With more precise price forecasts, purchasing timings can be planned much more accurately.
– Christian Jabs, CEO
Mr. Jabs, pacemaker.ai sees itself as a "digital pacemaker for global supply chains." Where do supply chains most frequently lose their rhythm?
pacemaker.ai is a provider of digital supply chain solutions with which we make supply chains more efficient and controllable using machine learning and natural language processing. As part of thyssenkrupp, we emerged directly from the concrete requirements of industry – from real customer problems along global value chains. Today, supply chains are characterized by global networks, volatile markets, and strongly fluctuating prices. This "jungle" can no longer be mastered with classic, static models and pure gut feeling. This is exactly where we come in.
A current focus of pacemaker.ai is AI-powered commodity price forecasting. Why are AI-based forecasts superior to traditional market analyses?
A central problem that many of our customers address is the high volatility of commodity prices and the question of how to sensibly hedge against them – for example, through hedging or long-term pricing decisions. This is where our AI-powered commodity price forecasting comes into play. Our models incorporate historical market data, supply and demand signals, as well as current price-relevant news. Methodologically, we work with an ensemble of different AI approaches that map different patterns and time horizons. Traditional market analyses remain important but reach their limits in volatile conditions. AI is the next evolutionary stage because it systematically links these analyses and continuously translates them into reliable price signals. This allows us to achieve a forecast accuracy of 97 to 99 percent for daily forecasts on a monthly basis – creating a solid foundation for strategic decisions in an increasingly volatile environment.
What changes at the strategic level through reliable commodity price forecasts?
The biggest leverage lies in better timing. With more precise price forecasts, purchasing timings can be planned much more accurately. In the short term, this means: buying earlier when commodity prices are rising, and consciously buying later when prices are falling. In the long term, algorithmic forecasting enables the development of robust price hedging strategies that systematically account for volatility rather than merely reacting to it.
What does practical implementation look like? Can you cite typical use cases?
Take aluminum purchasing in the automotive sector: There, the aluminum price directly determines the unit costs of a vehicle, and often for several years. Buyers must determine early on at what costs body and structural parts will go into series production, even though the commodity price changes daily. Our commodity price forecasting provides reliable price signals over the relevant time periods. Buyers recognize early whether a price increase is emerging and can specifically lock in volumes, tighten price adjustment clauses, or build hedges. This transforms purchasing from reactive renegotiation to an active control function for unit costs and margins over the entire model cycle.
A PUBLICATION BY SMART MEDIA