System and method for vessel risk assessment
Provided are systems and methods for vessel risk assessment. This includes determining a risk assessment associated with a vessel, including receiving vessel data from at least one source, generating at least one vessel profile based on the vessel data, wherein each vessel profile provides indication of expected behavior events for one vessel and abnormal behavior events for one vessel, determining at least one abnormal behavior event of the vessel based on the at least one vessel profile, each event in the at least one abnormal behavior event having a time of occurrence, determining at least one frequency of occurrence of abnormal behavior events of the vessel based on the time of occurrence of each event, using at least one model to determine a risk assessment associated with the vessel based on the at least one frequency of occurrence of abnormal behavior events of the vessel.
1. A computer-implemented method for providing a risk assessment user interface associated with a vessel, the method comprising:
receiving, at a processor, vessel data from at least one source, the vessel data comprising vessel tracking data received from a vessel tracking device associated with the vessel, the vessel tracking data comprising a plurality of Automatic Identification System (AIS) messages, the vessel tracking device comprising an AIS transceiver associated with the vessel, the at least one source comprising at least one vessel tracking system;
generating, at the processor, at least one vessel profile based on the vessel data, wherein each vessel profile provides indication of expected behavior events for one vessel and abnormal behavior events for one vessel;
determining, at the processor, at least one abnormal behavior event of the vessel based on the at least one vessel profile, each event in the at least one abnormal behavior event having a time of occurrence;
determining, at the processor, at least one frequency of occurrence of abnormal behavior events of the vessel based on the time of occurrence of each abnormal behavior event of the vessel;
using at least one machine-learning model to determine a risk assessment associated with the vessel based on the at least one frequency of occurrence of abnormal behavior events of the vessel, a first machine-learning model in the at least one machine-learning model selected from the group of a Naïve Bayesian model, a linear regression model, a multiple class classifier, and a Neural Network;
outputting a user interface comprising a map, a vessel icon positioned on the map based on the vessel tracking data, and the risk assessment associated with the vessel, wherein the map, the vessel icon and the risk assessment are updated automatically based on the updated vessel data.
2. The method of claim 1 wherein the at least one source includes at least one selected from the group of a vessel incidents information source, a vessel information source, a vessel tracking data source, a regional boundaries source, and a crime-related activity source.
3. The method of claim 2 further comprising:
determining, at the processor, enhanced vessel tracking data based on the vessel data; and
tagging the enhanced vessel tracking data based on vessel identification.
4. The method of claim 3 , wherein the determining the enhanced vessel tracking data comprises merging the vessel data with region boundary data.
5. The method of claim 4 wherein the at least one vessel profile comprises statistical information for a first vessel for a first geographical region and wherein the determination of a first abnormal event of the first vessel for the first geographical region is based on the statistical information.
6. The method of claim 5 wherein an alarm is generated based on the determination of the first abnormal event.
7. The method of claim 6 , wherein the at least one abnormal behavior event is determined based on the at least one vessel profile and a behavior data of the vessel for at least one route segment in the first geographical region.
8. The method of claim 7 , further comprising:
determining a cluster of anomalies in the at least one anomaly.
9. The method of claim 8 , wherein the determining the risk assessment comprises performing outlier detection based on the at least one abnormal behavior event and the at least one frequency of occurrence of abnormal behavior events.
10. The method of claim 1 , wherein the Neural Network is a Recurrent Neural Network.
11. The method of claim 10 , wherein:
the at least one machine-learning model comprise two or more models; and
the determining the risk assessment associated with the vessel comprises:
determining a candidate risk assessment for each of the two or more models; and
performing an election of the candidate risk assessments from the two or more models to determine the risk assessment.
12. The method of claim 11 , wherein the at least one machine-learning model is determined for a single vessel.
13. The method of claim 12 , wherein the at least one vessel profile includes at least one selected from the group of a vessel MMSI profile, a vessel name profile, a vessel destination profile, a vessel visit duration profile, a vessel trip duration profile, a vessel movement profile, a vessel speed profile, a vessel tracking transmission profile, a vessel tracking position accordance profile, a vessel sea route profile, a vessel crew size profile, a vessel incident profile and a vessel rendezvous profile.
