Files in this item



application/pdfChang_Song_2007_Haikou_Final.pdf (226kB)
paper for 5th China-Japn-US Symposium on Lifeline Earthquake EngineeringPDF


Title:Matrix-based System Reliability Analysis of Urban Infrastructure Networks: A Case Study of MLGW Natural Gas Network
Author(s):Chang, Liang; Song, Junho
Subject(s):lifeline network infrastructure system reliability GIS earthquake
Abstract:Urban infrastructure systems such as utility networks for electricity-, water-, sewage- and gas-services, transportation systems and telecommunication networks are critical backbones of modern societies. However, these systems are often susceptible to natural and man-made hazards. Structural damages of components in these infrastructure networks not only disrupt residential and commercial activities, but also impair post-disaster response and recovery efforts, resulting in substantial socio-economic losses. Therefore, estimating the reliability of these infrastructure systems is vital to urban stakeholders such as utility companies, urban planners, and policy makers as well as to residents and business owners. Evaluation of the performance and connectivity of such urban infrastructure systems is complex in nature due to the large number of network components, complex network topology and component/system interdependencies. Due to this complexity, network reliability analysis is often performed by repeated network analyses for random samples of hazard scenarios and component status, which prevents rapid risk assessment and near-real-time risk-informed decision making. In this paper, an analytical, i.e. non-sampling-based network reliability analysis method is proposed for urban infrastructure systems. First, a review of previous research on network reliability analysis is given, followed by a brief summary of the recursive decomposition algorithm (RDA) and the matrix-based system reliability (MSR) method that constitute the proposed network reliability analysis methodology. As a case study, the proposed methodology is applied to a Memphis Light, Gas and Water (MLGW) natural gas network of Shelby County of Tennessee. Based on seismic hazard maps from the Mid-America Earthquake Center’s risk management software MAEviz and the topological characteristics of an MLGW gas network, the reliability of the network components are computed by use of a geographic information system, ArcGIS. All the disjoint cut sets and link sets of the simplified MLGW gas network are efficiently identified by use of the RDA. The MSR method, which can account for statistical dependence between components and incomplete information, is employed to evaluate the connectivity reliability of the gas network in an efficient manner. The results of the system reliability analysis are presented for earthquake scenarios with different magnitudes. By integrating the MSR method with an advanced network analysis algorithm such as the RDA, we can perform network reliability analysis for a complex infrastructure system without random samplings. This approach will enable us to perform risk analysis and various statistical inferences rapidly. Based on the results of the case study, further research is in progress to (1) account for the statistical dependence between seismic intensities at adjacent network components, (2) estimate the average downtime and socio-economic losses, (3) account for the interdependency between different infrastructure systems, and (4) provide useful information for decision making on disaster planning, response, recovery, and mitigation using the proposed methodology.
Issue Date:2007-11-26
Citation Info:Chang, Liang and Junho Song. Matrix-based System Reliability Analysis of Urban Infrastructure Networks: A Case Study of MLGW Natural Gas Network, The Fifth China-Japan-US Trilateral Symposium on Lifeline Earthquake Engineering, Haikou, China, November 26-30, 2007.
Genre:Conference Paper / Presentation
Publication Status:published or submitted for publication
Peer Reviewed:is peer reviewed
Date Available in IDEALS:2008-04-11

This item appears in the following Collection(s)

Item Statistics