Tidal analysis and arrival process mining using Automatic Identification System (AIS) data
This work presents a method for extracting vessel arrival times and arrival processes from Automatic Identification System (AIS) data. This work employs the methodology presented by Mitchell and Scully (2014) for inferring tidal elevation at the time of vessel movement and calculating the tidal dependence (TD) parameter to 23 U.S. port areas for the years 2012–2014. Tidal prediction stations and observation reference lines are catalogued for considered ports. AIS data obtained from the U.S. Coast Guard, and 6-minute tide predictions, obtained from the National Oceanographic and Atmospheric Administration, are used to rank relative tidal dependence for arriving cargo and tank vessel traffic in studied locations. Results include relevant tide range and elevation threshold observations for each year and location studied. AIS-derived arrival processes, including arrival frequency, arrival rate, and interarrival time are visualized using several techniques with comparative discussion between ports to highlight implications for understanding seasonal traffic trends or port resiliency. The ports with the highest and lowest TD value, Portland, ME, and Los Angeles, CA, respectively, are discussed with regard to weekly arrival patterns and interarrival time. Cargo composition and value obtained through the Channel Portfolio Tool is also considered.
Coastal and Hydraulics Laboratory (U.S.)Engineer Research and Development Center (U.S.)
Channels; Data processing; Dredging--Planning; Harbors; Inland navigation; Remote sensing; Ships--Automatic identification systems; Tides
Technical Report (Engineer Research and Development Center (U.S.)) ; no. ERDC/CHL TR-17-2
Approved for Public Release; Distribution is Unlimited