14. The method of claim 13 , wherein the detected at least one anomaly includes at least one selected from the group of a speed anomaly, a location anomaly, a vessel tracking transmission anomaly, and a rendezvous anomaly.
15. A computer-implemented system for providing a risk assessment user interface associated with a vessel, the system comprising:
a memory, the memory comprising:
a model for determining a risk assessment;
a processor in communication with the memory, the processor configured to:
receive vessel data from at least one source, the vessel data comprising vessel tracking data received from a vessel tracking device associated with the vessel, the at least one source comprising at least one vessel tracking system, the vessel tracking data comprising a plurality of Automatic Identification System (AIS) messages, the vessel tracking device comprising an AIS transceiver associated with the vessel;
generate at least one vessel profile based on the vessel data, wherein each vessel profile provides indication of expected behavior events for one vessel and abnormal behavior events for one vessel;
determine at least one abnormal behavior event of the vessel based on the at least one vessel profile, each event in the at least one abnormal behavior event having a time of occurrence;
determine at least one frequency of occurrence of abnormal behavior events of the vessel based on the time of occurrence of each abnormal behavior event of the vessel;
use at least one machine-learning model to determine a risk assessment associated with the vessel based on the at least one frequency of occurrence of abnormal behavior events, a first machine-learning model in the at least one machine-learning model selected from the group of a Naïve Bayesian model, a linear regression model, a multiple class classifier, and a Neural Network;
output a user interface comprising a map, a vessel icon positioned on the map based on the vessel tracking data, and the risk assessment associated with the vessel, wherein the map, the vessel icon and the risk assessment are updated automatically based on the updated vessel data.
16. The system of claim 15 wherein the at least one source includes at least one selected from the group of a vessel incidents information source, a vessel information source, a vessel tracking data source, a regional boundaries source, and a crime-related activity source.
17. The system of claim 16 , wherein the processor is further configured to:
determine enhanced vessel tracking data based on the vessel data; and
tag the enhanced vessel tracking data based on vessel identification.
18. The system of claim 17 , wherein the determining the enhanced vessel tracking data comprises merging the vessel data with region boundary data.
19. The system of claim 18 , wherein the at least one vessel profile comprises statistical information for a first vessel for a first geographical region and the determination of a first abnormal event of the first vessel for the first geographical region is based on the statistical information.
20. The system of claim 19 , wherein an alarm is generated based on the determination of the first abnormal event.
21. The system of claim 20 , wherein the at least one abnormal behavior event is determined based on the at least one vessel profile and a behavior data of the vessel for at least one route segment in the first geographical region.
22. The system of claim 21 , wherein the processor is further configured to:
determine a cluster of anomalies in the at least one anomaly.
23. The system of claim 22 , wherein the determining the risk assessment comprises performing outlier detection based on the at least one abnormal behavior event and the at least one frequency of occurrence of abnormal behavior events.
24. The system of claim 15 , wherein the Neural Network is a Recurrent Neural Network.
25. The system of claim 24 , wherein the at least one machine-learning model comprises at least two models and the processor is further configured to:
perform an election of the candidate risk assessments from the two or more models to determine the risk assessment.
26. The system of claim 25 , wherein the at least one machine-learning model is for a single vessel.
27. The system of claim 26 , wherein the at least one vessel profile includes at least one selected from the group of a vessel MMSI profile, a vessel name profile, a vessel destination profile, a vessel visit duration profile, a vessel trip duration profile, a vessel movement profile, a vessel speed profile, a vessel tracking transmission profile, a vessel tracking position accordance profile, a vessel sea route profile, a vessel crew size profile, a vessel incident profile and a vessel rendezvous profile.
28. The system of claim 27 , wherein the detected at least one anomaly includes at least one selected from the group of a speed anomaly, a location anomaly, a vessel tracking transmission anomaly, and a rendezvous anomaly.