Maritime AI & Data Foundations Series : Part 1/4. What Is the SFI Code — And Why Every Maritime AI & Data Career Starts Here
Everyone in maritime is talking about AI. Predictive maintenance. Voyage optimization. Smart ship platforms. Autonomous vessels. The maritime AI sector has seen rapid investment growth across fleet management, port automation, and smart ship platforms.
But here is the problem nobody talks about: AI can't fix bad data. And most ship data is badly structured. Before you build a machine learning model, before you design a dashboard, before you write a single line of Python for a maritime use case — you need to understand how ship data is organized at its foundation. That foundation is the SFI Code.
Section Ⅰ — What Is the SFI Code?
SFI stands for Skipsteknisk Forskningsinstitutt — the Ship Research Institute of Norway, now part of SINTEF Ocean (formerly MARINTEK). Originally developed in 1972, the SFI Group System is the most widely-used classification standard for organizing technical information on ships, offshore platforms, and maritime structures worldwide.
Think of it as the taxonomy of a ship — a structured numbering system that gives every system, subsystem, and component onboard a unique, hierarchical identity.
Section Ⅱ — How the SFI Code Structure Works
The SFI system uses a hierarchical numerical code organized into four levels:
Section Ⅲ — Why SFI Is the Foundation of Maritime Data & AI
A modern vessel can have 200+ Computer-Based Systems (CBS) onboard — from engine monitoring to navigation, automation to safety systems. Each stores data differently, with different naming conventions, units, and update frequencies.
"I spent years in maritime cybersecurity analyzing ship systems under IACS UR E26/E27. One of the most consistent failure points I encounter during vessel assessments is asset inventory — ships that cannot produce an accurate, up-to-date list of their own CBS. The reason is almost always the same: no standardized data classification. SFI is the answer to that problem. If you want to work in maritime data or AI, learn SFI before you learn TensorFlow."
Section Ⅳ — SFI vs. Other Classification Systems
Section Ⅴ — SFI in Smart Ships & IACS UR E26/E27
IACS UR E26 mandates CBS (Computer-Based System) inventory management: every vessel must maintain an accurate, up-to-date record of all onboard digital systems. While IACS UR E26 does not explicitly prescribe SFI as the required format, SFI codes have become the widely adopted industry structure for building and maintaining compliant CBS inventories.
Section Ⅵ — Career Paths Using SFI Knowledge (Near Future)
Build data pipelines that ingest, normalize, and classify ship sensor data using SFI as the master taxonomy. Leading maritime technology companies all work with SFI-structured data as their foundation.
Train machine learning models on SFI-tagged maintenance records and sensor data to predict equipment failures before they happen. One of the most in-demand roles in ship management right now.
Use SFI-structured data to compare fuel consumption, maintenance frequency, and component reliability across vessels in a fleet — surfacing optimization opportunities that would be invisible without standardized data.
Map CBS inventory using SFI-based classification for IACS UR E26 compliance. This is a specialized role with very few qualified candidates globally — high demand, premium compensation.
Build or contribute to CMMS, ERP, or vessel intelligence platforms that rely on SFI as their underlying data structure. Every maritime software company needs engineers who understand both code and ship systems.
Section Ⅶ — How to Start Learning SFI
Know what falls under each top-level code. This gives you the mental map of how a ship's data is organized before diving into sub-levels.
If you have access to AMOS or a similar CMMS through your institution or internship, browse the equipment tree. Every item has an SFI code attached — see the classification in action.
Practice by taking a simplified ship equipment list and assigning SFI codes to each item. This exercise builds the classification instinct that's essential for maritime data work.
Once you understand the SFI structure, ISO 14224 gives you the failure mode taxonomy — essential for any predictive maintenance AI project. The SFI + ISO 14224 combination is particularly powerful.
The CBS inventory requirements in UR E26 are pushing the entire industry toward SFI-structured data. Understanding both positions you at the intersection of compliance and technology — a rare and valuable combination.
The SFI Code is not just a legacy maintenance standard from 1972 — it is the universal data spine of the smart ship era. Before you train a model, build a dashboard, or audit a CBS inventory, you need to understand how ship data is classified at its foundation.
IACS · Mandatory for newbuilds contracted from July 2024 · Requires CBS inventory — SFI widely used as the implementation structure · iacs.org.uk
Originally developed 1972 · Most widely used maritime equipment classification standard · sintef.no
ISO · 2016 · Provides failure mode & maintenance data taxonomy — used alongside SFI for reliability analysis · iso.org
IMO · 2017 · Integrated into ISM Code from January 2021 · imo.org
ShipPaulJobs · Download IACS, BIMCO, NIST, IEC 62443 compliance PDFs · shippauljobs.com
Comments
Post a Comment