AI has the potential to deliver significant benefits to the maritime industry, but several challenges must be overcome to fully unlock its capabilities. In the next few articles in this series of twelve, we explore some of these areas in greater detail.
Some of these challenges include:
- Data suitability and maturity
- Data consent
- Trust, transparency, and user acceptance
- Regulation
- Ethical concerns
Data suitability and maturity
Data is fuel for AI algorithms. Without access to high-quality and mature data, AI systems cannot learn, adapt, provide insights, and make decisions. But one of the biggest challenges for AI platforms is the lack of suitable data to train the systems.
Consider autonomous driving as an example. If the data available to an AI model is not of high quality, the model may struggle to distinguish between similar objects, such as trees and people. According to Dr. Hideyuki Ando, Director of MTI Co. Ltd, accurate and reliable data is critical for autonomous operations. “Autonomous vessels are complex human-machine systems consisting of advanced automatic control systems, crews, and shore-based operators. It is extremely important to ensure the safety of the system during the design, development, construction, and operation of autonomous ships,” he noted.
“Data quality control is extremely important. We have implemented DQMS (Data Quality Management System) in the use of ship IoT data for engine system anomaly detection and CBM, and we believe that utilising the knowledge gained so far will be extremely important in the future implementation of autonomous ships,” he added.
This means that data must be optimised as well as the neural networks that are used to detect and identify objects. With autonomous cars, this is crucial in preventing collisions with all objects, especially pedestrians. According to Shogo Yamada, Chief Engineer and Deputy General Manager
of the Marine Group at NYK Line, data maturity in the context of autonomous shipping can be assessed through two key indicators. “First, the ability to consistently acquire clean, timely, and granular data without gaps, delays, or sensor anomalies.
Second, the strategic investment in collecting data that has long been recognised as valuable but was previously unattainable due to cost constraints.” He says that this involves investing in the necessary sensors and infrastructure to capture that data. “When these foundational elements are in place, the benefits are twofold. End-users can reliably monitor vessel performance using high-quality
data, ensuring safer and more efficient operations. Simultaneously, developers can leverage this robust data to create more sophisticated monitoring algorithms, driving advancements in autonomous technology.”
Yamada explained that in this regard, data quality becomes the cornerstone of the entire data utilisation process. Without it, the potential of autonomous systems cannot be fully realised. “High-quality, reliable data not only supports the present operational needs but also enables the future development of more advanced autonomous systems. As the industry progresses, investing in data quality will be a decisive factor in achieving true autonomy in shipping,” Yamada noted.
The other issue tied to data suitability is the lack of standardisation. In other industries there is access to large quantities of standardised data sets, enabling stakeholders to work with huge volumes of data. In the maritime sector, there is some concern that the capability of machine learning will be limited without standardised data sets.
To learn more about the challenges and opportunities in AI discussed in this article, read our thought leadership report, produced in collaboration with Lloyd’s Register titled, “Beyond the Horizon, Opportunities and Obstacles in the Maritime AI Boom” This report offers an in-depth analysis of the current state of operational AI in the shipping industry, shedding light on the latest trends, groundbreaking developments, and successful implementations. You can download a copy of the report here.
