Have we overcomplicated the role of AI in factory process optimization? The answer might surprise you. While grand visions of AI-driven smart factories dominate headlines, the real power of industrial AI lies in its ability to enhance existing workflows, enabling process engineers and metallurgists to work faster and more efficiently.
Although the promise of using artificial intelligence is still valid for creating a smarter Industry 4.0 factory, most AI end users - process and production engineers, plant managers, and data scientists don’t have the bandwidth to take on a solution that would require them to alter their current workflow.
They barely have the bandwidth to dedicate to a proof of concept pilot period. Much of their bandwidth is spent resolving the non-mechanical issues that disrupt their operations and quality measures.
However, reducing the time spent resolving these issues is where AI can add its greatest and most immediate value.
Speed: The Key to Economic and Environmental Gains
The steel industry, known for its energy-intensive processes, stands to benefit significantly from AI-driven optimizations. By focusing on speed-to-solution, steel manufacturers can reduce costs and minimize their environmental impact without overhauling trusted methodologies.
Accelerating Lean Six Sigma
Industrial AI doesn't replace traditional Lean Six Sigma processes; it supercharges them. Supported with the power of AI, engineers can perform diagnostics, run regression models, and conduct root cause analyses in minutes rather than weeks. This acceleration means less time spent in suboptimal production states, directly translating to reduced energy consumption and material waste.
Energy Optimization: A Case in Point
Energy expenditure typically accounts for 20-40% of total production costs in steel manufacturing. AI-driven optimizations in this area can lead to substantial economic and environmental benefits:
- Blast Furnace Optimization: AI algorithms analyzing real-time data have helped companies like ArcelorMittal reduce energy consumption by 5% while improving production efficiency.
- Electric Arc Furnace Efficiency: Swedish manufacturer SSAB employed AI to optimize their electric arc furnaces, cutting energy usage by 7% and significantly reducing carbon emissions.
- Rolling and Finishing Processes: AI-driven process control in these stages can lead to energy savings of 3-5%.
Real-World Impact: Gerdau's Success Story
Gerdau, a leader in responsible steelmaking, has leveraged industrial AI to optimize its use of ferroalloy additives. Prior to using AI, they would assume the worst-case scenario with their production recipes, including a buffer for failure to account for any uncertainties. By using AI to predict the lowest amount of alloys needed to meet specifications, Gerdau reduced their alloy costs by $3 per ton and increased quality variation by 15%.
This optimization not only improved their bottom line but also reduced their environmental footprint by minimizing the amount of additives that they needed to purchase, ship, and store, which decreased their Scope 1 and 2 emissions. Using AI to ensure their production stays in-spec dramatically increased their speed to solution to function adaptively rather than reactively.
The Power of Prediction
AI's ability to increase forecast accuracy has far-reaching implications. More reliable quality predictions mean fewer lab tests, reduced production delays, and less energy waste.
One large chemical manufacturer reported that their complex lab tests require a 24-hour turnaround. This time period is not unsubstantial, noting that a single day of bad production could cost them £35,000 in wasted energy consumption.
Using AI to develop more regular and accurate forecasts would save this plant a significant amount of revenue expenditure each month. It would dramatically increase their speed to market in addition to ensuring they’re producing the high quality product that their customers expect.
Simplifying AI's Role to Streamline Pilots
The key to unlocking AI's potential in steel and chemical manufacturing lies not in complete system overhauls, but in empowering teams to work faster within their existing frameworks. By focusing on speed-to-solution, manufacturers can:
- Increase their speed to insight
- Reduce daily energy expenditure
- Minimize scrapped batches
- Optimize raw material usage
- Streamline operations to reduce carbon capture needs
- Increase their speed to market
- Simplify and accelerate AI pilot criteria
McKinsey & Company reported that 74% of industrial technology pilots do not make it out of pilot purgatory. The reasons for this are simple, most pilots are too long, too complex, and too targeted.
When a pilot is simplified to prove that a platform can increase a team’s speed of function and speed to insight, the pilot requirements become more precise, and the value of the platform becomes broader.
In 2025 and beyond, the most impactful change for steel manufacturers may be the simplest: equip teams with Fero Labs' industrial AI to accelerate their current processes.
This approach not only accelerates pilot requirements, it boosts margins and significantly reduces a plant’s environmental impact – that’s a win-win for both the plant and the planet.
[Originally published: Industry4.0 Magazine]
