Home – Demand Planning at Danone Polska, Nutricia Polska, and Żywiec Zdrój

Demand Planning at Danone Polska, Nutricia Polska, and Żywiec Zdrój

We implemented a comprehensive demand planning system for three key Danone business units in Poland: Danone Polska, Nutricia Polska, and Żywiec Zdrój.

Demand Planning

About the project

We implemented a comprehensive demand planning system for three key Danone business units in Poland: Danone Polska, Nutricia Polska, and Żywiec Zdrój. Danone is a global leader in specialized nutrition (for infants, children, and adults), yogurts (dairy and plant-based), and bottled water. Its most recognized brands include Bobovita, Nutricia, Danone, Activia, Fantasia, Actimel, and Żywiec Zdrój.

The project covered various business segments, each with unique demand dependencies—ranging from fresh products with short shelf lives, requiring daily-level precision planning, to categories strongly influenced by seasonality and promotions. Differences in sales dynamics, order cycles, and market strategies required a flexible forecasting approach and the integration of data from multiple sources. The implementation of a unified system enabled process automation, improved forecast accuracy, and optimized inventory levels tailored to each product category.

Project objectives and challenges

Improving forecast accuracy

Manual forecasting did not consider all key variables, leading to errors and inefficient inventory management.

Reducing demand planners’ workload

A significant amount of manual effort was required to prepare demand forecasts.

Optimizing stock quality

Ensuring proper inventory levels, reducing shortages, and improving stock turnover.

Migrating to a unified demand planning tool – IBP (Integrated Business Planning)

Standardizing planning processes across a single platform.

Ensuring proper input data for forecasting

Providing consistent and reliable data sources.

Automating forecasting processes

Implementing statistical and machine learning models.

Full integration of financial and promotional data

Automatically incorporating promotions and market trends into demand forecasts.

Diverse sales channels

Each product category was distributed through multiple channels, such as modern trade, traditional retail, e-commerce, and HoReCa, requiring demand forecasting to be adapted to different ordering patterns.

External factors impacting demand

Forecasts had to account for macroeconomic factors such as inflation, changes in consumer purchasing power, and evolving health and sustainability trends.

Seasonality and demand volatility

Categories like bottled water and dairy products were highly dependent on seasonality and promotional activities, necessitating flexible planning and historical sales analysis.

Complex product portfolio

A broad range of products with different life cycles, including new market launches, required precise forecast management to avoid both stock surpluses and out-of-stock situations.

Our Approach & Solutions

Implementation of the IBP (Integrated Business Planning) system

✔ Customization of IBP to meet specific business needs.
✔ Aggregation and disaggregation rules enabling forecasting at various levels tailored to product segment requirements.
✔ Automated data cleansing mechanisms and seamless data uploads.
✔ Machine learning and statistical forecasting models customized for different business divisions.
✔ A unified system covering all processes, eliminating issues with multiple units of measure and inconsistent data structures.

Statistical Models & Machine Learning

✔ Time-series analysis to select the best forecasting model per business unit.
✔ Wide range of statistical models, from traditional techniques to machine learning algorithms, which consider more variables than historical data alone.
✔ Automated filtering to select the optimal statistical model, with both automated and manual forecasting options available.

Achievements & Implementation Results

Fully automated demand planning supported by a single harmonized tool for all business divisions.

Improved forecast accuracy with ML models:

Gratka: from 90.2% to 99.2%
Activia: from 88.3% to 98.5%
Danonki: from 90.2% to 98%
Alpro: from 92.6% to 92.9%

✔ Reduction in product obsolescence losses – Crucial for the dairy industry, where products have a short shelf life.
✔ Minimizing stockouts – Improved product availability increased customer satisfaction and reduced lost revenue.
✔ Optimized inventory turnover – Faster stock rotation led to higher sales and improved profitability.
✔ Migration of all interfaces – All data available in one tool, regardless of units of measure.
✔ Time savings per month for demand planners:
Nutricia Polska – 120 hours
Danone Polska – 60 hours
Żywiec Zdrój – 30 hours

Business Benefits

✔ Reduced workload – Less time spent on manual forecast creation thanks to automated statistical forecasting.
✔ Greater predictability – More accurate forecasts based on full data integration, allowing demand planners to focus on analysis rather than manual corrections.
✔ Optimized inventory & better turnover – Reduced capital tied up in inventory while ensuring products were available in line with sales forecasts.

Conclusion

The implementation of IBP (Integrated Business Planning) enabled full automation and harmonization of demand planning at Danone Polska, Nutricia Polska, and Żywiec Zdrój. By leveraging advanced statistical models and machine learning, the project significantly improved forecast accuracy, reduced inventory losses, and minimized manual workload for demand planners.

See more realizations