Safety performance measurement framework for offshore oil and gas platforms in Malaysia / Daniel Tang Kuok Ho

Daniel Tang , Kuok Ho (2018) Safety performance measurement framework for offshore oil and gas platforms in Malaysia / Daniel Tang Kuok Ho. PhD thesis, University of Malaya.

	PDF (The Candidate's Agreement) Restricted to Repository staff only Download (1694Kb)
Preview	PDF (Thesis PhD) Download (3236Kb) | Preview

Abstract

The Macondo blowout in the Gulf of Mexico on the 20th April 2010 which caused 11 fatalities and numerous serious injuries turned attention again to safety of offshore oil and gas activities. Findings of the US Chemical Safety Hazard Investigation Board pointed to a lack of focus on process safety addressing major accident hazards, over-reliance on lagging indicators and overemphasis on personal safety. The findings also highlighted the use of both leading and lagging safety indicators has great potential in major accident prevention. Safety management on offshore installations is divided into process and personal safety, leading to fragmented safety performance measurement which overemphasizes on personal safety and the lagging aspects of process safety. From analysis of offshore accident data, the EU Commission recommended pooling of data to provide well-rounded picture of offshore safety, inclusion of near misses in accident databases, and common formatting to facilitate data and experience sharing. This study presents a comprehensive safety performance measurement framework for offshore oil and gas platforms in Malaysia which combines both leading and lagging safety indicators to monitor major aspects of process and personal safety. It identifies 70 leading and lagging safety indicators grouped under 14 safety factors most pertinent to offshore oil and gas platforms via literature review and inputs of industrial practitioners. It stages an integrative approach to unify the relevant offshore safety indicators from past studies and systematically apply them for performance measurement. The first phase of the study involved compilation of a list of indicators and development of questionnaire to gauge the perception of safety and health practitioners in 10 major oil and gas companies in Malaysia on the importance of the indicators and the perceived risk of failing to observe the indicators. The second phase involved statistical analyses of the survey data to yield descriptive statistics of the indicators, hence the safety factors, as well as the correlations between the safety factors demonstrated via factor analyses, hierarchical clustering and Pearson correlation. Weights of the safety indicators were also derived in this phase. The third phase of the study centered on development and validation of the safety performance framework. The framework consists of two components, i.e. a scoring system to generate the scores of the respective safety factors, hence the overall safety score of an offshore installation, as well as a fuzzy inference system to generate a composite safety performance index based on scores of the safety factors and the rules established by safety experts. The framework functions to pull safety data together and presents them in a common format which is responsive to experience gained, emergence of new indicators and changes in performance targets and standards. An alternative architecture of the fuzzy inference system with intermediate models of correlated safety factors was also proposed to simplify rule-setting of fuzzy inference system. The framework was finally validated against facility status reports and actual lagging performance of offshore platforms. The validation demonstrated reliability and applicability of the framework for offshore safety performance measurement, reporting and benchmarking. The findings showed the ability of the framework to highlight major contributors of offshore incidents, demonstrate interactions between safety factors, monitor well-being of platform’s safety management system and reveal causation of physical safety system failure.

Actions (For repository staff only : Login required)