@article{Availability:11072,
      recid = {11072},
      author = {Rajabally, E and Savage, I},
      title = {Autonomous Surface Vessels and Design for Availability},
      journal = {Conference Proceedings of EAAW},
      address = {2023-11-29},
      abstract = {Equipment faults and failures are generally tolerated  more-so in maritime than in other transportation domains  given the heightened safety-criticality of equipment  onboard air and land vehicles as compared to sea-going  vessels. As a result, ensuring availability of conventional  vessel equipment can require bespoke and complex  maintenance and repair tasks to be carried out particularly  for long-duration operations. However, a future  proliferation of Autonomous Surface Vessels (ASVs) demands  a different philosophy to designing for availability  without a conventional crew. This paper outlines the  challenges to delivering equipment availability that are  particular to ASVs, potential solutions and developmental  work required.

For long duration unattended operation,  addressing “holistic” health is the paramount challenge so  both structural and machinery related and ensuring hull,  mechanical and electrical availability. To address this  challenge, a structured, comprehensive and detailed upfront  analysis of equipment faults using a suite of techniques  and backed up by modelling and real-world data can be used  to focus the design effort around high-probability triggers  and high-severity consequences. Design efforts should then  consider mitigations across the cause-effect chain and  engineer-out single-points-of-failure whilst engineering-in  fault recoverability e.g. degraded subsystems operating in  concert with still healthy subsystems. For untrodden areas  of equipment automation, what-ifs based upon common-causes  of failures could help to plug knowledge gaps. Modern and  emerging technologies could then be better exploited to  augment condition monitoring such as IoT based sensors,  drone-based inspection, machine-learning based prognostics  and augmented reality. At the other end of the chain, the  same can be said of equipment repair and replacement, where  the full potential of onboard robots and 3D printing for  example, remains to be understood.

Turning to ASV  equipment availability and without the benefit of access to  unrestricted, unplanned human intervention, the relative  pros and cons of the three prevailing maintenance  strategies should be objectively compared to strike the  best balance between: corrective maintenance,  schedule-based/preventative maintenance, and  condition-based/predictive maintenance. The three  strategies fall on a continuum ranging from reactive to  proactive where there is tension between cost and benefit,  such as the lower expense but longer downtime for the  former case compared to the higher expense but shorter  downtime for the latter. Hence the need arises for an  evidence-based assessment weighing up of the alternatives  to target the “sweet spots” between them.},
      url = {http://library.imarest.org/record/11072},
      doi = {https://doi.org/10.24868/11072},
}