Analyzing Global Port Congestion

Live Dashboard

Having trouble interacting with the embed? Click here to open the full dashboard in its own page →

KPIs

Project overview

Port congestion directly affects delivery timelines and operational efficiency, but the root causes (vessel size, fleet characteristics, or regional inefficiency) are often unclear. Without that visibility, it's hard to know where to intervene to improve throughput. This project examines the relationship between vessel capacity, fleet age, and port dwell time across global maritime operations to identify which factors actually drive delay.

Methodology

Findings

• Vessel capacity turned out to be the strongest driver, by far. Larger vessels correlate with longer dwell time (correlation ≈ 0.54), strong enough that I stopped chasing other explanations first.
• Region mattered too, more than I expected. Congestion varies widely port to port, which points to real differences in infrastructure and operational efficiency rather than one global cause.
• Vessel age showed up, but as a distant second. Once I accounted for capacity, age only explained a small amount of what was left.
• Delay didn't scale evenly. It jumped at the high end. That pattern reads like ports hitting a real capacity ceiling, not just a straight line getting steeper.

Recommendations

I'd prioritize congestion-reduction work at the ports handling the largest vessels first. Capacity is the dominant lever here, not age or any single regional factor. I'd also use the regional variation as a benchmarking tool: ports with better dwell-time performance at a similar vessel-capacity level are the ones worth studying for what they're doing right.

What this demonstrates

• Correlation-driven root-cause analysis on real operational data, not just descriptive reporting.
• A full Python → SQL → Tableau pipeline built end to end.
• Translating a global, multi-variable problem into a small number of clear, actionable drivers.