Title: Developing a monitoring system for long-distance pipeline leakage incorporating fusion of conflicting evidences

Authors: Desmond Adair; Hosam Emara-Shabaik; Martin Jaeger

Addresses: Department of Mechanical Engineering, School of Engineering, Nazarbayev University, Astana, 010000, Republic of Kazakhstan ' Department of Mechanical Engineering, School of Engineering, Nazarbayev University, Astana, 010000, Republic of Kazakhstan ' Department of Civil Engineering, School of Engineering, University of Tasmania, Sandy Bay Campus, TAS7001, Australia

Abstract: Quite often, during long-distance pipe leakage, critical decisions have to be made quickly and with as much certainty as possible using conflicting and uncertain sources of information. Key to providing quality decision options is an appropriate method of combining, or fusing, these heterogeneous evidence sources. Since the development of belief function theory introduced by Shafer in the 1970s many combination rules have been proposed in the literature because in highly conflicting situations the emblematic Dempster's rule generates counter-intuitive and unacceptable results in practical applications. In this work, the Dezert-Smarandache theory (DSmT) will be explored, in particular the PCR5 and PCR6 rules of proportional conflict redistribution, and a simple long-distance pipe leakage monitoring method is developed to help in the development of a method for more complicated situations.

Keywords: pipeline leakage; evidence fusion; conflicting evidence; pipeline monitoring; Dezert-Smarandache theory; proportional conflict redistribution; leakage monitoring; critical decisions; uncertainty.

DOI: 10.1504/IJPSE.2014.070084

International Journal of Process Systems Engineering, 2014 Vol.2 No.4, pp.291 - 302

Received: 12 Aug 2014
Accepted: 25 Dec 2014

Published online: 26 Jun 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article