INTEGRATED EFFICIENCY AND PRODUCTIVITY ANALYSIS OF CONTAINER TERMINALS: A COMPREHENSIVE FRAMEWORK USING DEA, SBM, MPI, AND PREDICTIVE MODELING

Abstract

Container terminals are critical nodes in the global supply chain, connecting maritime and inland transport systems. This study evaluates the efficiency and productivity of eight Malaysian container terminals over a 15-year period (2003–2018) using an integrated framework of Data Envelopment Analysis (DEA), Slack-Based Model (SBM), Malmquist Productivity Index (MPI), and Machine Learning (ML) techniques. It benchmarks performance, diagnoses inefficiencies, tracks productivity trends, and predicts future efficiency. DEA and SBM reveal disparities, with high-performing terminals near the efficiency frontier and underperformers showing resource slack and throughput shortfalls. MPI highlights the role of innovation in driving long-term competitiveness, while predictive modeling using ML provides actionable insights for proactive planning. This study bridges traditional efficiency analysis with modern predictive tools, offering recommendations to optimize terminal operations and sustain competitiveness.