A modified method for the safety factor parameter: The use of big data to improve petroleum pipeline reliability assessment

A modified method for the safety factor parameter: The use of big data to improve petroleum pipeline reliability assessment

Petroleum industry

Pipelines

Gasoline

Decision making

Big data

Principal component analysis

Safety factor

Accident prevention

Correlation methods

Reliability analysis

Due to the potential severity of oil and gas pipeline accidents, accurate assessments on the reliability and viability of pipelines in the petroleum industry is of paramount importance. Nevertheless, the safety factor (SF) parameter in some well-established assessment standards are limited in their applications. This paper proposes a modified method for the SF parameter to better assess petroleum pipeline reliability. The proposed method improves upon current methods in that the SF is derived from multiple critical factors based on pipeline big data rather than the calculation of only the pressure of the pipeline. Data from an in-service pipeline is used as a case study to demonstrate how the proposed modified SF parameter is calculated. Comparative analysis with the existing method's results provide clear evidence that the proposed modification method is more accurate as it shows how the SF parameter changes according to different regional levels. This modified method, which incorporates Correlation Analysis, Mutual Information Principal Component Analysis (MIPCA), and Weighted Aggregated Sum Product Assessment (WASPAS), is in accordance with the widely accepted American Society of Mechanical Engineers (ASME) Manual for Determining the Remaining Strength of Corroded Pipelines (B31G-2012). With that said, the effectiveness of our modified method is directly related to the factors and case-based values being used. Therefore, although generally applicable to any pipeline, any form of SF analytics must be on a case-by-case basis. 2020 Elsevier Ltd

198

Zhang, Hewei

Dong, Shaohua

Ling, Jiatong

Zhang, Laibin

Cheang, Brenda

Reliability Engineering and System Safety

2020

journalArticle

9518320

10.1016/j.ress.2020.106892

http://dx.doi.org/10.1016/j.ress.2020.106892