The Descartes Labs Platform Creates a Digital Twin of the Brazilian Iron Ore Supply Chain
The global iron ore market experienced significant disruption in early 2019 in the weeks following a major tailings dam failure near Brumadinho, Brazil. In order to mitigate the risk of additional dam failures, the Brazilian government halted all mining operations at mines with similar upstream tailings dams. Since Brazil represents approximately 25% of the global iron ore market share, the operational halt caused a major rebalancing in global iron ore flows.
The Descartes Labs Platform helped a key iron ore transport company determine when Brazilian mining activity would come back online and estimate when flows would return to normal. With our platform, the company was able to incorporate computer vision data, satellite imagery and AIS shipping traces to develop a digital twin of the affected mines and nearby delivery facilities that mapped and measured commodity flows and timing before, during and after the event.
With a multi-modal digital twin of the iron ore supply chain at each mine and port, the company captured critical insights into future commodity flows with computer vision model accuracy above 90%. This enabled them to respond effectively to a shorter-than-expected supply disruption.
Featured Video – Reservoir Modeling
Art of the Possible
With the Descartes Labs Platform, business and data science teams can rapidly hypothesis, test, and deploy predictive analytics. These examples showcase some of our customers’ success, but the possibilities are endless.
Datasets: Automatic Identification System (AIS) shipping traces
Summary: Descartes Labs uses AIS data to track ship movement and characterize ship status. We clean and fuse multiple AIS datasets together to create global a snapshot of dry bulk carriers classified by navigation status at any point in time. This data can be used to track the volume of transported commodities between ports across the world.
Quantify multimodal transport
Datasets: High resolution imagery, internal customer data
Summary: Goods awaiting shipment represent a demand factor for transport. With the Descartes Labs Platform, imagery and radar data can characterize ports, railyards and storage fields all around the globe. These data and the resulting analytics help our customers improve the quantification of market supply and demand beyond their own operations.
Estimate reservoir levels
USGS/NASA Landsat Program 2020
Summary: Companies with large physical networks, from mine operations to inland water transportation systems, rely on water as a critical resource. But low reservoir levels and lack of runoff can cause unanticipated challenges. Data scientists have used the Descartes Labs Platform to model reservoir water storage and predict water flows weeks in advance.
Datasets: Sentinel-1 and Sentinel-2, Airbus SPOT 6 and 7, Airbus Pleiades
Summary:The number and class of ships under construction today will impact the shipping market over the next 18 – 24 months. Using the Descartes Labs Platform, data scientists created a computer vision model over key shipyards to determine the ship type, current construction stage and percent complete for each vessel under construction.
Datasets: Sentinel-1 and Sentinel-2, Airbus SPOT 6 and 7, Airbus Pleiades, Automatic Identification System (AIS) shipping traces
Summary:Users of the Descartes Labs Platform can easily combine different data types in a single analysis. In one example, a data scientist located all dry bulk ports in the world using computer vision and AIS shipping data. The model identified dry bulk ships and bulk commodity storage piles to find a candidate set of locations, which were then verified by arrival and departure information from AIS data.
Customers include shipping and logistics companies, commodity producers and service providers that deal with all aspects of global logistics: