1. Forecasting
Chapter 11

2. Chapter 11 Outline A Forecasting Framework
Qualitative Forecasting Methods
Time-Series Forecasting
Moving Average
Exponential Smoothing
Forecast Errors
Advanced Time-Series Forecasting
Causal Forecasting Methods
Selecting a Forecasting Method
Collaborative Planning, Forecasting and Replenishment

3. A Forecasting Framework
Focus of the chapter is on the forecasting of demand for output from the operations function.
Demand may differ from sales
Difference between forecasting and planning
Forecasting: what we think will happen
Planning: what we think should happen
Forecasting application in various decision areas of operations (capacity planning, inventory management, others)
Forecasting uses and methods (See Table 11.1)

4. Use of Forecasting: Operations Decisions

5. Use of Forecasting: Marketing & Finance

6. ‘Qualitative’ Forecasting Methods
Based upon managerial judgment when there is a lack of data. No specific model.
Major methods:
Delphi Technique
Market Surveys
Life-cycles Analogy
Informed Judgment (naïve models)

7. Time-Series Forecasting
Components of time-series data:
Trend—general direction (up or down)
Seasonality—short term recurring cycles
Cycle—long term business cycle
Error (random or irregular component)
“Decomposition” of time-series
Data are broken into the four components
Moving Averages
Exponential Smoothing

8. Moving Average Assumes no trend, seasonal or cyclical components.
Simple Moving Average:
Weighted Moving Average:

9. Moving Average Period Actual Demand Forecast 1 10 2 18 3 29 4 19
Compute three period moving average (number of periods is the decision of the forecaster)
Period Actual Demand Forecast
1 10
2 18
3 29
( )/3 = 19
Period 5 will be (18+29+actual for period 4)/3

10. Note: The more periods, the smoother the forecast.
Time-Series Data Plot
Note: The more periods, the smoother the forecast.

11. Exponential Smoothing
The new average is computed from the old average:
The value of the smoothing constant () is a choice. It determines how much the calculation smooths out the random variations. Its value can be set between zero (0) and one (1). Normally it is in the 0.1 to 0.2 range.

12. Simple Exponential Smoothing
The forecast:
F=forecast of demand (both this period and next)
D = actual demand (this period)
t = time period
No trend, cyclical or seasonal components.
Note: we are adjusting Ft to get Ft+1

13. Exponential Smoothing-calculation
Facts:
September forecast for sales was 15
September actual sales were 13
Alpha ( α) is 0.2
What is the forecast for October?
Calculation
October Forecast = September forecast + α(September actual-September forecast)
=15+0.2(13-15)=15+0.2(-2)=15-0.4=14.6

14. Forecast Errors
Cumulative Sum of Forecast Error (CFE) and Mean Error (ME)
Mean Square Error (MSE)
Mean Absolute Deviation (MAD)—measure of deviation in units.
Mean Absolute Percentage Error (MAPE)
Tracking Signal (TS)—relative measure of bias

15. Forecast Errors: Formulas
Cumulative sum of
Forecast Errors
Mean Absolute
Percentage Error
Mean Square Error
Tracking Signal
Mean Error
Mean Absolute
Deviation

16. Tracking Signal
Analogous to control charts in quality control, viz. if there is no bias, its values should fluctuate around zero.
Is a relative measure, i.e. the numbers mean the same for any forecast.

17. Advanced Time-Series Forecasting
Adaptive exponential smoothing
Smoothing coefficient () is varied
Box-Jenkins method
Requires about 60 periods of past data

18. Time Series vs. Causal Models
Time series compares data being forecast over time, i.e. Time is the independent variable or x- axis or x-variable.
Causal models compare data being forecast against some other data set which the forecaster may think is a cause of the forecasted data, e.g. population size causes newspaper sales.

19. Causal Forecasting Models
The general regression model:
Other forms of causal model:
Econometric
Input-output
Simulation models

20. Example of Time Series Model
Yt = a + b(t)
Dt = actual sales
Ft = forecasted sales
t = time period (e.g. year)
F7 = (7) = = sales forecast for next year

21. Example of Causal Model
Yt = a + b(t)
Dt = actual sales in year t
Ft = forecasted sales
It = median family income (000’s)
F7 = (7) = = sales forecast for next year (year 7)

22. Selecting a Forecasting Method
User and system sophistication
People reluctant to use what they don’t understand
Time and resources available
When is forecast needed?
What is value of forecast?
Use or decision characteristics, e.g. horizon
Data availability and quality
Data pattern
Don’t force the data to fit the model!

23. Forecast Horizons and Forecast Accuracy
The longer the forecast horizon, the less accurate the forecast
Long lead times require long forecast horizons
Lean, responsive companies have the goal of decreasing lead times so they are shorter than the forecast horizon

24. Collaborative Planning, Forecasting and Replenishment (CPFR)
Aim is to achieve more accurate forecasts
Share information in the supply chain with customers and suppliers.
Compare forecasts
If discrepancy, look for reason
Agree on consensus forecast
Works best in BtoB with few customers

25. Summary A Forecasting Framework Qualitative Forecasting Methods
Time-Series Forecasting
Moving Average
Exponential Smoothing
Forecast Errors
Advanced Time-Series Forecasting
Causal Forecasting Methods
Selecting a Forecasting Method
Collaborative Planning, Forecasting and Replenishment

26. End of Chapter Eleven