Planning Decisions

A practical, engineering-first explanation of integer programming formulations and classical solution approaches for planning, allocation, and scheduling under constraints.

Read Article

Maintenance Management Systems That Scale with Operational Complexity

A structured playbook for building maintenance systems around service level, work-order quality, and CMMS-enabled execution.

Read Article

Looking for applied research collaboration in reliability engineering or operational analytics? Let's discuss your use case.

Contact

Start a Conversation

Share your reliability, maintenance, or analytics challenge and I will respond with a practical next-step framework.

EmailLinkedInGitHubGoogle ScholarTelegramWhatsApp
Mailto Fallback

Insights Preview

Latest Technical Insights

Integer Programming

Formulating mixed-integer programs to optimize warehouse order-picking efficiency and reduce routing overhead.

Machine Learning

Applying machine learning and deep reinforcement learning to decision-making under uncertainty in engineering systems.

Maintenance Management

Bridging stochastic POMDP models with physical system constraints to optimize real-world maintenance scheduling.