Please use this identifier to cite or link to this item: http://theses.ncl.ac.uk/jspui/handle/10443/2529
Full metadata record
DC FieldValueLanguage
dc.contributor.authorRowland, Adewumi-
dc.date.accessioned2015-02-26T14:24:05Z-
dc.date.available2015-02-26T14:24:05Z-
dc.date.issued2011-
dc.identifier.urihttp://hdl.handle.net/10443/2529-
dc.descriptionPhD Thesisen_US
dc.description.abstractIn reported pipeline failures globally, third-party interference (TPI) has been recognised as a dominant failure mechanism in the oil and gas industry, although there has been limited research in this area. The problem is receiving considerable attention within the oil and gas industry, because of the industry threats (e.g. Al Qaeda's capabilities) and the natural vulnerability of pipelines because of their long distance network distribution. The ability to predict and secure pipelines against TPI is a valuable knowledge in the pipeline industry, and especially for the safety of the millions of people who live near pipelines. This thesis develop an understanding of the relationships between the many and various contributory factors leading to potential TPI, frequently resulting in mass deaths, economic losses, and widespread destruction to property. The thesis used GIS-based spatial statistical methodologies, first, based on hotspot and cold spot cluster analyses to explain pipeline incident patterns and distributions; and a geographically weighted regression (GWR) model to investigate the determinants of TPI and to identify local and global effects of the independent variables. Secondly, a generalized linear model (GLMs) methodology of Poisson GLMs and Logistic Regression (LR) procedures, by using a combination of land use types, pipeline geometry and intrinsic properties, and socioeconomic and socio-political factors to identify and predict potentially vulnerable pipeline segments and regions in a pipeline network. The GWR model showed significant spatial relationship between TPI, geographical accessibility, and pipeline intrinsic properties (e.g. depth, age, size), varying with location in the study area. The thesis showed that depth of pipeline and the socio-economic conditions of population living near pipeline are the two major factors influencing the occurrence of TPI. This thesis have prompted the need for selective protection of vulnerable segments of a pipeline by installing security tools where most needed. The thesis examined available literature and critically evaluated and assessed selected international pipeline failure databases, their effectiveness, limitations, trend, and the evolving difficulties of addressing and minimising TPI. The result of the review showed irregular nomenclature and the need for a universal classification of pipeline incidents database. The advantages and disadvantages of different detection and prevention tools for minimising TPI, used in the pipeline industry are discussed. A questionnaire survey was developed and employed, as part of the thesis, for the employees and managers in the pipeline industry. The results of the data analysis has contributed to the body of knowledge on pipeline TPI, especially the industry perceptions, prevention strategies, capabilities and complexities of the various application methods presently being implemented. The thesis also outlined the actions that governments and industry can and should take to help manage and effectively reduce the risk of pipeline TPI. The results of this study will be used as a reference to develop strategies for managing pipeline TPI. The results of the thesis also indicated that communications with all stakeholders is more effective in preventing intentional pipeline interference, and that the government's social responsibility to communities is the major factor influencing the occurrence of intentional pipeline TPI.en_US
dc.description.sponsorshipPetroleum Technology Development Fund (PTDF), Nigeria:en_US
dc.language.isoenen_US
dc.publisherNewcastle Universityen_US
dc.titleGIS-based prediction of pipeline third-party interference using hybrid multivariate statistical analysisen_US
dc.typeThesisen_US
Appears in Collections:School of Marine Science and Technology

Files in This Item:
File Description SizeFormat 
Rowland A. 2011.pdfThesis7.24 MBAdobe PDFView/Open
dspacelicence.pdfLicence43.82 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.