From Ocean Visibility to Ocean Predictability

We are all familiar with the recent Suez Canal incident involving the ship Ever Given causing havoc in multiple ways. Hundreds of ships to stuck in a tailback, vessels being rerouted to avoid the Suez Canal, adding around eight days to their total journeys.

Data from Lloyd’s List showed the stranded ship was holding up an estimated $9.6bn of trade along the waterway each day. That equates to $400 million and 3.3 million tonnes of cargo an hour, or $6.7 million a minute.

All this, caused by one single incident.

Now imagine the ongoing costs accrued to operational inefficiency, poor data visibility and vessels hitting ports later than their scheduled ETAs.

As the ocean becomes more and more crowded, supply chains get disrupted, and (when something like the Pandemic hits, hic!), it is up to maritime management to ensure that all vessels are safe, that they arrive on time, and that disruptions are managed as best as possible.

This can be a difficult task considering there are so many vessels and cargo in the water at any given time, businesses depending on crucial parts and/or products, and more.

Thankfully, using data allows you to monitor vessels, streamline operations, and manage supply chains better.

The three-legged foundation for efficient data management is this:

1. Visibility (access to real-time data)

2. Analytics (processing data to help support key decisions)

3. Predictions (using AI and Machine Learning to make predictions that help streamline logistics, maritime management, and to strengthen supply chain operations).

Visibility data is the stepping stone for predictability (e.g. data about vessel locations, container movements, port and terminal events). When you have all this data in place you can start adding data scientists and machine learning to come up with predictions on what will happen in the future.

The result? A better experience for all.

We have been working for many years to create our maritime data foundation. How does 273 billion data points + 150m new data points everyday sound? Thanks to all that data, we have the basics to build predictions.

Here are some use cases and examples of using machine learning for the maritime industry. For instance, we can estimate predictive ETA (estimated time of arrival) for container vessels at ports based on our predictive models.

What’s more exciting, you ask? We have a roadmap in place and we are working hard to build several predictive models.

Here’s a sneak peak (of what we are working on):

– Predictive Gate Out: When do we expect the container to be ready for pickup. Taking custom clearance, congestion, etc into consideration.

– Predictive Transhipment Miss: Is your container going to miss the transhipment? When is the transhipment due, where is the container now (which vessel) and where is the vessel is it going on)?

While accidents are inevitable sometimes, and no amount of planning, data insights, or predictive analysis can help, most other operational wrinkles can be ironed out.

Data and predictive analysis is gaining ground, as we write this: By 2023, 50% of leading global enterprises plan to invest in real-time transportation visibility solutions, according to The Gartner Market Guide for Real-Time Transportation Visibility Platforms.

project44, for instance, helps increase operations agility, minimize costs and wait times, get live info about disruptions with predictive tracking and analytics.

GlobalNewsWire recently reported that FourKites was awarded a patent for Smart Forecasted Arrival Engine for tracking and managing freight, transportation, trucking operations.

Using predictive analysis helps with better planning, proactive actions, sharper decision-making, insight-driven SOPs, and countless hours saved due to less manual work. 

This leads to big gains for cargo owners, transport companies, maritime management operators, and large businesses that depend on stable supply chain management.