Operations Management Chapter 8: Forecasting

Four successful judgment methods are as follows:

(1) sales force estimates (2) executive opinion (3) consumer market research (4) the Delphi method.

Advantages of the naive forecast include:

- It is easy for users to understand. - It can serve as an accuracy standard for other techniques

Disadvantages of the naive forecast include:

- It is not very accurate. Because the forecast just traces the actual data - It does not smooth random variations

The cyclical pattern is difficult to predict

- because it is affected by national or international events - because of a lack of demand history reflecting the stages of demand from product development to decline

Forecasts can help a manager to do all of the following:

- reduce uncertainty in planning - design a system - plan the use of a system - schedule the use of a system

The Delphi method is useful:

- when no historical data are available - when you develop long-range forecasts and technological forecasting.

Three methods for averaging are:

1. simple moving average, 2. weighted moving average 3. exponential smoothing

Disadvantage of simple moving method

All values in the average are weighted equally

SSR / SST =

Explained Variation / Total Variation

Advantage of sales force estimates

Forecasts of individual salesforce members can be combined easily to get regional or national sales totals

Advantage of simple moving method

It is easy to compute and easy to understand

Demand Patterns

The repeated observations of the demand form a pattern known as a time series.

Exponential smoothing

To use the simple or weighted moving average methods, companies must keep huge databases dating back many years; the most frequently used formal forecasting method

simple linear regression analysis

When we have only two variables (a dependent variable and an independent variable)

Regression analysis

aims to determine the extent to which the independent variable (I.V.) can explain the variation in the dependent variable (D.V.)

If the correlation coefficient is negative,

an increase in "X" will result in a decrease in "Y"

If the correlation coefficient is positive

an increase in "X" will result in an increase in "Y"

The exponential smoothing method requires

an initial forecast

Sales force estimates

are forecasts compiled from estimates made periodically by members of a company's salesforce

averaging methods

are useful if we can assume that market demands will stay fairly steady over time

CFE

cumulative sum of forecast errors

weighted moving average method

each demand has the same weight in the average, namely 1/n;is similar to the simple moving average, except that it assigns more weight to the most recent values

The value of r can range

from -1.00 ≤ r ≤ 1.00

seasonal pattern

indicates that a time series increases or decreases due to calendar or climatic changes.

Executive opinion

is a forecasting method in which the opinions and experiences of high-level managers are summarized to arrive at a single forecast

cyclical pattern

is a pattern in annual data that tends to repeat every several years; is only used when making very long-range forecasts (more than 1 year or decades); It usually results from changes in economic conditions

Delphi method

is a process of gaining consensus from a group of experts while maintaining their anonymity; It uses a series of anonymous questionnaires to achieve a consensus forecast; Responses are anonymous; No group discussion is required.

Consumer market research

is a systematic approach to determine external customer interest in a service or product by creating and testing hypotheses through data-gathering surveys

trend pattern

is a systematic increase or decrease in an average demand; is a general upward or downward movement of the data for a long period of time

naive method

is a time-series method that uses a single previous value of a time series as the basis of a forecast

random variation

is one aspect of demand that makes every forecast inaccurate

advantage of the weighted moving average method

is that the weighted moving average is more reflective of the recent observations than the simple average forecast

forecast error for a given period t

is the difference found by subtracting the forecast from actual demand

The primary difference between the seasonal pattern and the cyclical pattern:

is the duration of the repeating patterns

horizontal demand pattern

is the fluctuation of data around a constant mean; data cluster

Time-series analysis

is the forecasting technique that attempts to predict the future demand by using only historical data

Forecasting

is the process of predicting the future

dependent variable

is the variable that we wish to predict or explain

independent (explanatory) variable

is the variable used to explain the variation in the dependent variable

Qualitative methods (forecasting technique)

judgment methods

MAD

mean absolute deviation; merely measures the average of the absolute forecast errors over a series of time periods

MAPE

mean absolute percent error; a measure that relates the forecast error to the level of demand and is useful for putting forecast performance in the proper perspective

MSE

mean squared error; measures the average of the squares of the forecast errors

correlation coefficient (r)

measures a direction and strength of the relationship between the independent variable and the dependent variable

coefficient of determination (R2)

measures the proportion of variation in the dependent variable that is explained by an independent variable(s) in a regression model

The zero value of r indicates

no linear relationship exists

estimated regression model

provides an estimate of the population regression model

simple moving average method

simply involves calculating the average demand for the "n" most recent periods and using it as the forecast for future time periods

σ

standard deviation of the errors; measure the dispersion of forecast errors attributed to trend

The closer to -1

the stronger the negative linear relationship

The closer to 1

the stronger the positive linear relationship

The closer to 0

the weaker the linear relationship

Judgment methods (forecasting technique)

translate the opinions of managers, expert opinions, consumer surveys, and sales force estimates into quantitative estimates; are widely used when introducing new products, redesigning existing products, and promoting sales on internet

The ratio of SSR / SST indicates

what proportion of the total variation of y can be explained by an independent variable in a regression model

Quantitative methods (forecasting technique)

• Causal methods (e.g., Linear regression) • Time-series methods (e.g., naive forecast)

Two steps are required to forecast the seasonal demand as follows:

• Step 1: Compute a seasonal index . 1. The seasonal index is a ratio of sales to a centered average. 2. The centered average should be positioned in the middle of the periods. • Step 2: Multiply an average seasonal index by the estimate of average demand.