Vietnam Maritime University

Abstract Nondestructive Testing (NDT) is an essential methodology for ensuring the structural integrity and safety of critical engineering structures such as nuclear power plans, railways, and aerospaces. Leveraging advanced signal analysis and Artificial Intelligence (AI) is necessary to enhance detection capabilities, but the development of bespoke AI models presents a significant bottleneck, requiring deep, cross-disciplinary expertise, and extensive manual effort. To address this challenge, we introduce the AI Co-Scientist, a novel framework designed for human-AI cooperation throughout the entire model development lifecycle. Within this human-in-the-loop system, AI agents and human experts work in partnership to perform data analysis, iterative AI model development, and rigorous model evaluation.The effectiveness of this collaborative framework was rigorously verified by applying it to a real-world Gaussian-Pulsed Eddy Current Testing (GPECT) system for the task of detecting corrosion in an aluminum specimen. The framework autonomously performs signal analysis and converges on a high-performance model. The proposed approach achieved a recall of 99.6%, an F1-score of 95.1%, and an AUC-ROC of 99.9% on a highly imbalanced real-world GPECT dataset, demonstrating its suitability for safety-critical corrosion detection. This result validates our approach, demonstrating that the AI Co-Scientist framework can significantly accelerate research and empower domain experts to focus on high-level strategic oversight.

TThis paper presents a comparative study between the three-element tuned mass damper (T-TMD) and the conventional tuned mass damper (TMD) for vibration mitigation of offshore platforms subjected to ground excitations. The T-TMD is an enhanced configuration of the TMD, in which a viscous damper is connected in series with an additional spring. The equations of motion of offshore platforms equipped with the T-TMD are derived using the Lagrangian method. The constraint conditions of the T-TMD are analyzed to formulate the optimization problem, in which two objective functions are considered: maximizing vibration suppression efficiency and minimizing the device mass. The T-TMD parameters are optimized for a specific offshore platform based on these objectives. Numerical simulations demonstrate that the T-TMD provides superior performance and enhanced robustness compared to a conventional TMD of equal mass.

Purpose The purpose of this study is to examine how servant leadership influences firm performance through the sequential mediating roles of dynamic capabilities and frugal innovation, with environmental turbulence as a moderating factor. Design/methodology/approach This study used a two-stage partial least squares structural equation modelling (PLS-SEM) approach to analyse data collected from 448 Vietnamese small and medium-sized enterprises (SMEs). Findings The results show that servant leadership significantly enhances dynamic capabilities, which in turn foster frugal innovation and lead to improved firm performance. Furthermore, dynamic capabilities partially mediate the relationship between servant leadership and frugal innovation, while frugal innovation partially mediates the link between dynamic capabilities and firm performance. The moderating effect of environmental turbulence also amplifies the impact of dynamic capabilities on performance. Research limitations/implications This study provides practical guidance for SMEs in dynamic, resource-constrained environments by emphasizing servant leadership, dynamic capabilities and frugal innovation as key drivers of adaptability and performance. It encourages leadership development and a culture of learning to enhance strategic agility and innovation under uncertainty. Originality/value This study advances theory by introducing a novel serial mediation model and examining environmental turbulence as a moderator – an approach not previously applied to Vietnamese SMEs. It extends existing research by clarifying how servant leadership fosters performance through capability-building and resource-efficient innovation.

This study investigates emissions and pollution control strategies at Hai Phong Port, a key container terminal in Vietnam. Using a bottom-up methodology and AIS data, it estimates emissions during the hoteling phase from 2018 to 2022, focusing on pollutants like NOx, CO2, SO2, PM10, PM2.5, CO, and HC. The findings show that CO2 and NOx dominate emissions, while SO2 and particulate matter have declined significantly since the implementation of IMO MARPOL Annex VI standards in 2020. Increasing vessel sizes and improved cargo handling efficiency have reduced emissions per TEU, despite rising throughput and ship calls. The study explores links between vessel size, cargo handling productivity, and emissions, recommending strategies such as optimizing vessel size, reducing berthing times, redistributing ship calls, and adopting low-emission technologies.

The paper evaluates the stress-strain relationships of different structural element groups of ship structures subjected to longitudinal bending moments at the ultimate limit state (ULS). The incremental-iterative method proposed by the International Association of Classification Societies (IACS) is applied. In this method, the hull cross-section is divided into three groups of structural elements, each assigned to one of six common failure modes. Bulk carriers (BCs) are employed to illustrate the calculation results. Accordingly, the paper presents the stress-strain curves of the different structural element groups, as well as the stress and strain experienced by each structural element at the ULS under hogging and sagging conditions. One notable finding is that the sagging condition generates higher stresses in the ship structure compared to the hogging condition. The top-side tank region experiences the highest stress levels, reaching the material’s yield stress when the ship approaches its ultimate longitudinal bending capacity. Additional discussions are also provided in this paper.