Development of Reliable Condition Monitoring Technology for Maritime Using FMECA and Bayesian Network Modeling

Abstract

Condition monitoring technique has been widely applied in Maritime to ensure safe operation and minimise unscheduled downtime. However, in practice, ship operators need to assure that a failure mode is indeed monitored by the sensor intended for it, and the sensor has sufficient accuracy and precision for its purpose. Additionally, for a reliable condition monitoring technique, issues such as sensors degradation or drift that will reduce the data quality over time must be addressed. All these require that ship owners to select a monitoring system with the best suitable sensors technology while is economically viable. In this paper, tunnel thruster was used as a case study to demonstrate the basic approach to develop a reliable condition monitoring technique through Failure Mode, Effects and Criticality Analysis (FMECA). Based on failure modes, four types of condition monitoring techniques were identified including Vibration Monitoring, Acoustic Emission Monitoring, Wear Debris /Water in Oil Monitoring, and Thermal Monitoring, where vibration monitoring is discussed in detail as an example for defining the sensor specification. For a reliable condition monitoring technique, prediction of sensor reliability will be especially useful in the situation where sensors systems can degrade over time in service. Using temperature sensors as an example, a Bayesian network (BN) modeling approach has been carried out for assessing sensor reliability affected by aging.