Académique Documents
Professionnel Documents
Culture Documents
FORECAST IS DIFFICULT
Prediction is very difficult, especially if it's about the future -Niels Bohr" It is often said there are two types of forecasts ... lucky or wrong!!!! -in "Control" magazine published by Institute of Operations Management"Forecasting is the art of saying what will happen, and then explaining why it didn't! -Anonymous (communicated by Balaji Rajagopalan)-
FORECASTING
Prediction, projection, estimation
occurrence of uncertain future events or levels activity
What? Aggregat? Product? Item? Sparepart? Where? Which geographic area should include? When? Forecast horizon? How? Divide forecast into period? Frequency to update? Accuracy needed?
COST
RESPONSE
SIMPLICITY
PRINCIPLES OF FORECASTING
Forecast involves error >>> they are usually wrong Family forecast are more accurate than item forecast. Aggregate forecasts are more accurate. Short-range forecasts are more accurate than long-range forecasts A good forecast is more than a single number
5. Technological innovations
DEMAND MANAGEMENT
Where possible, calculate demand rather than forecast. If not possible...
Independent Demand
(finished goods and spare parts)
Dependent Demand
(components)
B(4)
C(2)
D(2)
E(1)
D(3)
F(2)
Demand Estimates
Sales Forecast
Time Span
Long-Range
Years
MediumRange
Months
Short-Range
Weeks
FORECAST METHODS
QUALITATIVE
Sales force composite Market Research Management Decision (Jury of executive opinion) The Delphi Method
QUANTITATIVE
Time series
Ft 1 Ft 1
1 ( Dt Dt 1 Dt 1n ) n 1 t Di n i t 1n
Note that the n past observations are equally weighted. Issues with moving average forecasts:
All n past observations treated equally; Observations older than n are not included at all; Requires that n past observations be retained; Problem when 1000's of items are being forecast.
Ft 1 ( wt Dt wt 1Dt 1 wt 1n Dt 1n )
Re-do the above example, using 3 past data, each given a weight of 0.5, 0.3, and 0.2 (larger for more recent data)
Exponential Smoothing
New Forecast = (current observation of demand) + (1-) (last forecast) Or Ft = (Dt-1) + (1-)Ft-1
And
Ft-1 = (Dt-2) + (1-)Ft-2, dst Sehingga pada model ini, semua data historis terwakili pada forecast terakhir dengan bobot yang semakin kecil (untuk data yang semakin lama)
Exponential Smoothing
Include all past observations Weight recent observations much more heavily than very old observations:
weight
0 1
Decreasing weight given to older observations
(1 ) (1 ) 2 (1 )
today
3
Exponential Smoothing
Notes:
Only 2 values (Dt and Ft-1 ) are required, compared with n for moving average Parameter determined empirically (whatever works best) Rule of thumb: < 0.5 = 0.1 to = 0.3
Ft k Ft
Exponential Smoothing
Example:
Exponential smoothing and a constant model are being used for forecasting. The smoothed average at the end of period zero was 80. The actual demand in period 1 was 104. The smoothing constant is 0,1. What is the forecast for period 2 made at the end of period 1?
Persamaan MA dan ES
Sama-sama mengasumsikan demand bersifat stationary Keduanya tergantung pada 1 nilai parameter, N pada MA dan pada ES. Kalau ada trend, kedua-duanya terlambat dalam merespon
Perbedaan MA dan ES
MA mengakomodasikan lebih banyak data ES hanya menyimpan dua data: forecast terakhir dan actual demand terakhir, sedang MA menyimpan N data demand terakhir