Christian Velasco-Gallego

The maritime sector is undergoing an unprecedented digital transformation driven by rapid advances in artificial intelligence (AI), data analytics, cyber–physical systems, and edge computing

When accidents occur, official investigations are carried out, and reports are generated, which are usually reviewed for safety improvements. The retrieval of information is typically performed manually, which can lead to biases, errors, and poor judgement. Moreover, manual reviews can be tedious and highly time-consuming tasks. For these reasons, the implementation of LLMs has been analysed in this context. However, previous studies have been limited, and no proper justification for the implemented LLMs has been provided. Consequently, this work proposes a comparative framework to assess LLM candidates across two main dimensions: cybersecurity and performance. Specifically, a total of 9 LLMs from different providers were analysed, and 18 prompt injection techniques were implemented across 7 categories based on OWASP LLM01:2025 and previous academic studies. Additionally, a RAG system based on these results is introduced to validate the potential of these models in supporting experts in the retrieval of information from maritime accident reports. For validation purposes, a case study on the Marine Accident Investigation Branch (MAIB) reports was conducted. Results show that a comparative framework is required, as model selection may vary depending on the task being performed, which is critical from both a performance and cybersecurity perspective.

Quantum Machine Learning (QML) is emerging as a promising technology for tackling complex computational challenges, although its practical implementation faces significant obstacles owed to the inherent noise in current quantum devices. This paper presents a hybrid architecture that combines classical and quantum elements for the development and training of QML models under noisy conditions. This research evaluates the impact of noise on superconducting systems through emulation. The study shows that, compared to noise-free configurations, certain noise levels tend to allow for a reduction in the number of qubits, thus simplifying the architecture of the quantum neural network, which has a direct impact on the computational cost and execution times. Experimental validation was performed by applying three biomedical datasets related to breast cancer detection. The experimental findings revealed that the variations in accuracy between noiseless configurations and those subjected to noisy conditions were minimal, with deviations ranging from 0.11% to 1.68%. Additionally, it was observed that the incorporation of noise during training contributed positively to the efficiency of the process for the datasets under test, achieving improvements in training execution times ranging from a factor of 1.61 to 4.39, when the proposed architecture was emulated on Qaptiva 802.

The lack of fault data is still a major concern in the area of smart maintenance, as these data are required to perform an adequate diagnostics and prognostics of the system. In some instances, fault data are adequately collected, even though the fault labels are missing. Accordingly, the development of methodologies that generate these missing fault labels is required. In this study, Markov-CVAELabeller is introduced in an attempt to address the lack of fault label challenge. Markov-CVAELabeller comprises three main phases: (1) image encoding through the application of the first-order Markov chain, (2) latent space representation through the consideration of a convolutional variational autoencoder (CVAE), and (3) clustering analysis through the implementation of k-means. Additionally, to evaluate the accuracy of the method, a convolutional neural network (CNN) is considered as part of the fault classification task. A case study is also presented to highlight the performance of the method. Specifically, a hydraulic test rig is considered to assess its condition as part of the fault diagnosis framework. Results indicate the promising applications that this type of methods can facilitate, as the average accuracy presented in this study was 97%.

The application of artificial intelligence models for the fault diagnosis of marine machinery increased expeditiously within the shipping industry. This relates to the effectiveness of artificial intelligence in capturing fault patterns in marine systems that are becoming more complex and where the application of traditional methods is becoming unfeasible. However, despite these advances, the lack of fault labelling data is still a major concern due to confidentiality issues, and lack of appropriate data, for instance. In this study, a method based on histogram similarity and hierarchical clustering is proposed as an attempt to label the distinct anomalies and faults that occur in the dataset so that supervised learning can then be implemented. To validate the proposed methodology, a case study on a main engine of a tanker vessel is considered. The results indicate that the method can be a preliminary option to classify and label distinct types of faults and anomalies that may appear in the dataset, as the model achieved an accuracy of approximately 95% for the case study presented.

Additive manufacturing (AM) has emerged as a transformative technology in the fabrication of intricate structures, offering unparalleled adaptability in crafting complex geometries. Particularly noteworthy is its burgeoning significance within the realm of medical prosthetics, owing to its capacity to seamlessly replicate anatomical forms utilizing biocompatible materials. Notably, the fabrication of porous architectures stands as a cornerstone in orthopaedic prosthetic development and bone tissue engineering. Porous constructs crafted via AM exhibit meticulously adjustable pore dimensions, shapes, and porosity levels, thus rendering AM indispensable in their production. This systematic review ventures to furnish a comprehensive examination of extant research endeavours centred on the generation of porous scaffolds through additive manufacturing modalities. Its primary aim is to delineate variances among distinct techniques, materials, and structural typologies employed, with the overarching objective of scrutinizing the cutting-edge methodologies in engineering self-supported stochastic printable porous frameworks via AM, specifically for bone scaffold fabrication. Findings show that most of the structures analysed correspond to lattice structures. However, there is a strong tendency to use organic structures generated by mathematical models and printed using powder bed fusion techniques. However, no work has been found that proposes a self-supporting design for organic structures.

Climate change has highlighted the need to promote renewable energies. The offshore wind industry is growing exponentially because of some political strategies supported by various organizations, such as the European Union. The implementation of these strategies is commonly associated with significant investments, public acceptance, or achieving better installations and greater cumulative capacities. To ensure that offshore renewable energy projects could reach their ambitious targets, this study promotes the implementation of political strategies or planning decisions using data mining techniques and analytical tools. Strategic decisions based on real data analysis could help to achieve more suitable and optimal infrastructures. The scour phenomenon jeopardizes the operability of offshore wind farms, making it necessary to study its evolution over the years. In this work, extensive research on the scour phenomenon in offshore wind farms using real data (from the Lynn and Inner Dowsing offshore wind farms located in the UK) was performed, which revealed an evident lack of consideration of this phenomenon for data-driven decision-making processes. As a novelty, this research develops a detailed geospatial analysis of data, studying the possible autocorrelation of scour data measured from each turbine between 2011 and 2015. The conclusions obtained could be used to improve future planning tasks in offshore wind farms.