International Port Dependencies and Resilience to Supply Chain Disruptions

The ability to identify and analyze spatiotemporal relationships is critical for cargo vessel shipments, understanding supply chain dynamics, diagnosing port congestion, and making effective recommendations. The spatiotemporal dependencies between different international shipping partners is an important aspect in understanding port supply and demand. The main limitations of the existing analytical methods in transportation research are that they fail to capture directional or nonlinear dependencies. The current techniques for evaluating ports rely on statistical relationships of individual ports and not the dependencies among shipping and receiving ports. Here, the research team proposes using two novel causal discovery algorithms for the first time for extracting the spatiotemporal dependencies of operational performances between international ports using large volumes of real-world port traffic datasets. The team propose a data-driven approach to analyze the cargo ship network, representing a complex system defined as the network of ports that are connected by links of vessel travel paths. This research will further identify the most central ports in the network and determine groups of highly interconnected ports. Next, the team will develop an alternative functional network of ports in which a directed link (or edge) connects two ports to capture the directional dependency between their activities. In particular, the team will use information-theoretic approaches, such as transfer entropy on time series data to quantify the directional relationships. The transfer entropy approach is model-free and has an advantage over simple cross-correlation which is restricted to linear and non-directional assumptions. The team will finally analyze the properties of the functional network using graph-theoretic measures. Specifically, the team will compute ‘centrality measures’ to identify the influential ports and ‘global measures’ to analyze the dynamics and detect the common critical properties. In other words, the goal is to obtain a high-level description of the structure of the system through the analysis of low-level relational data, where the high-level description identifies various kinds of patterns in the set of relationships. This work will find applications in resolving supply-chain related issues in the cargo ship network.


    • English


    • Status: Completed
    • Funding: $133378
    • Contract Numbers:


    • Sponsor Organizations:

      Office of the Assistant Secretary for Research and Technology

      University Transportation Centers Program
      Department of Transportation
      Washington, DC  United States  20590
    • Managing Organizations:

      Maritime Transportation Research and Education Center

      University of Arkansas
      Fayetteville, AR  United States  72701
    • Performing Organizations:

      Louisiana State University

      3660G Patrick F. Taylor Hall
      Civil and Environmental Engineering
      Baton Rouge, LA  United States  70803
    • Principal Investigators:

      Wolshon, Brian

    • Start Date: 20220401
    • Expected Completion Date: 20230930
    • Actual Completion Date: 20230930
    • USDOT Program: University Transportation Centers

    Subject/Index Terms

    Filing Info

    • Accession Number: 01841198
    • Record Type: Research project
    • Source Agency: Maritime Transportation Research and Education Center
    • Contract Numbers: 69A3551747130
    • Files: UTC, RIP
    • Created Date: Mar 30 2022 9:46PM