Michel Plourde, Systems Director at ArcelorMittal Canada, recently briefed ARC Advisory Group about the supply chain challenges surrounding the company’s decision in 2012 to expand its iron ore pellet production in Canada. These challenges largely involved eliminating logistics bottlenecks created by an inability to adapt quickly to disruptions and changes to plans.
ArcelorMittal, the world’s largest steelmaker, operates two interdependent companies in Quebec: ArcelorMittal Mines Canada and ArcelorMittal Infrastructure Canada. Following a recent 10 million metric ton expansion project and subsequent ramp-up in 2013-14, the company’s Canadian operations now extract, process, haul, and ship roughly 26 million metric tons of iron ore products yearly to support their steel production operations.
ArcelorMittal’s geographically dispersed operations in Quebec include open pit iron ore mines and associated processing facilities in Mont-Wright and Fire Lake; a pelletizing plant and private shipping port in Port-Cartier, Quebec; and a 420 km railway connecting the mines and the port.
At the time of the expansion project, ArcelorMittal had already been using a business intelligence (BI) platform for over 15 years. This core BI solution and database – the OSIsoft PI System – was used, and continues to be used, for collecting real-time data and transforming that data into actionable information.
However, when the company significantly expanded their production of iron ore pellets, a bottleneck emerged in logistics. By implementing a new information strategy to support the expansion project, ArcelorMittal was able to improve visibility and eliminate the bottlenecks. Significantly, this new information strategy, enabled by the company’s BI platform, did not require a major capital expenditure.
To help remove the logistics bottlenecks and improve resiliency, ArcelorMittal used the PI System to transform and correlate data from its disparate systems to produce intelligence to support both strategic decisions and day-to-day decisions related to optimizing the use of the company’s heavy mining equipment, railways, and port transportation infrastructure.
At its expansive open pit mining operation in Mont-Wright, the company extracts iron ore using a huge electric excavator and a fleet of 50 Caterpillar 240-ton and 400-ton haul trucks.
ArcelorMittal processes the ore at the mines and then loads the iron ore concentrate onto rail cars to transport it to the company’s pelletizing plant located at its port facility in Port-Cartier, Quebec. This railway is over 420 kilometers and passes through some very demanding geographical areas. The company operates approximately 36 GE and EMD locomotives and over 1,200 ore cars. They track this rolling stock on a daily basis.
The port, excavated out of rock and built in the late 50s, has only 54 feet of water at high tide. Loading facilities at the port can only allow loading one ship at a time. Getting the ship in and out based on tides represents another challenge.
This IT project’s goal really surrounded visibility and better planning: to be able to determine when production targets were met, when ore was being delivered, the right ships to load the iron ore products onto, the right tonnage to load, and the optimum times to do so. Workers needed to know when something went wrong and receive actionable intelligence to help resolve the issues quickly and efficiently.
Prior to the capacity expansion, the company had over 20 years of data stored in its PI System, encompassing over 50,000 data points from their distributed control system (DCS) and programmable logic controllers (PLCs). The company was good at integrating, diagnosing, and analyzing data from an operational perspective; but needed additional visibility to be able to predict and confirm what it wanted to happen along the supply chain. In other words, the company wanted to move more to a predictive environment for its critical logistics.
Because its logistics supply chain was only geared for 16 million metric tons, the company ran into severe bottlenecks at the port following the expansion to 26 tons. This meant loading an additional 150 ships per year…one ship at a time.
ArcelorMittal was able to optimize its mining and rail operations, but encountered bottlenecks at its port because workers there didn’t have adequate visibility into the tonnage being produced and transported to Port-Cartier via rail. The company needed to align its business metrics with mining and logistics operations so that workers could understand the plan, budget and determine the tonnage that would need to be extracted, processed, transported via rail, then shipped from the port in a given month.
ArcelorMittal used the OSIsoft architecture already in place to add business context and visibility to the data to provide actionable intelligence. This included automatic and consistent supply chain tracking. The company used the PI System infrastructure as the consistent information enterprise HUB to integrate data from its many systems.
The company integrates data from the shipboard automatic identification system (AIS) systems with PI System data and then graphically overlays the information on Google Earth, enabling port personnel to easily visualize the port and ship traffic and trigger ship loading, equipment maintenance, and other activities as needed. The colors on the graphical port display change as events occur. Asset and event data were also integrated into the maintenance systems to optimize maintenance scheduling based on market requirements for its fixed and moveable assets.
By using this geographical port visualization, the company can see all ship movements, manage the associated tugboat and loading operations, and combine this intelligence with tidal information to optimize port efficiencies. They can also get railway updates, see what tonnage is coming in, and what the downstream crushers in production are doing and more.
This visual representation of the port, supported with KPI-based planning, helps workers to make better decisions by understanding constraints and better optimizing operational parameters. For example, workers use the solution to adjust production rates and decide whether to stockpile ore or pellets or move them along the supply chain.
Users can answer questions such as, “How do I push the load to the right ships at the right time?” and “How do inefficiencies in the process affect income?” The model is used to determine real-time data targets and costs. Out of tolerance events initiate actions. According to Mr. Plourde, “The data now leads us to clear and precise decisions with information to the right people at the right time in the right context.”
The benefits? ArcelorMittal increased its shipping capacity from 13 million metric tons to 26 million tons, to meet its new throughput targets. The company has met its targets for increased throughput. The company estimates that, in 2015, the additional 10-million metric tons shipped was equivalent to $120 million USD in additional revenue based on current commodity prices.
According to Mr. Plourde, “It was amazing to see how operators viewed the data, wanted to help, and even asked for additional information.” With minimal investments, the company was able to solve its supply chain bottlenecks and develop a new collaborative mindset with more engaged employees. The company’s success with this project is influencing other projects across its supply chain. For additional information readers can view the presentation.
