chapter 4 - forecasting mr. david p. blain. c.q.e. management department unlv
TRANSCRIPT
Chapter 4 - Forecasting
Mr. David P. Blain. C.Q.E.
Management Department
UNLV
(Principles of Operations Management, Heizer & Render, 5th Edition)
Outline
Global Company Profile: TupperwareWhat is Forecasting?Types of ForecastsSeven Steps in the Forecasting SystemForecasting Approaches
Overview of Qualitative & Quantitative MethodsTime-Series ForecastingMonitoring and Controlling Forecasts
(Principles of Operations Management, Heizer & Render, 5th Edition)
Forecasting at Tupperware
Each of 50 profit centers around the world is responsible for computerized monthly, quarterly, and 12-month sales projections
These projections are aggregated by region, then globally, at Tupperware’s World Headquarters
Tupperware uses techniques discussed in text
(Principles of Operations Management, Heizer & Render, 5th Edition)
Three Key Factors for Tupperware
The number of registered “consultants” or sales representatives
The percentage of currently “active” dealers (this number changes each week and month)
Sales per active dealer, on a weekly basis
(Principles of Operations Management, Heizer & Render, 5th Edition)
Tupperware - Forecast by Consensus
Although inputs come from sales, marketing, finance, and production, final forecasts are the consensus of all participating managers.
The final step is Tupperware’s version of the “jury of executive opinion”
(Principles of Operations Management, Heizer & Render, 5th Edition)
What is Forecasting?
Process of predicting a future event
Underlying basis of all business decisions Production Inventory Personnel Facilities
Sales will be $200 Million!
(Principles of Operations Management, Heizer & Render, 5th Edition)
Forecasting Time Horizons
Process of predicting a future eventShort-range forecast
Job scheduling, worker assignments
Medium-range forecast Sales & production planning, budgeting
Long-range forecast New product planning, facility location
(Principles of Operations Management, Heizer & Render, 5th Edition)
Types of Forecasts
Economic forecasts Address business cycle, e.g., inflation rate,
money supply, etc.Technological forecasts
Predict technological change Predict new product sales
Demand forecasts Predict existing product sales
(Principles of Operations Management, Heizer & Render, 5th Edition)
Forecasting Approaches
Used when situation is stable & historical data existExisting productsCurrent technology
Involves mathematical techniquese.g., forecasting sales of
color televisions
Quantitative Methods Used when situation is
vague & little data existNew productsNew technology
Involves intuition, experiencee.g., forecasting sales
on Internet
Qualitative Methods
(Principles of Operations Management, Heizer & Render, 5th Edition)
Overview of Qualitative Methods
Jury of executive opinion Pool opinions of high-level executives, sometimes
augment by statistical modelsSales force composite
Estimates from individual salespersons are reviewed for reasonableness, then aggregated
Delphi method Panel of experts, queried iteratively
Consumer Market Survey Ask the customer
(Principles of Operations Management, Heizer & Render, 5th Edition)
Overview of Quantitative Approaches
Naïve approachMoving averagesExponential smoothingTrend projection
Linear regression
Time-series models
Associative models
(Principles of Operations Management, Heizer & Render, 5th Edition)
TrendTrend
SeasonalSeasonal
CyclicalCyclical
RandomRandom
Time Series Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
Naive Approach
Assumes demand in next period is the same as demand in most recent period If May sales were 48, then June sales will
be 48Sometimes can be cost effective &
efficient
(Principles of Operations Management, Heizer & Render, 5th Edition)
Moving Average Method
MA is a series of arithmetic means Used if little or no trendUsed often for smoothing
Provides overall impression of data over time
Equation
MAn
n Demand in previous periods
(Principles of Operations Management, Heizer & Render, 5th Edition)
Weighted Moving Average Method
Used when trend is present Older data usually less important
Weights based on intuition Ranges between 0 & 1, & sum to 1.0
Equation
WMA =Σ((Weight for period n) () (Demand in period n))
ΣWeights
(Principles of Operations Management, Heizer & Render, 5th Edition)
Exponential Smoothing Method
Form of weighted moving average Weights decline exponentially Most recent data weighted most
Requires smoothing constant () Ranges from 0 to 1 Subjectively chosen
Involves little record keeping of past data
(Principles of Operations Management, Heizer & Render, 5th Edition)
Ft = Ft-1 + (At-1 - Ft-1) Use for computing forecast
Ft = At - 1 + (1-)At - 2 + (1- )2·At - 3
+ (1- )3At - 4 + ... + (1- )t-1·A0
Ft = Forecast value
At = Actual value
= Smoothing constant
Exponential Smoothing Equations
(Principles of Operations Management, Heizer & Render, 5th Edition)
^XY i i= +a b
Shows linear relationship between dependent & explanatory variables Example: Sales & advertising (not time)
Dependent (response) variable
Independent (explanatory) variable
SlopeY-intercept
Trend & Linear Regression Model
(Principles of Operations Management, Heizer & Render, 5th Edition)
Linear Regression Equations
Equation: ii bxaY
Slope:
xnx
yxnyxb
i
n
i
ii
n
i
Y-Intercept: xbya
(Principles of Operations Management, Heizer & Render, 5th Edition)
Slope (b) Estimated Y changes by b for each 1 unit
