Files in this item



application/pdfMILLER-THESIS-2015.pdf (5MB)
(no description provided)PDF


Title:Integrating geographic information systems with the Level 3 Probabilistic Risk Assessment of nuclear power plants to advance modeling of socio-technical infrastructure in emergency response applications
Author(s):Miller, Ian M
Department / Program:Nuclear, Plasma, & Rad Engr
Discipline:Nuclear, Plasma, Radiolgc Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Risk-informed emergency preparedness
Level 3 probabilistic risk assessment
social vulnerability
Abstract:Explicit incorporation of social and organizational factors into Level 1 Probabilistic Risk Assessment (PRA) has been theoretically and methodologically improved and now is in the process of development for Nuclear Power Plant (NPPs) applications. The goal of this study is to initiate the same paradigm of research for Level 3 PRA. Explicit incorporation of social factors, most specifically location-specific social factors into Level 3 PRA, can drastically affect decisions related to emergency planning, preparedness, and response (EPPR). With concerns about population response from a radiological accident such as the one that occurred at the Fukushima Daiichi Nuclear Power Plant in 2011, understanding the implications of the social makeup of the population in the vicinity of an NPP has the potential to give decision makers information about the effects of their decisions. This research proposes theoretical and methodological approaches to explicitly consider the social factors of the local population in NPP accident consequence modeling. In a Level 3 PRA, the MELCOR Accident Consequence Code System (MACCS2) developed by Sandia National Laboratory, is used by the U.S. Nuclear Regulatory Commission and nuclear industry in order to estimate the damages to public health and environment in the case of an NPP severe accident leading to a large radiological release into the atmosphere. The goal for this research is to derive and incorporate location-specific human and organizational factors, socio-political/ socio-economic climate, and community-specific characteristics into a Level 3 PRA. This has been done “externally” by the integration of MACCS2 with Geographic Information Systems (GIS). Esri’s ArcGIS Version 10.2 software is utilized to operationalize this study. A Bayesian Belief Network (BBN) methodology is also proposed as an approach to “internally” incorporate social risk-contributing factors into a Level 3 PRA code. In this research, social vulnerability construct is used, as a surrogate for a causal model, to integrate social factors with a Level 3 PRA. There have been over five decades of research dedicated to the development of quantifiable social vulnerability factors and models that point toward a prediction of consequences to a population, given a specific hazard. Most of these studies have been concentrated on natural hazards; yet, none have been applied to the man-made hazard (i.e., radiation) related to NPPs. This research study combines social and technical contextual factors with radiation and contamination hazard characteristics based on a specific NPP in order to advance risk assessment and management for NPP severe accidents. Specific demographic information is integrated into social vulnerability and includes house value, age, minority status, and gender. This social vulnerability is associated with the population’s ability to evacuate the area, namely to define evacuation delay time and evacuation speed within the population evacuation model. This research spans two very diverse areas of study; (1) Probabilistic Risk Assessment (PRA) as originated in nuclear engineering, and (2) social vulnerability analysis which is primarily conducted in geography and the social sciences. The contributions of this research include: 1. Theoretical contributions to support applying social vulnerability frameworks to NPP accident consequence analysis, covered in chapter 2. This research is the first of its kind to bridge the gap between social vulnerability theories and nuclear power risk analysis, and consists of a thorough literature review spanning many diverse areas of research 2. Methodological contributions toward combining an accident consequence code such as MACCS2 with the quantification of social vulnerability in the form of a social vulnerability index, covered in chapter 3. This methodology has been established in natural hazards research, and never in the context of probabilistic nuclear accident consequence codes. 3. Methodological contributions toward the integration of an accident consequence code such as MACCS2 with Geographic Information Systems (GIS) to visualize risk information and to explicitly and externally integrate social factors with MACCS2. This has been demonstrated in chapters 3 and 4. 4. Methodological contributions to explicitly and internally merge social vulnerability indices with the evacuation module in MACCS2, using Bayesian Belief Network (BBN). This has been explained in chapter 5. 5. Practical contributions including explicit consideration of location-specific social factors in Level 3 PRA that will help develop: (i) more realistic modeling of population response and, therefore, a more accurate estimation of NPP severe accident risk; and (ii) more advanced management of NPPs severe accident risk by facilitating the analysis of the effects of change in risk due to changes in the underlying socio-technical risk contributing factors. This will certainly help advance models and applications of risk-informed EPPR, particularly in focusing on location-specific populations who rank highest with respect to risk. A further contribution is to visualize location-specific radiological risk around a NPP in order to improve risk communication with the public and policy makers.
Issue Date:2015-05-01
Rights Information:Copyright 2015 Ian Miller
Date Available in IDEALS:2015-07-22
Date Deposited:May 2015

This item appears in the following Collection(s)

Item Statistics