liner_shipping — Global liner shipping networks (2015-2018)

Description

Networks of the world's container ports connected by regular liner shipping services, constructed by M. Xu and collaborators from the database of liner shipping service routes of Alphaliner (a commercial database of the container shipping industry, whose raw data are not public). Nodes are ports, and two ports are connected by an undirected edge if at least one liner service route calls at both of them: each service route, a fixed sequence of ports of call served by container ships, becomes a clique of its ports, and the network is the union of these cliques.

The entry contains the versions of the network published by the authors with their articles:

• 2015: the network of April 2015 of Xu et al. (Nature Communications 2020), from all service routes, with 977 ports and 16,680 edges (the source data of Fig. 1d of the article). The edges are unweighted.
• 2015_international and 2017_international: the networks of 2015 and 2017 of Xu et al. (Royal Society Open Science 2020), from the international service routes (1,316 routes in 2015), with 777 and 770 ports. The ports are identified only by numbers (which differ between the two years) and their countries. The edges have six weights, computed as the sum over the service routes that call at both ports of either 1, $1/(n-1)$, $1/[n(n-1)/2]$, $C$, $C/(n-1)$ or $C/[n(n-1)/2]$, where $n$ is the number of ports of the route and $C$ its total traffic capacity, in twenty-foot equivalent units (TEU).
• 2018: the network of 2018 of Xu et al. (Global Networks 2024), with 887 ports, weighted edges and the great-circle distances between the ports. In the data, each edge is listed in both directions, with the same values; it is included once.

The networks are not comparable in every respect, since they were constructed from different selections of the routes; the port names of the 2015 and 2018 networks are as given (e.g. Antwerp, Port Chalmers).

The networks have the following properties:

• name (2015, 2018): the name of the port.
• port_id (2015_international, 2017_international): the identifier of the port in the data.
• country (2015_international, 2017_international, 2018): the ISO 3166-1 alpha-3 code of the country of the port.
• country_name (2018): the name of the country of the port.
• Edge routes, weight_1n, weight_2nn, capacity, capacity_1n, capacity_2nn (2015_international, 2017_international): the six edge weights, the sums over the shared routes of 1 (the number of shared routes), $1/(n-1)$, $1/[n(n-1)/2]$, $C$, $C/(n-1)$ and $C/[n(n-1)/2]$, respectively.
• Edge weight (2018): the weight of the link as given by the authors (derived from the traffic capacities of the shared routes).
• Edge distance (2018): the great-circle distance between the two ports, in kilometers.

Tags

Transportation Economic Trade Weighted Metadata

Citation

• M. Xu, Q. Pan, A. Muscoloni, H. Xia, C. V. Cannistraci, "Modular gateway-ness connectivity and structural core organization in maritime network science." Nature Communications 11, 2849 (2020) 10.1038/s41467-020-16619-5 Sci-Hub
• M. Xu, Q. Pan, H. Xia, N. Masuda, "Estimating international trade status of countries from global liner shipping networks." Royal Society Open Science 7, 200386 (2020) 10.1098/rsos.200386 Sci-Hub
• M. Xu, Y. Zhu, W. Deng, Y. Shen, T. Li, "Assessing the efficiency and vulnerability of global liner shipping network." Global Networks 24(1), e12445 (2024) 10.1111/glob.12445 Sci-Hub

Upstream URL OK

https://github.com/Network-Maritime-Complexity/Structural-core
https://doi.org/10.5281/zenodo.4018584
https://github.com/Network-Maritime-Complexity/GLSN-efficiency-vulnerability

Upstream license

CC BY 4.0 (2015_international, 2017_international), MIT (2018)

Networks

Tip: click on a column header to sort the list, or hover over it for a description.

Column legend

$\left<k\right>$

Mean degree

$\sigma_k$

Standard deviation of the degrees

$\lambda_h$

Largest eigenvalue of the Hashimoto (non-backtracking) matrix

$\tau$

Random walk mixing time

$r$

Degree assortativity

$c$

Global clustering coefficient

$\oslash$

(Pseudo-) diameter

$S$

Relative size of the largest component

NPs

Node properties

EPs

Edge properties

Show all columns

Name	Nodes	Edges	$\left<k\right>$	$\sigma_k$	$\lambda_h$	$\tau$	$r$	$c$	$\oslash$	$S$	Kind	Mode	NPs	EPs	gt	GraphML	GML	csv
2015	977	16,680	34.15	41.03	84.34	14.73	-0.02	0.35	5	1.00	Undirected	Unipartite	name		37 KiB	73 KiB	72 KiB	61 KiB
2015_international	777	12,000	30.89	37.12	75.73	15.08	-0.03	0.36	5	1.00	Undirected	Unipartite	port_id country	routes weight_1n weight_2nn capacity capacity_1n capacity_2nn	153 KiB	190 KiB	180 KiB	194 KiB
2017_international	770	11,780	30.60	38.24	76.67	7.09	-0.06	0.35	6	1.00	Undirected	Unipartite	port_id country	routes weight_1n weight_2nn capacity capacity_1n capacity_2nn	166 KiB	196 KiB	185 KiB	211 KiB
2018	887	14,052	31.68	39.01	79.74	27.07	-0.03	0.36	6	1.00	Undirected	Unipartite	name country country_name	weight distance	164 KiB	226 KiB	222 KiB	213 KiB

Ridiculograms^*

^* These are automatically generated force-directed visualizations, and can be quite meaningless for networks both big and small. They should not be taken seriously as sources of scientific insight. See here for a discussion.