Supply Chain Management (SCM) 101: Demand Forecasting Forecasting Methods Qualitative Expert Opinion Delphi Quantitative Time Series Causal

What This Is

Forecasting is the process of predicting future demand or supply chain events. Accurate forecasting is crucial in supply chain management as it enables companies to make informed decisions about production, inventory, and logistics. For example, Amazon uses forecasting to determine the optimal inventory levels for its products, ensuring that customers can receive their orders on time.

Key Frameworks & Formulas

• Time Series Forecasting: A method that analyzes historical data to predict future demand. It can be further divided into:
  • Moving Average (MA): A simple method that calculates the average of past demand values.
  • Exponential Smoothing (ES): A method that gives more weight to recent data points.
• Causal Forecasting: A method that identifies the relationships between variables to predict future demand.
  • Regression Analysis: A statistical method that models the relationship between variables.
• Safety Stock: The additional inventory held to mitigate the risk of stockouts.
  • Safety Stock Formula: SS = Z × σ × √L, where Z is the Z-score, σ is the standard deviation, and L is the lead time.
• Economic Order Quantity (EOQ): The optimal order quantity that minimizes the total cost of inventory.
  • EOQ Formula: EOQ = √(2DS/H), where D is the demand, S is the ordering cost, and H is the holding cost.
• Fisher's Model: A framework that categorizes products into three categories based on their demand variability and lead time.
  • Product Categories:
    • Product A: High demand variability and short lead time.
    • Product B: Low demand variability and short lead time.
    • Product C: Low demand variability and long lead time.

Step-by-Step Application

1. Calculate Safety Stock: Determine the Z-score, standard deviation, and lead time for a product. Then, use the safety stock formula to calculate the additional inventory needed.
2. Implement Time Series Forecasting: Collect historical data for a product and use a moving average or exponential smoothing method to predict future demand.
3. Conduct Regression Analysis: Identify the relationships between variables that affect demand for a product. Then, use regression analysis to model the relationship and predict future demand.
4. Determine EOQ: Calculate the demand, ordering cost, and holding cost for a product. Then, use the EOQ formula to determine the optimal order quantity.
5. Categorize Products using Fisher's Model: Determine the demand variability and lead time for each product. Then, categorize the products into Product A, B, or C based on Fisher's Model.

Common Mistakes

• Mistake: Assuming that all products have the same demand variability and lead time.
• Correction: Use Fisher's Model to categorize products based on their unique characteristics.
• Mistake: Failing to account for seasonality and trends in demand.
• Correction: Use time series forecasting methods to account for seasonality and trends.
• Mistake: Ignoring the relationships between variables that affect demand.
• Correction: Conduct regression analysis to identify and model the relationships between variables.

Exam / Certification Tips

• Common Question Patterns: Time series forecasting, safety stock calculation, and EOQ determination are common topics on SCM exams.
• Tricky Distinctions: Understand the differences between push and pull strategies, efficient and responsive supply chains, and Incoterms responsibility.
• Focus on Formulas: Make sure to memorize and apply the formulas for safety stock, EOQ, and time series forecasting.

Quick Practice Problem

Scenario: A company sells 100 units of a product per month with a standard deviation of 10 units. The lead time is 5 days, and the desired service level is 95%. What is the safety stock needed for this product?

Answer: Safety stock = 2.33 × 10 × √5 = 33.5 units

Explanation: The company needs to hold 33.5 units of inventory to mitigate the risk of stockouts and meet the desired service level.

Last-Minute Cram Sheet

• ⚠️ Time Series Forecasting: Analyze historical data to predict future demand.
• Safety Stock Formula: SS = Z × σ × √L
• EOQ Formula: EOQ = √(2DS/H)
• Fisher's Model: Categorize products into Product A, B, or C based on demand variability and lead time.
• Regression Analysis: Identify and model the relationships between variables that affect demand.
• ⚠️ Push vs Pull: Push strategies focus on producing and storing inventory, while pull strategies focus on producing and delivering products to customers.
• ⚠️ Efficient vs Responsive: Efficient supply chains focus on minimizing costs, while responsive supply chains focus on meeting customer needs.
• Incoterms Responsibility: Understand the responsibilities of buyers and sellers under different Incoterms.