Sudipta Chowdhury

Research

My research centers on infrastructure system of systems resilience. Related work covers supply chain and logistics and technology in engineering education.

Infrastructure system of systems resilience

Resilience planning and disaster risk management for interdependent infrastructure systems. This work addresses post-disaster restoration planning, agent-based models of network and community recovery, and future-proofed infrastructure planning.

Representative publications

  1. Hughes, W., Chowdhury, S., Zhu, J., and Zhang, W. (2026). “Strengthening Resilience: An Agent-Based Model for Poststorm Restoration and Grid Hardening in Power Distribution Systems.” ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, 12(3), 04026038.
  2. Hatfield, J., Chowdhury, S., and Alzarrad, A. (2026). “Modeling Community Resilience Under Prolonged Disruption: An Agent-Based Framework Integrating Social Connectivity, Migration, and Policy-Driven Allocation.” Analytics, 5(3), 20.
  3. Zhu, J., and Chowdhury, S. (2022). “Building Resilience in Civil Infrastructure Systems: System-Level Biomimicry to Address Complex Managerial Challenges.” ASCE Journal of Infrastructure Systems.
  4. Chowdhury, S., Zhu, J., and Zhang, W. (2020). “Optimized Restoration Planning of Infrastructure System-of-Systems using Heterogeneous Network Flow Simulation.” ASCE Journal of Computing in Civil Engineering, 34(5).

All publications

  • Supply chain and logistics

    Value chain risk management, humanitarian logistics, and logistics service sharing.

  • Engineering education

    Design and application of innovative technologies in engineering education.

ATLAS Lab

I co-direct the Advanced Transportation, Logistics & AI for Resilient Systems (ATLAS) Lab.

Focus areas

  • Advanced transportation systems
  • Logistics optimization
  • Artificial intelligence
  • Resilient infrastructure

Theory and methods

Research theory
Information science, complex systems science, network science, decision theory, and behavioral science.
Research methods
Linear and non-linear programming, stochastic and robust optimization, AI informed decision analytics, and discrete and agent-based simulation modeling.
Programming
Python, Java, C++
Statistical software
R, MATLAB, SPSS, Minitab
Simulation software
AnyLogic
Cloud computing
Azure

Research experience

  • Marshall University

    Assistant Professor

    • Designing decision-support systems for railroad infrastructure maintenance under physical and environmental stressors
    • Developing community resilience models for small and medium enterprise (SME) manufacturing networks facing natural disasters
    • Creating virtual reality training modules for construction workforce safety and skill development
    • Conducting research on technology-enabled education frameworks to reduce attainment gaps among undergraduate engineering students
    • Analyzing and developing industrial ecology frameworks for end-of-life products, such as electric vehicle batteries

  • University of Connecticut

    Research Specialist and Graduate Assistant

    • Designed a future-proofed planning framework for predicting different climate risks (e.g., sea level rise) and analyzed its effects on transportation infrastructures
    • Analyzed the energy generation (e.g., natural gas based) and distribution landscape in Connecticut
    • Developed community centered power/electricity system restoration planning models
    • Analyzed consumer power demand trend during COVID-19
    • Conducted surveys and expert workshops to gather owner, engineering-contractor, and supplier firms' data and analyzed them to find hidden inefficiencies in existing business practices for construction industries
    • Designed a study for UConn undergraduate students to determine the effects of online and virtual learning during COVID-19 on class participation and understanding of the topics

  • Mississippi State University

    Research Assistant

    • Proposed drone routing models using mixed integer linear/non-linear programming (MILP/MINLP) for monitoring and surveillance of natural disaster impacted regions
    • Developed exact, meta-heuristics, and evolutionary algorithms based customized algorithms for enhancing computational efficiency
    • Analyzed CTU-13 dataset and utilized graph based feature clustering and artificial neural network to predict anomalies and provide disruption prevention strategies in large scale cyber networks
    • Developed supervised and unsupervised machine learning algorithms to analyze critical manufacturing process output data to predict discrepancies in test specimens without destructive testing

Research project collaborators

  • National Science Foundation (NSF)
  • U.S. Army Engineer Research and Development Center (ERDC)
  • National Aeronautics and Space Administration (NASA)
  • Occupational Safety and Health Administration (OSHA)
  • Civil-Military Innovation Institute (CMI2)
  • Pacific Northwest National Lab (PNNL)
  • U.S. Army Research Laboratory (ARL)
  • Transportation Infrastructure Durability Center (TIDC)
  • Center for Advanced Multimodal Mobility Solutions and Education (CAMMSE)
  • Construction Industry Institute (CII)
  • Eversource Energy
  • FedEx