Demand Forecasting in Supply Chain

7-2 Outline
 The role of forecasting in a supply chain
 Characteristics of forecasts
 Components of forecasts and forecasting methods
 Basic approach to demand forecasting
 Time series forecasting methods
 Forecasting in practice
7-3 Role of Forecasting
in a Supply Chain
 The basis for all strategic and planning decisions in a supply chain
 Used for both push and pull processes
 Examples:
 Production: scheduling, inventory, aggregate planning
 Marketing: sales force allocation, promotions, new production introduction
 Finance: plant/equipment investment, budgetary planning
 Personnel: workforce planning, hiring, layoffs
 All of these decisions are interrelated
7-4 Characteristics of Forecasts
 Forecasts are always wrong. Should include expected value and measure of
error.
 Long-term forecasts are less accurate than short-term forecasts (forecast
horizon is important) (e.g. monthly shopping of groceries)
 Aggregate forecasts are more accurate than disaggregate forecasts (e.g.
company growth vs sku level growth)
 Forecast error increases as the distance from customer increases (e.g.
bullwhip effect)
7-5 Forecasting Methods
 Qualitative: primarily subjective; rely on judgment and opinion (e.g. new
products)
 Time Series: use historical demand only
 Causal: use the relationship between demand and some other factor to
develop forecast (e.g. discounts)
 Simulation
 Imitate consumer choices that give rise to demand
 Can combine time series and causal methods
7-6 Basic Approach to
Demand Forecasting
 Understand the objectives of forecasting (who all are affected?)
 Integrate demand planning and forecasting throughout the supply chain
(retailer’s forecast must match the company’s forecast)
 Identify major factors that influence the demand forecast
 Understand and identify customer segments
 Determine the appropriate forecasting technique
 Establish performance and error measures for the forecast
7-7 Time Series
Forecasting Methods
 Goal is to predict systematic component of demand
 Multiplicative: (level)(trend)(seasonal factor)
 Additive: level + trend + seasonal factor
 Mixed: (level + trend)(seasonal factor)
 Static methods
 Adaptive forecasting
Risk Management in Forecasting
 Examples contributing to forecasting error
 Long lead times require forecast to be provided in advance, hence reliability of forecast reduces
 Seasonality
 Continuously shortening product life cycles
 Inconsistent customer base (e.g. kirana store in remote location)

 How do you mitigate the risk?

 Pooling of demand (e.g. consolidating warehouse)
 Increasing responsiveness of supply chain thereby reducing lead time, hence forecast can be given
for shorter durations.
 Collaborative forecasting
7-9 Forecasting in Practice
 Collaborate in building forecasts
 The value of data depends on where you are in the supply chain
 Be sure to distinguish between demand and sales
Thank you