Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Spatial Clustering Approach for Vessel Path Identification

Mohamed Abuella, M. Amine Atoui, Slawomir Nowaczyk, Simon Johansson, Ethan Faghan

TL;DR

The paper tackles vessel path identification from position data by introducing a spatial clustering framework with two complementary approaches: a distance-based method using average nearest neighbor distance (ANND) and a segmented Gaussian likelihood method. Through a real-world case study of Cinderella II in the Stockholm archipelago, the authors demonstrate that distance-based hierarchical clustering and k-means/GMM can classify voyages into five path classes with high accuracy, while the segmented Gaussian approach achieves perfect precision and recall across classes. The results show robustness to noise via ANND, interpretability of path similarity, and valuable segment-level insights into route alterations, supporting improved route planning and safety in maritime transportation. The framework emphasizes data-driven, scalable insights for route optimization, with clear indications of scalability considerations and future work to extend applicability and integration with maritime systems.

Abstract

This paper addresses the challenge of identifying the paths for vessels with operating routes of repetitive paths, partially repetitive paths, and new paths. We propose a spatial clustering approach for labeling the vessel paths by using only position information. We develop a path clustering framework employing two methods: a distance-based path modeling and a likelihood estimation method. The former enhances the accuracy of path clustering through the integration of unsupervised machine learning techniques, while the latter focuses on likelihood-based path modeling and introduces segmentation for a more detailed analysis. The result findings highlight the superior performance and efficiency of the developed approach, as both methods for clustering vessel paths into five classes achieve a perfect F1-score. The approach aims to offer valuable insights for route planning, ultimately contributing to improving safety and efficiency in maritime transportation.

Spatial Clustering Approach for Vessel Path Identification

TL;DR

The paper tackles vessel path identification from position data by introducing a spatial clustering framework with two complementary approaches: a distance-based method using average nearest neighbor distance (ANND) and a segmented Gaussian likelihood method. Through a real-world case study of Cinderella II in the Stockholm archipelago, the authors demonstrate that distance-based hierarchical clustering and k-means/GMM can classify voyages into five path classes with high accuracy, while the segmented Gaussian approach achieves perfect precision and recall across classes. The results show robustness to noise via ANND, interpretability of path similarity, and valuable segment-level insights into route alterations, supporting improved route planning and safety in maritime transportation. The framework emphasizes data-driven, scalable insights for route optimization, with clear indications of scalability considerations and future work to extend applicability and integration with maritime systems.

Abstract

This paper addresses the challenge of identifying the paths for vessels with operating routes of repetitive paths, partially repetitive paths, and new paths. We propose a spatial clustering approach for labeling the vessel paths by using only position information. We develop a path clustering framework employing two methods: a distance-based path modeling and a likelihood estimation method. The former enhances the accuracy of path clustering through the integration of unsupervised machine learning techniques, while the latter focuses on likelihood-based path modeling and introduces segmentation for a more detailed analysis. The result findings highlight the superior performance and efficiency of the developed approach, as both methods for clustering vessel paths into five classes achieve a perfect F1-score. The approach aims to offer valuable insights for route planning, ultimately contributing to improving safety and efficiency in maritime transportation.
Paper Structure (15 sections, 8 equations, 14 figures, 2 tables)

This paper contains 15 sections, 8 equations, 14 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Background and Motivation
  3. Aims and Contributions
  4. Related Work
  5. Methodology
  6. Problem Formulation
  7. Distance-Based Method
  8. Segmented Gaussian Likelihood Method
  9. Case Study
  10. Data Collection
  11. Data Preparation and Analysis
  12. Results and Discussion
  13. Evaluation of Path Clustering
  14. Discussion
  15. Conclusion

Figures (14)

  • Figure 1: Framework of vessel path identification. (a) Flowchart of distance-based method. (b) Flowchart of segmented Gaussian likelihood method.
  • Figure 2: Part of the distance matrix, in this case showing 12 paths.
  • Figure 3: Image of the Cinderella II ship marinetrafffic_Cind.
  • Figure 4: Map of vessel route.
  • Figure 5: The vessel route before and after applying path identification
  • ...and 9 more figures