Measuring Port Disruptions with Automatic Identification System Data

Study efforts have focused on understanding the behavior of individual ports in the U.S. following episodic disruptions caused by hurricanes. Event driven timelines are analyzed and a method demonstrated for identifying the boundary between recovery and a return to normal operation. The role of ports within a waterway network is also explored through disruptive events with the goal of understanding the changing relative importance of in-network ports. The methods presented, relying on machine learning and big-data techniques provide a uniform, objective, scalable approach to gaging the level of port disruption and the rapidity with which services are restored. The presented methods benefit from the availability of a wide-scale datasource documenting vessel operations that can be applied at a variety of temporal and spatial scales, and is suitable for a variety of disruptive events for which ports must be prepared. Understanding the evolution of ports individually and in context with waterway network through event driven disruption timelines will ultimately inform resilience planning and recovery efforts for managers of ports and waterways.