TY  - GEN
AB  - Maritime supply chain disruptions over recent years stemming from causes such as piracy, the COVID-19 pandemic, blockage of the Suez Canal and the ongoing Red Sea crisis, underscore the pressures on navies and commercial ships for higher ship operational availability. Ships are sailing longer distances, at higher speeds and in more challenging environmental conditions. These settings are consequentially increasing demands for more effective ship machinery monitoring. However, although shipboard systems generate more data than ever, using that information to improve operational availability remains elusive; data from a ship’s electrical auxiliary and main propulsion systems are often disorganized, undefined, and not timely. Further, data is sometimes undiscoverable and frequently unusable by the ship’s information system to prevent or mitigate equipment failures. Moreover, increasing demands for more sophisticated analytics to improve machine reliability, with likely thousands of data points per system without any relationship model to help interpret this data continues to make machinery monitoring efforts more complex and costly. Creating the datasets is just one piece of the puzzle. These data points take on different meanings dependent on grouping. The lack of consistent data requirements definitions and context from one information system to another introduces other challenges to integrating machinery room operational data into the ship’s higher-level information system and further up into an organization’s fleet maintenance center. To this end, this paper explores two evolving areas of technology: 1) Machine learning schema for common hull, mechanical, and electrical system machinery equipment to improve contextualizing performance anomalies and that equipment’s baseline operations and 2) an AI Information model for machinery equipment that could advance the ability of crews to reduce unplanned failures, increase availability, and obtain an accurate representation of the ship’s readiness state. These activities will drive improved reliability, maintainability, and supportability of these systems and a higher readiness for a propulsion plant, electrical plant, damage control system, and ship’s auxiliaries.
AD  - Rockwell Automation
AD  - Thor Solutions
AU  - Johnson, W
AU  - Walker, J
DA  - 2024-11-05
DO  - 10.24868/11159
DO  - doi
ID  - 11159
JF  - Conference Proceedings of iSCSS
L1  - https://library.imarest.org/record/11159/files/.pdf
L2  - https://library.imarest.org/record/11159/files/.pdf
L4  - https://library.imarest.org/record/11159/files/.pdf
LK  - https://library.imarest.org/record/11159/files/.pdf
N2  - Maritime supply chain disruptions over recent years stemming from causes such as piracy, the COVID-19 pandemic, blockage of the Suez Canal and the ongoing Red Sea crisis, underscore the pressures on navies and commercial ships for higher ship operational availability. Ships are sailing longer distances, at higher speeds and in more challenging environmental conditions. These settings are consequentially increasing demands for more effective ship machinery monitoring. However, although shipboard systems generate more data than ever, using that information to improve operational availability remains elusive; data from a ship’s electrical auxiliary and main propulsion systems are often disorganized, undefined, and not timely. Further, data is sometimes undiscoverable and frequently unusable by the ship’s information system to prevent or mitigate equipment failures. Moreover, increasing demands for more sophisticated analytics to improve machine reliability, with likely thousands of data points per system without any relationship model to help interpret this data continues to make machinery monitoring efforts more complex and costly. Creating the datasets is just one piece of the puzzle. These data points take on different meanings dependent on grouping. The lack of consistent data requirements definitions and context from one information system to another introduces other challenges to integrating machinery room operational data into the ship’s higher-level information system and further up into an organization’s fleet maintenance center. To this end, this paper explores two evolving areas of technology: 1) Machine learning schema for common hull, mechanical, and electrical system machinery equipment to improve contextualizing performance anomalies and that equipment’s baseline operations and 2) an AI Information model for machinery equipment that could advance the ability of crews to reduce unplanned failures, increase availability, and obtain an accurate representation of the ship’s readiness state. These activities will drive improved reliability, maintainability, and supportability of these systems and a higher readiness for a propulsion plant, electrical plant, damage control system, and ship’s auxiliaries.
PY  - 2024-11-05
T1  - RESILIENT: Advance a Ship's HM&E resiliency through contextual information models and innovative ML/AI analytics At-The-Edge
TI  - RESILIENT: Advance a Ship's HM&E resiliency through contextual information models and innovative ML/AI analytics At-The-Edge
UR  - https://library.imarest.org/record/11159/files/.pdf
VL  - iSCSS 2024
Y1  - 2024-11-05
ER  -