Xin, X (2023) MODELLING AND SYSTEMATIC EVALUATION OF MARITIME TRAFFIC SITUATION IN COMPLEX WATERS. Doctoral thesis, Liverpool John Moores Univerisity.

Preview

Text 2023xurixinphd.pdf - Published Version Available under License Creative Commons Attribution Non-commercial. Download (10MB) | Preview

Abstract

Maritime Situational Awareness (MSA) plays a vital role in the development of intelligent transportation support systems. The surge in maritime traffic, combined with increasing vessel sizes and speeds, has intensified the complexity and risk of maritime traffic. This escalation presents a considerable challenge to the current systems and tools dedicated to maritime traffic monitoring and management. Meanwhile, the existing literature on advanced MSA methods and techniques is relatively limited, especially when it comes to addressing multi-ship interactions that may involve hybrid traffic of manned ships and emerging autonomous ships in complex and restricted waters in the future. The primary research question revolves around the challenge faced by current collision risk models in incorporating the impact of traffic characteristics in complex waters. This limitation hampers their effectiveness in managing complex maritime traffic situations.
In view of this, the research aims to investigate and analyse the traffic characteristics in complex port waters and develop a set of advanced MSA methods and models in a holistic manner, so as to enhance maritime traffic situation perception capabilities and strengthen decision-making on anti-collision risk control. This study starts with probabilistic conflict detection by incorporating the dynamics and uncertainty that may be involved in ship movements. Then, the conflict criticality and spatial distance indicators are used together to partition the regional ship traffic into several compact, scalable, and interpretable clusters from both static and dynamic perspectives. On this basis, a systematic multi-scale collision risk approach is newly proposed to estimate the collision risk of a given traffic scenario from different spatial scales. The novelty of this research lies not only in the development of new modelling techniques on MSA that have never been done by using various advanced techniques (e.g., Monte Carlo simulation, image processing techniques, graph-based clustering techniques, complex network theory, and fuzzy clustering iterative method) but also in the consideration of the impact of traffic characteristics in complex waters, such as multi-dependent conflicts, restricted water topography, and dynamic and uncertain ship motion behaviours.
Extensive numerical experiments based on real AIS data in the world's busiest and most complex water area (i.e., Ningbo_Zhoushan Port, China) are carried out to evaluate the models’ performance. The research results show that the proposed models have rational and reliable performance in detecting potential collision danger under an uncertain environment, identifying high-risk traffic clusters, offering a complete comprehension of a traffic situation, and supporting strategic maritime safety management. These developed techniques and models provide useful insights and valuable implications for maritime practitioners on traffic surveillance and management, benefiting the safety and efficiency enhancement of maritime transportation. The research can also be tailored for a wide range of applications given its generalization ability in tackling various traffic scenarios in complex waters. It is believed that this work would make significant contributions in terms of 1) improving traffic safety management from an operational perspective without high financial requirements on infrastructure updating and 2) effectively supporting intelligent maritime surveillance and serving as a theoretical basis of promoting maritime safety management for the complex traffic of mixed manned and autonomous ships.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	Maritime safety; Conflict detection; Graph partitioning; Multi-scale collision risk; AIS data
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD61 Risk Management T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Engineering
SWORD Depositor:	A Symplectic
Date Deposited:	20 Nov 2023 14:34
Last Modified:	20 Nov 2023 14:38
DOI or ID number:	10.24377/LJMU.t.00021761
Supervisors:	Yang, Z, Loughney, S, Wang, J and Liu, K
URI:	https://researchonline.ljmu.ac.uk/id/eprint/21761

View Item