increase in X• If b = 2, then sales (Y) is expected to increase by 2
for each 1 unit increase in advertising (X)
Y-intercept (a) Average value of Y when X = 0
• If a = 4, then average sales (Y) is expected to be 4 when advertising (X) is 0
Interpretation of Coefficients
(Principles of Operations Management, Heizer & Render, 5th Edition)
You want to achieve: No pattern or direction in forecast error
• Error = (Yi - Yi) = (Actual - Forecast)
• Seen in plots of errors over time
Smallest forecast error• Mean square error (MSE)
• Mean absolute deviation (MAD)
Selecting a Forecasting Model
(Principles of Operations Management, Heizer & Render, 5th Edition)
Mean Square Error (MSE)
Mean Absolute Deviation (MAD)
Forecast Error Equations
2
n
1i
2ii
n
errorsforecast
n
)y(yMSE
n|errorsforecast |
n
|yy|MAD
n
iii
(Principles of Operations Management, Heizer & Render, 5th Edition)
Measures how well the forecast is predicting actual values
Ratio of running sum of forecast errors (RSFE) to mean absolute deviation (MAD) Good tracking signal has low values
Should be within upper and lower control limits
Tracking Signal
(Principles of Operations Management, Heizer & Render, 5th Edition)
Chap 4 ForecastingLearning Objectives
Ability to Identify or Define:Forecasting and types of forecastsTime horizonsApproaches to forecasts
Ability to Describe or Explain:Moving averagesExponential smoothingTrend projections and regressionMeasures of forecast accuracyTracking signal
(Principles of Operations Management, Heizer & Render, 5th Edition)
Ch 4 Forecasting
4.6, 4.7
(Principles of Operations Management, Heizer & Render, 5th Edition)
Sum 218 78 650 1474
Average 18.17 6.5
Period (x) x 2 xy
1 1 20
2 4 42
3 9 45
4 16 56
5 25 65
6 36 96
7 49119
8 64144
9 81180
10 100200
11 121231
12 144276
Monthly Sales for Telco Batteries were as follows:Month Sales (y)
January 20
February 21
March 15
April 14
May 13
June 16
July 17
August 16
September 20
October 20
November 21
December 23
Problem 4.6 Forecasting methodsCh 4
Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
Sales
0
5
10
15
20
25
1 2 3 4 5 6 7 8 9 10 11 12
Month
Dolla
rsP 4.6 a) Plotting this data:
Ch 4 Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
P 4.6 b) Forecast January Sales using each of the following :
2) 3 month Moving Average
Using the last 3 months Oct, Nov, Dec = (20 + 21 + 23)/3 =21 .33
We would forecast January at 21 based on smoothing of 3 month average
1) Naive Method
Demand in next period same as last period
Therefore January will be 23 since those were December’s sales
Ch 4 Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
3) 6-month weighted average
Using weighting of .1, .1, .1, .2, .2 and .3;
with the heavier weights applied to the most recent months
( Oct, Nov, Dec)
(0.1 * 17) + (0.1 * 18) + (0.1 * 20) + (0.2 * 20) + ( 0.2 * 21) + (0.3 * 23) = 21.33 which gives us a January forecast of 21
Ch 4 Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
4) Exponential smoothing
sophisticated weighted moving average forecasting method New forecast = last periods forecast + alpha *
(last period's actual demand - last period's forecast)
Therefore with alpha of .3 and September’s forecast of 18
October forecast = Sept forecast + alpha times (Sept. actual - Sept. forecast)
October forecast = 18 + 0.3 (20-18) = 18.6November forecast = 18.6 + 0.3 (20-18.6) = 19.02December forecast = 19.02 + 0.3 (21-19.02) = 19.6January forecast = 19.6 + 0.3 (23-19.6) = 20.6
Ch 4 Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
5) Trend projection is a forecasting method that fits a trend line to a series of historical data points and then projects the line into the future for forecasting:
Ŷ = a + b*x (equation of the line)
Calculate slope and y-intecept
b = ∑xy – n x y a = y – bx
∑x² - nx²∑ = summation sign
Ŷ = (“y hat”) computed value of the variable to be predicted
(dependant variable)
a = y-axis intercept
b = slope of regression line (rate of change in y given change in x)
x = the independent variable
X = known values of the independent variable
Y = known values of the dependant variable
x = average of value of x’s
y = average of value of y’s
n = number of data points or observations
Ch 4 Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
5) Trend projection: Based on earlier data supplied: ∑ x = 78
X = 6.5∑ y = 218Ŷ = 18.2
Calculate slope
b = ∑xy – n x y
∑x² - nx²
b = 1474-(12*6.5*18.2) = 54.4 = 0.38
650 – (12*(6.5)2 ) 143
and y-intercept
a = y – bx
a = 18.2 – (0.38*6.5) = 15.73
Forecast is Ŷ = a + b*x = 15.73 + ( .38*13) = 20.76 , where January is 13th month
Ch 4 Forecasting
(Principles of Operations Management, Heizer & Render, 5th Edition)
P 4.7 Doug Moodie is the president of Garden Products Limited. Over the last 5 tears, he has asked both his vice president of marketing and his vice president of operations to provide sales forecasts. The actual sales and the forecasts are given below.
VP VP
YEAR SALES Marketing Operations
1 167,325 170,000 160,000
2 167,325 170,000 160,000
3 167,325 170,000 160,000
4 167,325 170,000 160,000
5 176,325 165,000 165,000
Using MAD which vice president is better at forecasting?
Mkt VP Oper VP
Error Error
2,675 7,325 5,362 10,322 7,464 2,536 23,268 18,268
11,325 11,325Totals 50,094 49,816
MAD VP Marketing = 50,094 /5 = 10,019
Ch 4 Forecasting
MAD VP Operations = 49,816 /5 = 9.963
Therefore, based on past data, the VP of operations has been presenting better forecasts.