dc.contributor.author	Hussin, Nor Afina Eidura
dc.date.accessioned	2024-05-19T07:07:53Z
dc.date.available	2024-05-19T07:07:53Z
dc.date.issued	2018
dc.description	Thesis (PhD. (Chemical Engineering))
dc.description.abstract	Currently, process safety accidents caused by process equipment failures have kept on happening in the chemical process industry due to lack of learning from accidents, comprehensive accident data, safety studies on learning from accidents and knowledge transfer, and experience and knowledge management-based loss prevention tools. Most available tools focused on risk assessment, utilized quantitative/hybrid analyses, and neglected safety experience and accident knowledge from the workforce. The industry less emphasized on engineering and management errors, hazard identification and safety management tools, and qualitative/semi-quantitative analyses. In the study, learning from accidents in the industry is improved by integrating safety experience and accident knowledge into a safety management method. About 1,337 accident contributors and 2,014 safety recommendations were identified from 468 accident reports (1990-early 2014) retrieved from reliable accident databases. The accidents involved piping system, storage tank, process vessel, reactor, heat transfer equipment and separation equipment failures. Data analysis was carried out on accident contributors, safety actions, and their interconnection and suitability. From knowledge management, data, information, and knowledge generated from the analysis and literature were transformed into the experience-based method. Based on the analysis, design errors (45%) were the highest contributors, followed by human (23%), technical (22%), organizational (9%) and external factors (1%). About 67% of the contributors were originated from design phases while 33% of them were from operation phases. The most recommended actions by the industry were procedural (56%), followed by inherent (18%), active (14%) and passive (12%) safety strategies. However, the interconnection analysis showed that the industry mostly preferred procedural safety (70%) but least preferred inherent safety (10%). To enable suitability analysis of the recommended actions against contributor classification, a suitability logic matrix was developed. Using the logic matrix, about 50% of the actions were considered unsuitable solutions. The developed systematic method comprises of process equipment selection, hazard identification of engineering and management aspects, problem solving, audit/monitoring, decision making and record keeping. Problem solving and decision making are determined by type, ranking, critical criteria, and origin of the contributors. For contributors originated from design phases, selection of the actions depends on reliability of process safety hierarchy; while for contributors originated from operation phases, the selection depends on the suitability logic matrix. The method was applied to the BP Texas City and the BP Deepwater Horizon case studies, and reported the average prediction accuracy of 89%. In conclusion, the established method is able to fulfill the gaps in loss prevention tools, improve learning from accidents and indirectly reduce accident rates in the industry
dc.description.sponsorship	Faculty of Chemical Engineering
dc.identifier.uri	http://openscience.utm.my/handle/123456789/1161
dc.language.iso	en
dc.publisher	Universiti Teknologi Malaysia
dc.subject	Chemical processes—Safety measures
dc.subject	Industrial safety—Databases
dc.subject	Industrial safety—Research
dc.title	An experience-based systematic method for process safety management in the chemical process industry
dc.type	Thesis
dc.type	Dataset

An experience-based systematic method for process safety management in the chemical process industry

Files

Original bundle

License bundle

Collections