Data-driven decision-making practice in response with drawworks maintenance notifications

Abstract

Offshore installations are complex and need to be maintained properly to keep expected performance. Critical failures on these installations might induce great threats on productivity, personnel safety, and environment. A maintenance strategy that combines corrective, preventive and predictive maintenance practices will be suggested to achieve reliability as well as cost-efficiency. During the operation and maintenance (O&M) activities, much data is collected, and it has great potential values to help understanding the condition of offshore facilities, and to help making reliable decisions. The thesis is designed to suggest maintenance decision-making practices that incorporate all data collected, analyzed and accumulated from O&M activities, failure histories and other data sources. The methodology used in the thesis is suggested by the author. Drawworks is selected as an example to explain the idea of achieving the target. The research will start from identification of the most critical failure modes of drawworks. This will be done in several ways at the same time to ensure most failure modes are included in the discussion. Then qualitative (Fault tree analysis) and quantitative analysis (reliability analysis, assignment of Monitoring Priority Number) will be implemented. The results from these analyses will provide some reference of risk criticality of potential failures. With the risk analysis results and data integrity management, comprehensive and straightforward data architecture could be built in purpose of providing the right data to the right person at the right place. In technical integrity management context, competence management, decision support system and integrated work process will also be studied to help identifying necessary and critical elements in a reliable and efficient decision making practice on maintenance notifications.