To overcome economic challenges and remain competitive, European steel producers are increasingly turning to AI and operations research (OR). This reduces costs and increases efficiency. But what are the challenges involved in implementing such technologies and how can they be successfully overcome?
The decline in automotive and industrial production, persistently weak economic growth, and changing industry trends have led to lower overall demand for steel. There is currently no sign of a short-term upward trend. In addition, unstable supply chains are putting further pressure on the steel industry. Instability is further exacerbated by geopolitical tensions and fluctuating energy costs. Furthermore, the market is becoming even more complex due to regulatory pressure and increasing protectionism. This is because trade barriers could restrict access to global markets or be accompanied by stricter environmental standards. Not only is profitability affected by the development, but steel manufacturers are also forced to overcome the challenges of the uncertain market in order to increase their efficiency and remain competitive.
AI and OR not only improve decision-making, but also help to optimize processes and make the company more adaptable. AI also automates (time-consuming/complex) tasks. For example, the creation of market reports and sales forecasts by analyzing sales data and identifying trends. Operational research can be used to identify bottlenecks and allocate resources efficiently, thereby eliminating planning problems in production, logistics and human resources.
Using modern approaches from artificial intelligence and operations research, steel manufacturers can make their processes more resilient, flexible and cost-efficient – and thus maintain their position on the market in the long term.
Forecasts for demand planning
When creating forecasts in the steel industry, AI and statistical models are used to analyze historical data, market trends, and external factors such as economic conditions and raw material prices. The input data includes past demand, production rates, and external indicators. Accurate demand forecasts are created using methods such as time series analysis, machine learning, and regression models. This creates the basis for more efficient production planning, lower stock levels (reduced NWC) and greater operational stability – with the aim of recognizing market fluctuations at an early stage and managing them with foresight.
Sales and operations planning (S&OP)
S&OP brings production, sales, and procurement together in an integrated planning process by transferring data from different areas of the company into a common basis for decision-making. The input data includes demand forecasts, production capacities, and possible supply restrictions.
Optimization algorithms can be used to automatically create production plans that take capacity restrictions into account. Scenario analyses then enable the simulation of various scenarios from which the most suitable one can be selected. The result is a coordinated operating strategy that increases efficiency, optimizes the use of resources, and significantly improves responsiveness to market changes.
Blending
Blending in steel production is the targeted process of mixing different raw materials, such as iron ore, scrap, and alloying elements, to ensure the desired steel quality at optimized production costs. The inputs include the properties of the raw materials, cost fluctuations in raw materials, and production requirements. Optimization algorithms can be used to calculate the optimum mixing ratios in order to achieve the desired results while taking resources and targets into account. Secondary blending using scrap offers a special lever: High-quality alloying elements contained in scrap can be used strategically to reduce the need for expensive primary alloys. Among other things, this creates considerable cost optimization potential while maintaining the same product quality. A simulation tool can help with the optimal selection of ingredients and predict the likely quality of the blend. The result is lower raw material costs, consistent product quality, more efficient use of resources, and greater sustainability thanks to the optimal use of scrap resources.
Digital twin for transparency in the supply chain and cost reduction
The digital twin technology creates a virtual image of the supply chain by integrating real-time data from logistics, procurement, and production. Inputs include IoT sensor data, transportation routes, and performance metrics from suppliers. Simulation models and AI-driven analyses identify inefficiencies, predict disruptions, and optimize procurement strategies. As a result, companies gain transparency along the supply chain, reduce transportation and procurement costs, and, at the same time, increase their resilience to disruptions.
Predictive maintenance
Predictive maintenance uses data analysis and real-time monitoring to prevent system failures and optimize maintenance plans in steel production. The inputs include sensor data, master data, and historical performance records. Advanced techniques such as anomaly detection and predictive modeling identify potential problems before they cause costly downtime. This leads to higher operating efficiency, reduced maintenance costs, and increased system reliability, which in turn enables uninterrupted production and improved profitability.
News Radar
Monitoring industry news and market trends ensures that steel manufacturers are aware of changes in regulations, raw material prices, and competitive dynamics. Market data includes news, economic reports, and political updates. Natural language processing and sentiment analysis allow relevant insights to be extracted from large amounts of information. This creates a reliable basis for decision-making, targeted risk management, and the necessary foresight to react proactively to market changes.
Several challenges can hinder implementation and effectiveness. One major hurdle is data quality, as traditional AI models require accurate, consistent, and well-structured data. Many steel producers therefore have to plan for expenses in order to optimize fragmented, incomplete, or partially outdated data records. Furthermore, stricter EU regulations on data protection and the ethical use of AI have added complexity as companies must ensure transparency, security, and legal compliance. Those who overcome these challenges will be able to exploit the full potential of AI in the steel industry.
After all, change management is crucial. Without acceptance, appropriate training, and the reduction of resistance, even promising AI integrations can fail.
A structured and strategic approach guarantees long-term benefits and acceptance. A clear strategy is also essential in order to define the role of AI in the company, align it with business objectives, and set measurable goals. Without a solid strategic foundation, AI initiatives run the risk of being applied in a fragmented manner and thus failing to deliver meaningful results. Companies should start with small pilot projects that address specific problems and enable controlled tests and iterative improvements. If these pilot projects prove successful, they can be gradually rolled out across the entire company to maximize their impact.
In addition, effective change management is essential to overcome resistance and ensure that employees and stakeholders understand the benefits of AI and feel included in the transformation process. Training, the promotion of cooperation, and transparent communication help to build trust and promote acceptance. Finally, companies must ensure integration with existing systems and processes and also take regulatory and ethical aspects into account, especially in industries with strict compliance standards such as in the EU. Through a combination of strategic vision, step-by-step implementation, and strong stakeholder engagement, companies can successfully leverage AI to benefit from efficiency, innovation, and competitive advantages.
As the competitive landscape changes, companies that fail to adapt could fall behind. AI-powered solutions, such as predictive analytics and digital supply chains, hold significant potential to improve resilience and reduce costs in a volatile market.
As your partner in this transformation, Horváth provides you with the expertise and support you need to make the most of the potential of AI in the steel industry. Find out more about our AI solutions here or contact our experts for a personal consultation.
Schneider, O. / Voß, F. / Poscher, B. / Wichert, S.
