In an era of digital transformation, one might assume that basic information about ships their size, type, and even total fleet numbers would be standardized and consistent. However, the reality reveals a surprising "identity crisis" in maritime data that challenges industry stakeholders and technology providers alike. That’s what we learned from Rory Proud, co-founder of Maritime Data, in his presentation called “Not all data is created equally” at OrbitMI FLOW.
Consider this startling fact: when asked to count merchant vessels above 100 gross tons, three leading maritime data providers recently gave vastly different answers ranging from 57,000 to 172,000 ships. This dramatic disparity raises a fundamental question: how can an industry digitize effectively when it can't agree on basic facts?
The problem extends beyond mere counting. Even vessel classification presents significant challenges. In one notable example, a single vessel was classified as a "tanker/floating production storage and offloading unit" by one provider and as a "ferry" by another. Such discrepancies aren't mere academic concerns they can significantly impact business decisions, risk assessments, and regulatory compliance efforts.
The roots of this identity crisis lie in multiple factors:
The implications of these inconsistencies are far-reaching. For commodity traders trying to track fleet capacity, or compliance officers monitoring vessel movements, these discrepancies can lead to significant blind spots or errors. A 100,000 DWT tanker misclassified as a ferry doesn't just represent a data error it could mean missing crucial market capacity or compliance risks.
Making matters more complex, vessel data isn't static. Ships can change roles, owners, or classifications throughout their lifecycle. Without robust audit trails, these changes can create confusion and contradictions across different data sets. Even something as seemingly straightforward as a port call can have multiple interpretations one recent study of vessels calling at Jeddah found 17 different ways to spell the port's name in AIS messages.
The industry's challenge now is not just to digitize, but to standardize. As one expert noted during the conference, "If we can't agree on what ship is called what, how can we build reliable AI systems on top of this data?"
Indeed. That’s why OrbitMI has doubled-down on our commitment to addressing data quality in maritime. In addition to working with partners such as Maritime Data, we source only the highest quality data feeds in Orbit.
To watch Rory Proud's full presentation from OrbitMI Flow 2024, click here.
