Fraunhofer IWU builds AI demand forecasting for MÖVE textiles to stabilize planning

Textile companies have a classic planning problem: demand is noisy, capacity is expensive, and reality rarely matches the spreadsheet. Fraunhofer IWU stepped into that mess with an AI-powered demand forecasting tool built for frottana Textil GmbH & Co. KG, the company behind the MÖVE brand. The goal is straightforward and operationally brutal: forecast sales more reliably, plan production capacity more fully digitally, and feed employee know-how into the process in a systematic way.

At the core, the tool does what good forecasting systems always do. It intelligently analyzes historical sales data and then provides companies with a robust, data-driven basis for sales and order planning. That means fewer blind guesses when turning past performance into near-term expectations for what customers will actually buy. In a subsequent step, production planning could also be adapted, extending the impact from “what we expect to sell” to “what we plan to make.”

Why this matters right now is that textile operations live and die by execution. When forecasts are off, the pain shows up quickly across the supply chain. Orders can arrive that production did not prepare for, leading to either stockouts, expedited logistics, or rushed manufacturing changes. Or the opposite can happen: capacity gets scheduled for demand that never materializes, which can leave inventory sitting longer, tying up working capital and compressing margins. The source frames the AI tool as a way to give planning teams a more reliable foundation. In practice, that is how you reduce planning volatility, stabilize order commitments, and make digital planning workflows more trustworthy.

There is also a second layer here that boards and CFOs tend to care about: reliability is not just an operational metric, it is financial risk control. Sales and order planning are the early inputs into downstream commitments, from material purchasing to production schedules. If those inputs are based on inconsistent signals, companies end up paying for corrections later. The Fraunhofer IWU approach focuses on structured analysis of historical sales data. The point is not magic forecasting, it is improving the decision pipeline, so planning outputs are more consistent and can be managed digitally rather than patched by manual work.

For context, most industrial planning systems have a tension that companies know too well. Humans hold know-how that models do not automatically capture, such as how customer relationships shift or how product mix trends behave in specific markets. Meanwhile, digital systems excel at processing volume and generating repeatable workflows. The source explicitly positions the tool to integrate employee know-how systematically at the same time as forecasts are improved. That is important because it signals an implementation mindset. The project is not only about producing a forecast number; it is about embedding the forecasting approach into how the organization plans, learns, and updates.

On the regulatory and standards side, the textile industry is increasingly shaped by product, sourcing, and reporting requirements across markets. Even when the forecasting use case is internal, improved planning tends to support compliance-friendly operations because it can reduce waste from overproduction and improve traceability of production decisions. While the source does not cite specific regulations, the broader compliance environment is one reason planning reliability has become more than a cost issue. If you are producing more efficiently and planning capacity with clearer demand signals, it can help reduce the operational churn that often makes reporting harder.

Finally, the second-order implication for peers is competitive pressure that does not show up in marketing. When one operator stabilizes planning, it often improves customer responsiveness too. Better sales and order planning can translate into more accurate delivery expectations, fewer disruptions, and smoother production execution. And because the tool can feed into subsequent production planning adaptation, the architecture points toward a fuller digital planning loop. For other textile and consumer-goods manufacturers, the lesson is not “install AI and win.” It is that AI becomes valuable when it strengthens the full planning chain, from historical data to orders, and then into production schedules.

In short, Fraunhofer IWU has built an AI-based demand forecasting tool for frottana Textil GmbH & Co. KG (MÖVE) that analyzes historical sales data to improve data-driven sales and order planning, with production planning adaptation possible next. For decision-makers across manufacturing, that is a clean storyline: reduce forecast uncertainty, make planning more digital, and systematically bring employee know-how into the process, so operations can run with less surprise and more control.