Poland's Largest Steel Maker Uses AI, Vision to Keep Distribution on Track

ArcelorMittal Poland converted its rail car monitoring system to foster greater accuracy and productivity. Its hybrid solution combines computer vision, deep learning, and near-real-time processing and identifies rail cars throughout a factory with just two cameras per checkpoint, while its weighing system uses the Intel® Distribution of OpenVINO™ toolkit to rapidly match the cargo weight to an image of the car for efficient tracking. 

 

Article Key

Image credit: ArcelorMittal Poland

ArcelorMittal Poland, the country's largest steel producer, has more than a passing need for logistics management, transport and distribution monitoring, and automating routine processes. In addition to the railcar components and steel sheeting it manufactures, AMP is also one of the country's largest producers and exporters of coke, an industrial product used mainly in iron ore smelting.

"ArcelorMittal Poland moves large volumes of material," says Adrian Wisła, the company's digitalization officer. The types of materials that AMP ships are diverse and uniquely challenging: coal, iron ore, limestone, coke, slabs, coils, rails, profiles, sheet piles, wire rod and mining sections. Moving cargo around must also conform with the company's sustainable development policies with environmentally friendly rail transport, Wisla adds.

Materials must also be monitored throughout their journey. "For each car, a variety of data points must be tracked, such as material location, the condition of the materials, and how fast shipments are moving," Wisla explains.

Forging Ahead with AI

AMP is a subsidiary of ArcelorMittal, which was formed by the merger of steel giants Arcelor and Mittal in 2006. The merged entity is the world's leading steel and mining company, counting a presence in 60 countries and an industrial footprint in 17, supplying steel for automotive production, construction, household appliances and packaging; 2020 revenues totaled $53.3 billion.

ArcelorMittal Poland's operations are distributed across six plants that employ more than 10,000 people. Recent technology upgrades, which included artificial intelligence software and machine learning capabilities, have made the AMP's operations more efficient and streamlined. 

"Camera operators previously used video technology to categorize and tag incoming cars, which was a time-consuming process," Wisla says. "Using the new technology frees up operators to focus on other important tasks. The new solution has the capacity to log critical pieces of data quickly and accurately." 

Automation and sensors help move large rail cars

Image credit: ArcelorMittal Poland

Weighing In with Smart Technology 

Automating these processes and freeing employees from watching multiple video feeds and inputting critical data annually was essential to scaling AMP's operations. The technology solution also needed to be sufficiently hardy enough to remain operational in harsh weather and temperature extremes. harsh environmental conditions such as heat, rain, and cold temperatures.

On any given day, AMP must keep track of more than 1,000 rail cars in motion at a single factory. Each car's cargo can weigh in at more than 20 tons—so the inventory of materials in transit at any given time is non-trivial. Just the process of weighing the cars alone was complicated, with multiple checkpoints across a location where cars are weighed. 

AMP's hybrid solution combines computer vision, deep learning, and near-real-time processing, and identifies and recognizes rail cars throughout a factory with just a couple of cameras per checkpoint. The weighing system uses the Intel® Distribution of OpenVINO™ toolkit, whose algorithms match the cargo weight to an image of the car for tracking. 

The Intel® Distribution of OpenVINO™ toolkit also allows AMP developers and data scientists to accelerate and streamline AI development and deep learning applications and algorithms. The Polish steelmaker also used the OpenVINO™ toolkit to run inference and deep learning models accelerated by Intel® FPGAs to achieve greater efficiency.

It's this kind of innovation that's driving key industrial business and manufacturers to embrace Industry 4.0, using analytics, artificial intelligence, and the Internet of Things (IoT) to transform business processes, strategic decisions and productivity. Some 86 percent of factory executives reported big increases in shop-floor data collection during the last two years; 66 percent reported that data insights have led to quality improvements and efficiency savings of 10 percent or more, according to a study by Intel.

As AMP demonstrates, It's clear that digital transformation is a top objective for industrial organizations and manufacturers looking to innovate and stay competitive.

 

Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries.

Find Intel® IoT Market Ready Solutions

Search the Intel® IoT partner ecosystem and discover 130+ end-to-end IoT solutions.

Become a Member of the Intel® Partner Alliance

If you are delivering repeatable and scalable IoT offerings using Intel technology, the Intel® Partner Alliance program gives you more opportunities to build your business. Get started.