Two Forecasts, One Goal: How HOMAG Strengthens Its Planning Reliability with AI-Powered Forecasts

Together, the companies have implemented forecasting projects for the new machine business as well as for material requirements. The Demand Forecasting solution from pacemaker.ai uses a combination of raw data, industry knowledge, and machine learning to generate automated, highly precise forecasts. Both projects are now in productive SaaS operation.

Two Challenges, One Partner

Use Case 1: New Machine Forecast

HOMAG's new machine business is characterized by a high degree of customization. Each machine is tailored to specific customer requirements. The resulting variety in the underlying data makes the strategic planning of capacity, materials, and resources more difficult.

The goal was therefore defined as developing a monthly forecast at product series level over a rolling 12-month horizon in order to support planning with data-based forecasts.

Use Case 2: Material Forecast

In parallel, HOMAG faced the challenge of forecasting demand for externally procured production material more precisely. The volatility and complexity of the consumption patterns left room for improvement in forecast quality.

Here too, the goal was clearly defined: a reliable forecast of material requirements per plant in order to place supply chain management and production on a solid data basis.

From Data Thinking Workshop to Productive Solution

The joint journey began with a Data Thinking Workshop in which the requirements of both use cases were defined and prioritized. From this, the New Machine Forecast and the Material Forecast emerged as two clearly delineated projects.

In weekly coordination, the teams from HOMAG and pacemaker.ai worked closely together, from data validation through to fine-tuning to increase precision.

Within this structured collaboration, both use cases were transitioned into SaaS operation on schedule.

Initial Results

Both forecasts are already delivering measurable improvements. The New Machine Forecast reduces the forecast error, meaning the deviation between predicted and actual demand, by 50% per segment and month. This creates a significantly more robust foundation for capacity and resource planning.

The Material Forecast today captures around 25,000 materials instead of the previous 3,500 and thus also covers materials that had previously not been included in the forecast. At the same time, it increases forecast accuracy by 15%.

Outlook

With the transition of both projects into productive SaaS operation, HOMAG completes an important step toward data-driven planning and at the same time creates the basis for further optimization along the value chain.

The successful collaboration shows that AI-powered forecasts can contribute to planning reliability even in highly customized, project-based business models, provided the approach is developed jointly, implemented in a structured way, and consistently evaluated.