top of page

AI Case Study

CargoMetrics analyses satellite shipping imagery and plans to to identify investment opportunities based on the data with machine learning

CargoMetrics is an investment firm which uses satellite imagery to collect shipping data and is analyzed using proprietary algorithms in combination with other data to make commodities, equity index futures, and currencies trading decisions. It is building a machine learning system to identify investment decisions based on its various data sources.



Freight And Logistics

Project Overview

"CargoMetrics is building a “learning machine” that will be able to automatically profit from spotting any publicly traded security that is mispriced, using what he refers to as systematic fundamental macro strategies. CargoMetrics’ analytic models help find asset classes that are outliers. Those may include a publicly traded instrument such as oil, another commodity or an equity for which shipping information was a leading indicator during times when other asset classes marched in lockstep. The historical ship data is then blended with this new information to seek opportunities.

CargoMetrics is building a systematic approach that will work even when cargo cannot be identified — on containerships, for instance. It already knows a large percentage of the daily imports and exports into and out of China and island economies such as Japan and Australia. And although the firm cannot glean from its calculations on satellite AIS data the type of cargo, such as iPhones from China, it can measure total flow, which shows present economic activity. Cargo­Metrics’ data scientists are working on linking such activity to the firm’s data set of the past seven years to measure the evolving global economy. "

Reported Results

Planned; results not yet available


In-house prediction algorithms, unsure of machine learning specifics for prediction and cargo detection.
"Using satellite data with hundreds of millions of ship positions, CargoMetrics makes trillions of calculations to determine individual cargoes onboard the ships and then to aggregate the cargo flows and compare them with historical shipping data. All that leads to the final comparisons with historical financial market data to find mispricings. If CargoMetrics observes an appreciable decline in export shipping activity in South Africa, for example, its trading models will determine whether that is a significant early-warning sign by considering that information alongside other factors, such as interest rates. If Cargo­Metrics believes a decline in the rand is forthcoming, it might short it against a basket of other currencies. To make sure CargoMetrics’ algorithms for identifying cargo are valid, the firm spot-checks manifest data filed at ports and imposes statistical confidence checks to guard against spurious correlation."



Data Science


"CargoMetrics, a start-up investment firm, is not your typical money manager or hedge fund. It was originally set up to supply information on cargo shipping to commodities traders, among others. Now it links satellite signals, historical shipping data and proprietary analytics for its own trading in commodities, currencies and equity index futures.

CargoMetrics was one of the first maritime data analytics companies to seize the potential of the global Automatic Identification System. Ships transmit AIS signals via very high frequency (VHF) radio to receiver devices on other ships or land. Since 2004, large vessels with gross tonnage of 300 or more are required to flash AIS positioning signals every few seconds to avoid collisions. That allows Cargo­Metrics to pay satellite companies for access to the signals gleaned from 500 miles above the water. The firm uses historical data to identify cargo and aggregation of cargo flow, and then applies sophisticated analysis of financial market correlations to identify buying and selling opportunities."



satellite imagery, historical shipping data

bottom of page