By the end of the session, participants will be able to:

• Structure Sales and Inventory Data for Forecasting: Clean and organize historical sales and inventory data into a consistent, analysis-ready format that supports forecasting and operational decisions.
• Convert Sales to Demand: Understand the difference between observed sales and true demand and identify situations where sales may be constrained by stockouts, availability, or distribution limits.
• Identify Sales Patterns That Matter: Detect volatility, stability, and seasonality across products and channels, and understand the drivers behind these patterns.
• Choose the Right Forecasting Method: Decide when to use simple averages, seasonal indices, or basic smoothing techniques based on data behavior and business context.
• Build the Forecast: Create a practical forecast at the appropriate level (product, channel, or aggregate) and translate it into weekly or monthly demand projections.
• Monitor Accuracy: Track forecast performance over time using simple accuracy measures and learn how to adjust forecasts when patterns change.

Workshop Outline
I. Welcome & Context Setting
 Participant introductions and business context
 Common challenges in forecasting
 What this workshop will and will not cover
II. Understanding the Data
 Typical issues in sales and inventory data
 Structuring data for demand forecasting
 Sales vs. demand, understanding when sales do not reflect true demand
III. Identifying Patterns That Matter
 Volatile vs. stable demand
 Seasonal vs. non-seasonal behavior
 Channel-driven vs. product-driven patterns
 Why pattern recognition matters more than model complexity
IV. Hands-on data preparation practice
 For a set of dummy data, practice data clean up, structure and analyze data to identify patterns
V. Choosing the Right Forecasting Approach
 When an average is the right answer
 When and how to use seasonal indices
VI. Building the Forecast
 Creating a forecast at the right level (channel, product group, time horizon)
 Forecast aggregation and disaggregation
VII. Hands-on data preparation practice
 Using the same data, select the right forecasting algorithm and calculate the forecast for the next year
VIII. Monitoring and Adjusting Forecast Accuracy
 Simple accuracy measures
 Recognizing when a forecast needs adjustment
 Avoiding overreaction to short-term fluctuations
IX. Hands-on data preparation practice
 Adding dummy sales data to the same database, calculate forecast error and decide if and how to adjust the forecast
X. Wrap-Up & Next Steps
 Key takeaways
 Q&A

Facilitator: Helia Sohrabi, Ph.D.
Helia Sohrabi is a data scientist in the field of supply chain with over 18 years of experience working at the intersection of data analytics, retail operations, and academia. Her background spans hands-on leadership roles in retail and consulting, where she has worked closely with operations, IT and e-commerce teams.

She has led forecasting, inventory planning, and allocation initiatives for large retailers, such as Simons and Psycho Bunny, helping organizations improve forecast accuracy, reduce stockout risk, and translate complex analytical concepts into practical business decisions.

Helia holds a Ph.D. in Operations Management, which underpins her analytical approach and informs how she evaluates forecasting methods, data, and decision-making frameworks in real-world environments. Today, she runs MontWave Consulting, where she supports businesses in building realistic, data-driven forecasting and inventory strategies.

Helia is passionate about bridging theory and practice and about supporting professionals in making better decisions driven by data, without over-engineering or unnecessary complexity. This session draws on real retail and supply chain contexts, with a focus on practical application rather than abstract theory.