Marine Engine Running Hour Prediction Using AIS Data Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for collecting engine running hours data on marine vessels are prone to errors and inaccuracies, leading to unreliable predictions, which can result in inefficient maintenance and potential malfunctions.
Innovation Solution
A computer-implemented method that combines data from multiple sources, including engine installation information, customer asset management, and vessel identification and movement data, to generate a dynamic engine running hour model, which cleans and enriches the data to improve accuracy and reliability of predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If sensor-based data collection methods are used for engine running hours, then data collection is simple and cost-effective, but the data accuracy and reliability deteriorate due to errors and inaccuracies
Solution Approach 1:
The patent combines multiple data sources including sensor data, AIS vessel movement data, and customer asset management information to create a comprehensive running hours estimation model. This merging of diverse data sources compensates for the inaccuracies of individual sensor-based measurements while maintaining practical data collection methods.
Solution Approach 2:
The patent introduces an intermediary estimation model that processes and reconciles data from multiple sources. This model acts as a mediator between raw sensor data and final running hours predictions, filtering errors and combining information to produce accurate results without requiring direct reliance on potentially flawed sensor measurements.
2Reliability
If multiple data sources are combined to improve prediction accuracy, then running hours prediction reliability improves, but system complexity increases due to data integration requirements
Solution Approach 1:
The patent creates a multi-functional system that handles multiple data types (sensor data, AIS data, maintenance records) through a single integrated estimation model. This universal approach allows the system to process diverse inputs without requiring separate complex subsystems for each data source, reducing overall system complexity while maintaining high reliability.
Solution Approach 2:
The patent transforms multiple different data parameters into a unified running hours estimation framework. By changing the parameters of various data sources into a common format and using weighted combinations, the system integrates diverse information without requiring complex custom processing for each data type.
3Measurement precision
If manual data cleaning and verification processes are implemented, then data quality improves, but processing time and operational complexity increase
Solution Approach 1:
The patent implements self-service data cleaning through automated algorithms that detect and correct inconsistencies in the integrated data. The system automatically identifies conflicting information from multiple sources and resolves discrepancies without requiring manual intervention, maintaining high data quality while minimizing processing time and operational complexity.
Solution Approach 2:
The patent performs preliminary data validation and cleaning during the data integration process itself, before final estimation calculations. By preprocessing and validating data as it is collected and combined, the system ensures high data quality without requiring separate time-consuming manual verification steps later in the workflow.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The present invention relates to a computer-implemented method, a computing device and a computer program product for predicting running hour information of marine vessel engine. The method comprises receiving engine installation information, receiving engine running hour information from an engine running hours storage system, receiving vessel identification and movement data and generating a dynamic engine running hour model by combining the engine installation information, the engine running hour information and the vessel identification and movement data. Predicted running hour information is predicted based on the dynamic engine running hour model.