Answers are grounded in this paper and cite their sources.
The increasing availability of Automatic Identification System (AIS) data provides unprecedented opportunities for the fine-grained analysis of maritime traffic zones in complex port waters. However, conventional zone delineation approaches, relying on purely attribute-based statistics or predefined management divisions, struggle to ensure both traffic attribute homogeneity and spatial continuity. This study proposes a data-driven framework for delineating Spatially Homogeneous Traffic Zones (SHTZs). A dual-layer graph is constructed, in which the attribute layer captures multi-dimensional traffic similarities through a hybrid local-window and k -Nearest Neighbour sparsification strategy, while the spatial layer encourages spatial continuity via an eight-neighbour adjacency matrix. Constrained spectral clustering is then applied to jointly optimise attribute consistency and spatial continuity, followed by a spatial refinement procedure that guarantees zone connectivity and an adaptive multi-stage hyperparameter optimisation strategy that ensures solution stability. Experiments using AIS data from Ningbo-Zhoushan Port show that the framework delineates zones with homogeneous traffic characteristics and continuous spatial structures, capturing key traffic patterns and revealing areas with elevated potential conflict risk. Compared with five graph-based and spatially constrained baselines, the proposed approach achieves the best balance between attribute homogeneity and spatial continuity. The framework provides quantitative support for fine-grained traffic pattern analysis and intelligent maritime traffic management.
No takes yet. Share an insight, caveat, or question.
Xin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: