Subscribe Button 1
SUBSCRIBE

Unlocking the Value of Data at the Edge

With more artificial intelligence and machine learning data created at the edge than ever before, businesses are increasing their reliance on some serious computational power, storage and analytics to do the heavy lifting as it’s processed and put into the hands of their users in real time. As a result, users can remotely manage, administer and gain insight on their IT infrastructure wherever it’s deployed. The emphasis on extracting value from data at the edge and using it to improve IT outcomes, drive transformation and increase overall business agility has never been greater.

Dell Technologies has announced a series of ways its solution set and partnerships are helping businesses get more value from their data with real-time analytics at the edge. One such example is their work with Duos Technologies Group, Inc.

Through a partnership, Dell are providing the foundation for Duos to deploy innovative artificial intelligence solutions at the edge for remote, automated railcar inspections. This first-of-its-kind technology, used by Class I railroads including CSX, designed to increase safety, enhance maintenance accuracy and reduce customers’ overall costs.

“Our goal at Duos Technologies is to transform industries through intelligent solutions, bringing machine learning, artificial intelligence and video analytics together – the infrastructure behind these solutions is absolutely critical to our business,” said Charles Ferry, chief executive officer at Duos Technologies. “Dell Technologies provides an essential combination of solutions, support and services that, at the end of the day, help our rail customers unlock the value of data at the edge. Together, we’re able to not only keep trains moving, we’re getting them there faster and safer.”

Hundreds of trains cross North American rail lines every day, moving approximately $700 billion worth of imports, exports and domestic shipments in 2020 alone. The traditional inspection process takes an average of eight minutes per railcar, resulting in unscheduled stops that cost rail companies millions of dollars annually. Duos Technologies’ Railcar Inspection Portal, or rip, is able to review more than 120 railcars in the same amount of time – all while the trains are moving at full speed.

The rip Railcar Inspection Portal is a modular intelligent visualization system that provides real-time detailed 360-degree imaging at high speeds. The rip can be used for a variety of inspection criteria on mainlines or in yards. The included Linear Panorama Generator assembles images gathered from cameras and stiches all frames to create a continuous view of the entire consist. Operators can quickly select the side of interest and scroll through the continuous panoramic view. This also provides operators with a geo-special view of the train.

Automated Equipment Inventory (AEI) consist data is the primary methodology used to synchronize captured images of each railcar. The system is searchable using the AEI tag number or sequence in the train. Duostech high definition Imaging utilizes megapixel line scan cameras to provide and average image resolution of 224 megapixels per railcar. Images are crisp and highly detailed. A range of features and options are available offering maximum flexibility and scalability for end users. Machine learning algorithms (AI) can also be incorporated to provide an even higher level of maintenance inspection.

The Railcar Inspection Portals cover more than 140,000 miles of rail track throughout North America. Each portal, placed 50 to 100 miles from an inspection point, contains a micro-data center, designed together with Dell Technologies OEM Solutions, and stacked with Dell EMC PowerEdge servers , specifically designed to endure remote and tough conditions, handle the demands of image capturing. Additional servers with multiple NVIDIA GPUs provide the lightning-speed throughput needed to process and analyze data in microseconds.

For more information: www.delltechnologies.com

HOME PAGE LINK