• 2019

    • Twitter and Disasters: a Social Resilience Fingerprint. Rachunok, Bennett, Nateghi IEEE Access 2019

    • The Sensitivity of Electric Power Infrastructure Resilience to the Spatial Distribution of Disaster Impacts. Rachunok, Nateghi. Reliability Engineering & System Safety (In Press). 2019.

    • Evaluation of Parametric Wind Power Scenarios. Rachunok, Staid, Watson, Woodruff. (Under Review)

    • A Path Forward for Leveraging Social Media to Improve the Study of Community Resilience. Rachunok, Bennett, Flage, Nateghi. (Under Review).

    • Decoding Regional Climate Attitudes by Integrating Social Media and Survey Data. Bennett, Rachunok, Flage, Nateghi. (Under Review).

    • The Fallacy of Symmetric Temperature Response of Electricity Demand: a Case Study for California. Kumar, Rachunok, Silva, Nateghi. (Under Review)

  • 2016

    • Rachunok BA, Mayorga ME, 2016. UAVs Provide Lifesaving Medical Care. Proceedings of the 2016 IIE Annual Conference, May 2016. Manuscript Copy

in preperation

  • Improved Assessment of Sustainability: Away from Ad Hoc Rankings. Paulvannan, Obringer, Rachunok, Nateghi. (November 2019)

  • Appraisal Theory in Community Resilience. Rachunok, Zhu, Nateghi. (October 2019).

  • The Cost of Poisson Assumptions in Hurricane Risk Models. Rachunok, Staid, Nateghi. (September 2019).



  • INFORMS 2019. Seattle Washington. Come see my session on Data Driven Disaster Resilience on Monday October 21st at 11:00am in the Yakinma room 57.

  • SRA 2019. Crystal City, VA. Hosting a session on social media and community resilience, details TBD.

  • 2019

    • IISE, May 2019, Orlando, FL. Understanding the importance of generalization in network models.

  • 2018

    • Society of Risk Analysis, New Orlines, LA. December 2018

    • INFORRMS Annual Meeting, Phoenix, AZ. November 2018. The sensitivity of electric power infrastructure resilience to the spatial distribution of disaster impacts informs2018.pdf

    • CoDA, March 2018, Santa Fe, NM Probabilistic Wind Power Scenarios POSTER

  • 2017

    • December, Society of Risk Analysis Annual Conference,Transportation Network Recovery Analysis, an equilibrated perspective. Arlington, VA SLIDES

    • October, INFORMS Annual Conference. Modeling Uncertainty, The cost of wrong assumptions. Houston, TX SLIDES

    • October, INFORMS Annual Conference.Assessing Changes in Residential Electric Power Consumption Due to Emerging Changes in Climate and Technology. Houston, TX

    • April, EEE Seminar Series, Risk Analysis for Power Systems Planning. Purdue University SLIDES

  • 2016

    • September, IE 590, Nature Inspired Computing. Class presentation on Empirical Hardness Modeling. SLIDES

    • May, IISE Annual Conference, Anaheim, CA. 15 minutes research presentation.

    • April, NCSU Research Symposium. 15 minute research presentation. SLIDES


  • 2018

    • Lee Chaden Fellowship, Purdue University

  • 2017

    • Estus and Vashti Magoon Award for Teaching Assitant Excellence, Purdue University

  • 2016

    • IISE Annual Conference Operations Research Undergraduate Research Dissemination Contest: 1st place

in the news

twitter and disasters


  • I was a caller on Science Friday in December of 2016. The episode can be heard here. My call starts after the 26 minute mark.

  • My undergraduate research on UAVs and their potential use delivering defibrillators to patients experiencing cardiac arrest was featured in a few popular news outlets.

WNCN picture

the lab

  • News about LASCI can be seen here!

other things

previous research

Briefly, the research looked at the feasibility of using UAVs (drones) carrying defibrillators to respond quicker to cardiac arrest. Results showed adding one UAV to a medium-large city (test case Charlotte, NC) could reduce the avereage EMS response time to out of hospital cardiac arrests by approximately 15 seconds. A more detailed explination is available at my previous website.

Latent Variable Symbolic Regression

As a part of a course on Nature Inspired Computing taken in Fall of 2016, I worked on a project using latent variables to reduce the dimensionality of a large dataset with the goal of predicting the number of Tropical Cyclones in a year. The idea was based on work done by Trent McConaghy in 2009 (this book chapter is freely available on google). Put simply, the LVSR method attempts to shrink a dataset with a large number of variables down such that Symbolic Regression (which works best on smaller datasets) can be applied to develop a model. A vector as long as the number of variables is used to shrink each observation to a scalar. Symbolic Regression is applied and the residuals are used to develop the next model. This is done iteratively until the model converges. The full paper is attached here. Code and data are available upon request.