OM 1 NAIVE METHODS- A simple but widely used approach to

CHAPTER 3: FORECASTING forecasting.

NAIVE FORECAST- A forecast for any period that equals
Forecasting- is an important component of yield the previous periods actual value.
management, which relates to the percentage of capacity TECHNIQUES FOR AVERAGING
being used. 1. Moving average- forecast tends to smooth and lag
Forecasts in Business Organizations: changes in the data.
Accounting Marketing 2. Weighted moving average- more recent values in a series
Finance. MIS are given more weight in computing a forecast.
Human Resource. Operations 3. Exponential smoothing- a weighted averaging method
Product service/design basedvon previous forecast plus a percentage of the
Two (2) uses for forecast forecast error.
a. Plan the system- a long-range plans about the types of Techniques for trend- analysis of trend
product and services to offer, what facilities and equipment LINEAR TREND EQUATION- used to develop forecast
to have, where to locate, and so on. when trend is present.
b. Plan the use of system- a short-range and intermediate- TREND ADJUSTED EXPONENTIAL SMOOTHING-
range planning, which involves tasks. variations of exponential smoothing used when a time
 FEATURES COMMON TO ALL FORECASTS series exhibits a linear trend.
- a wide variety of forecasting techniques are in use. TECHNIQUES FOR SEASONALITY
1. Techniques generally assume Seasonal variations- regularly repeating movements in
2. Forecast are rarely perfect series values that can be tied to recurring events.
3. For groups of items tend to be more accurate than Seasonal relative- percentage of average and trend.
forecast Centered moving average- a moving average positioned at
4. Accuracy decreases as the time period covered by the the center of the data that were used to compute it.
forecast- the time horizon- increases TECHNIQUES FOR CYCLES- Cycles are up-and- down
ELEMENTS OF A GOOD FORECAST movements similar to seasonal variations.
1. Timely 2. Accurate 3. Reliable 4. Meaningful units ASSOCIATIVE FORECASTING TECHNIQUES- rely on
5. In writing 6. Simple to understand and use 7. cost- identification of related variables that can be used to predict
effective values of the variaable of interest.
STEPS IN THE FORECASTING PROCESS Predictor variables- variables that can be used to predict
1. Determine the purpose of the forecast. values of the variable of interest.
2. Establish a time horizon. Regression- technique for fitting a line to a set of points.
3. Select a forecasting technique Least squares line- minimizes the sum ofvthe squarred
4. Obtain, clean and analyze appropriate data vertical deviations around the line.
5. Make the forecast Straight line- is fitted to a set of sample points.
6. Monitor the forecast Correlation- a measure of the strength and direction of
APPROACHES TO FORECASTING relationship between two variables.
a. Qualitative methods- consist mainly of subjective ACCURACY AND CONTROL OF FORECASTS
b. Quantitative methods- projection of historical data ERROR- difference between the actual value and the value
Three (3) forecasting techniques that was predicted for a given period.
a. Judgmental forecast- use subjective inputs such as Accuracy- is based on the historical error performance of a
opinions from customer surveys, sales, staff, managers, forecast.
executive and experts. MAD- Mean absolute deviation. The average absolute
b. Time- series forecast- project patterns identified in recent forecast error.
time- series observations MSE- Mean squarred error. The average of squarred-
c. Associative models- that uses explanatory variables to forecast errors.
predict future demand MAPE- Mean absolute percent error. The average percent
error.
FORECAST BASED ON JUDGMENT AND OPINION Controlling the Forecast- many forecast are made at regular
a. Executive Opinions- a small group of upper-level intervals.
managers may meet and collectively develop a forecasts. Control chart- a visual tool for monitoring forecast errors.
b. Salesforce Opinions- Members of the sales staff Tracking signal- the ratio of cumulative forecast error to
c. Consumer surveys- consumers who ultimately the corresponding value of MAD used to monitor a
determined demand forecast.
d. Other approaches Bias Persistent- tendency for forecast to be greater or less
Delphi Method- an iterative process in which managers and thaan the actual values of a time series.
staff complete a series of questionnaires, each developed CHOOSING A FORECASTING TECHNIQUE
from the previous one, to achieve a consensus forecast Two (2) Important Factors
FORECAST BASED ON TIME- SERIES DATA 1. Cost 2. Accuracy
- A time- ordered sequence of observations taken at regular USING FORECAST INFORMATION- A manager can
intervals take a reactive or a proactive approach to a forecast.
Behaviors: COMPUTERS IN FORECASTING- play an important role
1. Trend- a long-term upward or downward movement in in preparing forecast based on quantitative data.
data OPERATION STRATEGY- Forecast are the basis for
2. Seasonality- short- term regular variations related to the many decisions..
calendar or time or day.
3. Cycles- wavelike variations lasting more than one year
4. Irregular variations- caused by unusual circumstances
not reflect of typical behavior.
5. Random variations- Residual variations after all othet
behaviors are accounted for.