Date of Graduation

8-2024

Document Type

Thesis

Degree Name

Master of Science in Civil Engineering (MSCE)

Degree Level

Graduate

Department

Civil Engineering

Advisor/Mentor

Hernandez, Sarah V.

Committee Member

Mitra, Suman K.

Second Committee Member

Poddar, Subhadipto

Keywords

Freight movement data; Machine learning; Marine Automatic Identification Systems (AIS); Marine transportation; Pattern recognition; Waterways data

Abstract

Efficient management of inland waterways is essential for the economic and operational efficiency of transportation networks. Characterization and prediction of waterway vessel traffic flow patterns by time of day are critical for optimizing planned disruptive events like maintenance activities. This study identifies and predicts inland waterway traffic flow patterns along the Lower Mississippi River (LMR) using a modified clustering approach. A five-year period of Automatic Identification System (AIS) data, which tracks vessel movements in real-time, is used for model development and evaluation. The model first segments the river into approximately one-mile-long traffic message channels (TMCs) to estimate vessel counts and daily traffic patterns, represented as hourly volumes over a 24-hr period. Unique vessels in the TMC are counted towards the TMC volume. Then, daily traffic patterns of selected TMC are clustered using a modified clustering algorithm into groups to identify common patterns. Finally, using weighted averaging, common patterns are combined using weights based on the prediction month to estimate a time-of-day pattern and traffic volume for a future period. Data from 2018 to 2021 was used for model development and validation, and data from 2022 was used for model evaluation. The modified clustering approach enhances estimation accuracy, reducing the Mean Absolute Percentage Error (MAPE) from 34.6 (Mean) and 34.7% (Median) in baseline models to 30.6% and 29.8% with K-Means and DBSCAN, respectively. Moreover, the direction of vessel movements (upstream or downstream) had minimal impact on MAPE, suggesting that combined traffic data does not compromise prediction accuracy. Insights into time-of-day vessel traffic patterns aid in the optimization of maritime operations, enabling decision-makers to strategically plan waterway closures by determining the best timing and duration based on traffic patterns.

Download

Share

COinS