From Dock to Destination: Toward an End-to-End Simulation Study

Abstract

This study uses a discrete-event simulation model, built with open-source software, to analyze import container flows at the Port of New York/New Jersey. The model integrates input and parameter distributions derived from extensive data analysis of publicly available import records, enhanced by machine learning techniques, including Natural Language Processing for commodity classification using unstructured shipping manifest product descriptions. Initial results demonstrate the effectiveness of Gaussian Kernel Density Estimate (KDE) and Fourier models in representing container dwell times, reducing mean absolute error compared to normal distribution by up to 39.5% for dry containers and 24.8% for reefers. A fine-tuned BERT model achieves over 80% accuracy in commodity classification to the four-digit HS code level, enabling improved input data structuring for simulation. Initial scenario testing indicates increasing outbound rail freight from 15% to 25% of total containers reduces truck congestion by 11.5% and decreases median dwell time by 1.52% for dry containers and 2.55% for reefers. These findings highlight the potential for logistical adjustments to improve efficiency and reduce congestion at the Port of New York/New Jersey.