Affiliation:
1. National Research Tomsk Polytechnic University
2. National Research Tomsk Polytechnic University; Tomsk State University of Architecture and Building
Abstract
Purpose: The aim of this work is to analyze competing risks of engineering system failure using nonparametric prediction.Methodology: Two statistical methods are used to tackle the problem of the pipeline integrity. Nonparametric prediction is applied in the first one. The main attention is paid to the possible failure of the pipeline section due to a specific threat in terms of competing risks. The second method contains the analysis of the entire pipeline system. The focus is on rupture incidents, that provides the real data on the lifecycle of the pipeline section.Research findings: Nonparametric prediction is used to analyze competing risks of the pipeline failure. The lower and upper probability boundaries and survivability functions and competing risks of the pipeline section failure are introduced for a future ground section of the pipeline, which can be failed as a result of rupture.
Publisher
Tomsk State University of Architecture and Building