Review of synergy between machine learning and first principles models for asset integrity management

Cai, Tianxing; Fang, Jian; Daida, Sharath; Lou, Helen H.

doi:10.3389/fceng.2023.1138283

Cited by 1 publication

(1 citation statement)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to [24,28], risk is defined as the effect of uncertainty on goals or objectives. This impact can manifest as beneficial, detrimental, or a departure from the anticipated outcome.…”

Section: Fig 1 Safety and Health Injuries In The Oil And Gas Sector [17]mentioning

confidence: 99%

Improving Safety Performance in Nigeria’s Oil and Gas Industry: A Comparative Risk Model Analysis

Zacchaeus,

Ugbebor,

Agbagwa

et al. 2024

Asian J. Adv. Res. Rep.

View full text Add to dashboard Cite

This study examined a risk-based model for the enhancement of safety performance in Nigeria’s oil and gas sector. A non-probabilistic convenience sampling method was adopted to consider the selected five oil and gas companies in operations within this study area. A quantitative approach was employed using a comprehensive hazard checklist developed based on industry standards and expert insights, utilizing a structured questionnaire to collect data from industry experts, workers, and stakeholders on hazard likelihood and severity. The 392-sample size used was selected purposively to represent the total population of oil and gas workers in these companies. 392 copies of the questionnaire were administered, and 357 copies were used for the study, with a response rate of 91.07%. The analysis revealed key risk factors for oil and gas offshore operations, such as structural failure due to poor design, valve and seal failures, security threats, fire/ explosions, blowouts, and oil spills. Modelling the risk score against the likelihood showed a good model fit (R2 = 0.831), indicating that the model explains 83.1% of the variance, and the likelihood coefficient was positive and significant. Modelling the risk score against the severity also showed a reasonably good fit (R2 = 0.676), with the severity coefficient being positive and significant. However, modelling the risk score against both likelihood and severity showed excellent fit (R2 = 0.994), explaining 99.4% of the variance. The likelihood and severity coefficients were positive and highly significant; this indicated that an increase in either factor would lead to an increase in the risk score. These models provide a quantitative way to assess risk in the oil and gas industry based on the key factors of likelihood and severity of potential hazards. This study bridged theoretical risk management concepts with practical applications and provided actionable recommendations for policymakers, significantly enhancing safety performance in a crucial economic sector.

show abstract

Section: Fig 1 Safety and Health Injuries In The Oil And Gas Sector [17]mentioning

confidence: 99%

Improving Safety Performance in Nigeria’s Oil and Gas Industry: A Comparative Risk Model Analysis

Zacchaeus,

Ugbebor,

Agbagwa

et al. 2024

Asian J. Adv. Res. Rep.

View full text Add to dashboard Cite

show abstract

Review of synergy between machine learning and first principles models for asset integrity management

Cited by 1 publication

References 54 publications

Improving Safety Performance in Nigeria’s Oil and Gas Industry: A Comparative Risk Model Analysis

Improving Safety Performance in Nigeria’s Oil and Gas Industry: A Comparative Risk Model Analysis

Contact Info

Product

Resources

About