Decision Analysis for Toxic by
Inhalation Material Routing (DATIMR)

Applications Seminar/Research Project
Spring 2008

Railroad Team
Matthew Albin
Ruben Luna
Danielle Martin
Danielle Obuchon
Andrew Ramsay

| Abstract | Documents | Selected References |

Sponsored by
Dr. Rajni Goel Howard University
Mr. Mark Hartong Federal Railroad Administration
Dr. Duminda Wijesekera George Mason University

Project Director
Dr. Kathryn B. Laskey
George Mason University
Department of Systems Engineering
and Operations Research

Fairfax VA

The views expressed in this report do not necessarily reflect those of the United States Government, the US Department of Transportation, the US Department of Energy, the Federal Railroad Administration, the Oak Ridge National Laboratory, or their associated contractors/subcontractors. Material in this report shall not be used for advertising or other commercial activities without the express written consent of the United States Government, the Department of Transportation, the Federal Railroad Administration, the Department of Energy, and the Oak Ridge National Laboratory.

In the United States, transportation of toxic by inhalation (TIH) chemicals by rail constitutes a hazard to the general public as can be observed in recent train incidents that resulted in loss of life, injuries, damage to infrastructure, activation of first responder resources, and multiple millions in costs for liability settlements. Furthermore, given the terrorist attack threat, railroads are vulnerable to malicious attacks with potential catastrophic effects to the public, economy, and national security. Carriers (railroads), shippers (chemical industry), federal and local authorities, and the general public have engaged in a national-level initiative to explore options to reduce this security risk. This paper investigates a simplified decision model to select train routes that minimize the security risk of an attack premised on using TIH materials as a weapon. The proposed model identifies basic transport cost, fatality, injury, and remediation as the primary parameters of interest to traffic managers. By combining indications and warning information from intelligence resources with these parameters, the model provides railroad traffic managers with an understanding of the effect of key drivers on route optimality. The model was applied to route analysis for the transport of chlorine between Alexandria, Virginia, and Philadelphia, Pennsylvania. It was found that in high risk situations, a railroad traffic manager should consider re-routing TIH material trains through longer but less populated areas to reduce risk. For low risk situations, the current, shorter route through more density populated areas constitutes a viable choice. This decision support model proposes a probabilistic, deterministic approach for estimating the chances and impacts of incidents and informs a simplified decision tree. An alternative approach suggests a stochastic solution should the deterministic approach prove computationally expensive.

Final Project Proposal
Capstone Presentation
Project Final Report

Selected References
Carter, M., Howard, M., Owens, N., Register, D., Kennedy, J., Pecheux, K., Newton, A., Effects of Catastrophic Events on Transportation System Management and Operations, Department of Transportation, July 2001.

Hartong, M., Goel, R., Wijesekera, D., A Risk Assessment Framework for TIH Train Routing, Federal Railroad Administration, Howard University, George Mason University, Dec 2007.

Johnson, P. E., Railroad Routing Visualization and Analysis (RRVA) Tool Userís Manual, Oak Ridge National Laboratory, Oak Ridge, TN, May 2006.

Kawprasert, A. and Barkan C. P.L., Reducing Railroad Hazardous Materials Transportation Risk by Route Rationalization, Railroad Engineering Program, University of Illinois at Urbana-Champaign, January 2008.

Parentela, E., Risk Modeling For Commercial Goods Transport Final Report California State University, Long Beach, CA, July 2002.

Pate-Cornell, E., Probabilistic Modeling of Terrorist Threats: A Systems Analysis Approach, Stanford, May 2004.