• Avoidance of navigation accident costs and consequences (average of 500-1000 ships per year totalling hundreds to billions dollars per year of insurance claims)

DHM provides the lowest life costs for the asset and at the same time improving its availability (readiness to do paid work) and utility (doing paid work). 

• Anomaly detection – is it working normally? If there is abnormality should the crew investigate? Promotes pro-active troubleshooting of suspected problems.
• Fault Detection – If there is persistent abnormality, what is exactly wrong? How bad is it? Prompt fault disambiguation and isolation including maintenance recommendations.
• Diagnostics – if there are concurrent faults, how bad are they? What is the consequence at equipment, system, vessel level? Generates multiple maintenance recommendations and operational advisories.
• Prognostics (Predictive) – Given the current fault progression, how much time before failure? Allows opportunistic maintenance, availability-utility trade-offs, continuous optimisation of the logistics - servicing/parts supply chain.

If the AI cannot provide complete coverage (e.g. incomplete understanding of the healthy states, but also failures state-space), can it be complemented by remote human experts? Allows AI-SME teaming with online help from remote experts for fault isolation, consequence analyses walkthroughs including resolution, repair advisories to the crew.

Cost savings relevant to test man-hours, logistics of testing tools in addition to minimising the loss of asset availability, utility during testing and inspection events.