In a significant stride towards greener and more efficient maritime operations, researchers have developed a novel framework that combines artificial intelligence (AI) and blockchain technology to predict and optimize ship fuel oil consumption. This innovative approach, detailed in a recent study published in the journal ‘Applied Ocean Research’ (translated from English), promises to revolutionize energy management in the shipping industry.
At the heart of this research is a decentralized, data-driven analytical framework that leverages federated learning (FL) and artificial neural networks (ANN) to predict CO2 emissions and categorize ships based on their energy consumption features. The lead author, Mihir Parekh from the Department of Computer Science and Engineering at the Institute of Technology, Nirma University in Ahmedabad, India, explains, “We first employed a regression model to predict CO2 emissions in ships. Based on the prediction, we created target labels in the dataset, i.e., ship with poor engine (1) and ship with good engine (0).”
The framework’s uniqueness lies in its use of blockchain technology to ensure the integrity and confidentiality of the predicted data. A smart contract was developed to guarantee that only valid FL-trained weights are shared with FL clients and the global model. This tamper-proof technology confronts data tampering attacks, a critical concern in the maritime industry where data integrity is paramount.
The proposed framework demonstrated impressive performance metrics, including a training accuracy of 98.74%, a training loss of 0.094, and a regression error rate ranging from approximately 24.15% to 32.12%. The blockchain’s transaction and execution cost were also evaluated, ranging from approximately 50,000 to 260,000 units.
The commercial impacts of this research are substantial. By optimizing energy consumption, shipping companies can significantly reduce fuel costs, which constitute a major portion of operational expenses. Moreover, improved energy efficiency translates to lower CO2 emissions, aligning with the industry’s increasing focus on sustainability and regulatory compliance.
The framework’s decentralized nature also offers opportunities for collaboration among shipping companies. By sharing data and insights without compromising confidentiality, companies can collectively enhance their predictive capabilities and drive industry-wide improvements in energy efficiency.
In the words of Parekh, “The synergy of AI and blockchain highlights their combined impact on revolutionizing energy consumption prediction in the maritime industry.” This research not only refines predictive accuracy but also ensures the confidentiality and integrity of the predicted data, paving the way for a more efficient and sustainable maritime sector.