Engineering

Philosophy

Transport engineering

Biology

Identification (biology)

Spots

DBSCAN

Canopy clustering algorithm

Data mining

Computer science

Botany

Geology

Computer network

Physical chemistry

Paleontology

Accident (philosophy)

Operations research

Segmentation

Epistemology

Stage (stratigraphy)

Black spot

Artificial intelligence

Horticulture

Cluster analysis

Chemistry

Correlation clustering

Resource (disambiguation)

Reliability Engineering & System Safety

Inland shipping plays a significant role in the integrated transport system. Maritime safety has been one of the top concerns due to its high-risk characteristics. The historical accident data is treated as a valuable source for identifying the riskiest waters (also called black-spots) where special attention is necessary. In view of this, a two-stage black-spot identification model is proposed in this paper to identify and locate waterways with higher accident rates. In stage 1, the dynamic segmentation and equivalent accident number methods are proposed to identify the preliminarily black-spots. In stage 2, Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is introduced to pinpoint the precise locations of the detailed black-spots based on the results from the first step. The model is further applied to the Jiangsu section of the Yangtze River based on the historical accident data between 2012 and 2016. The results show that altogether 12 preliminary black-spots and 5 detailed black-spots are identified in the investigated waters. This research provides helpful reference for optimizing the allocations of search and rescue resource as well as differentiated safety management of black-spot waters.

A two-stage black-spot identification model for inland waterway transportation