Search for Supply Chain.. !

Chapter Eight – Forecasting

CPIM Exam – Basics of Supply Chain Management Practice Study Sheet  Ch... thumbnail 1 summary

CPIM Exam – Basics of Supply Chain Management Practice Study Sheet 

Chapter Eight – Forecasting

Demand management is the function of recognizing and managing the demand for products. It includes 1) forecasting 2) order processing 3) making delivery promises 4) interfacing between manufacturing planning and control and the marketplace

Order processing occurs when a customer’s order is received

• Demand shows the need for an item; sales shows what was actually sold. Because demand cannot always be satisfied, demand is higher than sales

Demand patterns include four types:
1. Trend – increasing in a steady pattern of demand, or level. Can be geometric or exponential
2. Seasonality – fluctuates depending on the time of year
3. Random variation – many factors effect demand. Pattern of variation can usually be measured
4. Cycle – wavelike increases and decreases in the economy impact demand

Stable demand retains their shape and dynamic changes do not. The more stable the demand, the easier it is to forecast. The average demand can be the same as it is for stable and dynamic patterns. Usually the stable pattern is forecasted

Independent demand is not related to the demand for any other product or service. Dependent demand occurs where demand is derived from a second item. Only independent demand needs to be forecasted

Forecasting has four major principles:
1. Forecasts are usually wrong. Expect errors
2. Every forecast should contain an estimate of error
3. Forecasts are more accurate with families or groups
4. Forecasts are more accurate for nearer time periods

• Three principles of data collection are:
1. Record data in the same terms as needed for the forecast 1) data based on demand, not shipments 2) forecast time period should be the same as the schedule period 3) the items being forecasted should be controlled by manufacturing
2. Record the circumstances of the data. Other factors like sales promotions and competitors’ sales are important
3. Record demand separately for different customer groups – wholesale vs. retail may have different trends

Three forecasting techniques are 1) qualitative techniques 2) extrinsic techniques 3) intrinsic techniques

Qualitative techniques are projections based on judgment, intuition and informed opinions (SWAGS)

Extrinsic forecasting techniques are projects based on external factors (bricks to housing starts, tires to gasoline consumption). This is more useful for forecasting demand for a large family of products (cars)

Intrinsic forecasting techniques use historical data to forecast. These are the most important techniques, including:
1. average demand
2. moving averages
3. exponential smoothing (a moving average without retaining prior months) – 90% of the forecast is based on the prior months’ average and 10% on the current months’ average.

        New forecast = alpha * (latest demand) + (1 – alpha) * (previous forecast) with alpha between 0 and 1
      4. Seasonal index shows how high above or below an average for a product.  
          Seasonal index = period average demand / average demand for all periods. Average demand is “deseasonalized demand”. For seasonal demand, 1) only use deseasonalized data for forecast 2) forecast deseasonalized demand 3) apply the seasonality index to the deseasonalized forecast
Forecast error is the difference between actual demand and forecast demand. Due to bias and random variation

Bias exists when the cumulative actual demand varies from the cumulative forecast. Bias is a systemic error; need to change the forecast

Random variation – demands on the demand pattern of the project. Average error should be zero

• Mean absolute deviation (MAD) is a way to measure forecast error.

• Normal distribution by standard deviation +- 1 MAD (60%), +- 2 MAD (90%) +- 3 MAD (98%). MAD is a tracking signal to see if there is bias. Tracking signal = sum of forecast errors / MAD