Reliability-based Assessment of Operational Modes for Dynamic Positioning Systems of Drilling Vessels

This study presents a mode-dependent reliability modelling framework for Dynamic Positioning systems in offshore drilling vessels, aimed at supporting risk-informed operational decisions between Critical Activity Mode and Task Appropriate Mode. Fault Tree Analysis (FTA) and Event Tree Analysis (ETA) are employed to model failure combinations and consequence propagation, enabling the estimation of the probability of loss of position (Plop) under different operational demands and redundancy configurations. Reliability is formulated as a function of operational mode, explicitly capturing how redundancy requirements, degradation patterns, and component availability influence station-keeping capability. Bayesian updating is incorporated to revise component reliability parameters as new in-service data become available, allowing the probability of basic events in the FTA to evolve over time without altering the logical structure of the model. This probabilistic refinement enables the system-level reliability to reflect accumulated operational evidence and changing mission conditions. The drillship case study demonstrates that the probability of loss of position increases nonlinearly with operational demand, rising by nearly 4,700 % from low to moderate demand and by approximately 1,100 % from moderate to high demand, reflecting the strong sensitivity of system-level risk to redundancy thresholds and demand intensity.

Keywords: Dynamic positioning systems, Operational modes, Reliability modelling, Fault tree analysis, Event tree analysis, Bayesian updating.