Download presentation

Presentation is loading. Please wait.

Published byAnnika Brent Modified over 3 years ago

1
© 1997 Prentice-Hall, Inc. S2 - 1 Principles of Operations Management Forecasting Chapter S2

2
© 1997 Prentice-Hall, Inc. S2 - 2 Learning Objectives n Define forecasting n Describe types of forecasts n Describe time series n Use time series forecasting methods n Use causal forecasting methods n Explain how to monitor & control forecasts

3
© 1997 Prentice-Hall, Inc. S2 - 3 What Is Forecasting? n Process of predicting a future event n Underlying basis of all business decisions l Production l Inventory l Personnel l Facilities Sales will be $200 Million!

4
© 1997 Prentice-Hall, Inc. S2 - 4 Types of Forecasts by Time Horizon n Short-range forecast l Up to 1 year; usually < 3 months l Job scheduling, worker assignments n Medium-range forecast l 3 months to 3 years l Sales & production planning, budgeting n Long-range forecast l 3+ years l New product planning, facility location

5
© 1997 Prentice-Hall, Inc. S2 - 5 Types of Forecasts by Item Forecast n Economic forecasts l Address business cycle l e.g., inflation rate, money supply etc. n Technological forecasts l Predict technological change l Predict new product sales n Demand forecasts l Predict existing product sales

6
© 1997 Prentice-Hall, Inc. S2 - 6 n Used when situation is ‘stable’ & historical data exist l Existing products l Current technology n Involves mathematical techniques n e.g., forecasting sales of color televisions Quantitative Methods Forecasting Approaches n Used when situation is vague & little data exist l New products l New technology n Involves intuition, experience n e.g., forecasting sales on Internet Qualitative Methods

7
© 1997 Prentice-Hall, Inc. S2 - 7 Qualitative Forecasting Methods

8
© 1997 Prentice-Hall, Inc. S2 - 8 Naive Approach n Assumes demand in next period is the same as demand in most recent period n e.g., If May sales were 48, then June sales will be 48 n Sometimes cost effective & efficient © 1995 Corel Corp.

9
© 1997 Prentice-Hall, Inc. S2 - 9 Jury of Executive Opinion n Involves small group of high-level managers l Group estimates demand by working together n Combines managerial experience with statistical models n Relatively quick n ‘Group-think’ disadvantage © 1995 Corel Corp.

10
© 1997 Prentice-Hall, Inc. S2 - 10 Delphi Method n Iterative group process n 3 types of people l Decision makers l Staff l Respondents n Reduces ‘group- think’ Decision Makers (Sales?) (Sales will be 50!) Respondents (Sales will be 45, 50, 55) Staff (What will sales be? survey)

11
© 1997 Prentice-Hall, Inc. S2 - 11 Sales Force Composite n Each salesperson projects their sales n Combined at district & national levels n Sales rep’s know customers’ wants n Tends to be overly optimistic Sales © 1995 Corel Corp.

12
© 1997 Prentice-Hall, Inc. S2 - 12 Consumer Market Survey n Ask customers about purchasing plans n What consumers say, & what they actually do are often different n Sometimes difficult to answer How many hours will you use the Internet next week? © 1995 Corel Corp.

13
© 1997 Prentice-Hall, Inc. S2 - 13 Causal Models Quantitative Forecasting Methods Quantitative Forecasting Time Series Models Linear Regression Exponential Smoothing Trend Projection Moving Average

14
© 1997 Prentice-Hall, Inc. S2 - 14 What’s a Time Series? n Set of evenly spaced numerical data l Obtained by observing response variable at regular time periods n Forecast based only on past values l Assumes that factors influencing past, present, & future will continue n Example Year:19931994199519961997 Sales:78.763.589.793.292.1

15
© 1997 Prentice-Hall, Inc. S2 - 15 Trend Component n Persistent, overall upward or downward pattern n Due to population, technology etc. n Several years duration Mo., Qtr., Yr. Response © 1984-1994 T/Maker Co.

16
© 1997 Prentice-Hall, Inc. S2 - 16 Cyclical Component n Repeating up & down movements n Due to interactions of factors influencing economy n Usually 2-10 years duration Mo., Qtr., Yr. Response Cycle B

17
© 1997 Prentice-Hall, Inc. S2 - 17 Seasonal Component n Regular pattern of up & down fluctuations n Due to weather, customs etc. n Occurs within 1 year Mo., Qtr. Response Summer © 1984-1994 T/Maker Co.

18
© 1997 Prentice-Hall, Inc. S2 - 18 Random Component n Erratic, unsystematic, ‘residual’ fluctuations n Due to random variation or unforeseen events l Union strike l Tornado n Short duration & nonrepeating © 1984-1994 T/Maker Co.

19
© 1997 Prentice-Hall, Inc. S2 - 19 Moving Average Method n MA is a series of arithmetic means n Used if little or no trend n Used often for smoothing l Provides overall impression of data over time n Equation MA n n Demand in Previous Periods Periods

20
© 1997 Prentice-Hall, Inc. S2 - 20 You’re manager of a museum store that sells historical replicas. You want to forecast sales (000) for 1998 using a 3-period moving average. 19934 1994 6 19955 19963 19977 Moving Average Example © 1995 Corel Corp.

21
© 1997 Prentice-Hall, Inc. S2 - 21 Time Response Y i Moving Total (n = 3) Moving Avg. ( n = 3) 19934NANA 19946NANA 19955NANA 19963 4 + 6 + 5 = 15 15/3 = 5.0 19977 6 + 5 + 3 = 14 14/3 = 4.7 1998NA 5 + 3 + 7 = 15 15/3 = 5.0 Moving Average Solution

22
© 1997 Prentice-Hall, Inc. S2 - 22 Moving Average Graph Year Sales 0 2 4 6 8 939495969798 Actual Forecast

23
© 1997 Prentice-Hall, Inc. S2 - 23 Moving Average Thinking Challenge You work for Firestone Tire. You want to forecast sales using a 3-period moving average. 199320,000 1994 24,000 199522,000 199626,000 199725,000 AloneGroupClass

24
© 1997 Prentice-Hall, Inc. S2 - 24 Moving Average Solution* YearSalesMA(3) YearSalesMA(3) 199320,000 1994 24,000 199522,000 199626,000 199725,000 1998NA

25
© 1997 Prentice-Hall, Inc. S2 - 25 Disadvantages of Moving Averages n Increasing n makes forecast less sensitive to changes n Do not forecast trend well n Require much historical data © 1984-1994 T/Maker Co.

26
© 1997 Prentice-Hall, Inc. S2 - 26 Exponential Smoothing Method n Form of weighted moving average l Weights decline exponentially l Most recent data weighted most Requires smoothing constant ( ) Requires smoothing constant ( ) l Ranges from 0 to 1 l Subjectively chosen n Involves little record keeping of past data

27
© 1997 Prentice-Hall, Inc. S2 - 27 Exponential Smoothing Equations n F t = Forecast value next period n F t-1 = Forecast value last period n A t-1 = Actual value last period = Smoothing constant = Smoothing constant F t = F t-1 + ·(A t-1 - F t-1 ) F t = F t-1 + ·(A t-1 - F t-1 )

28
© 1997 Prentice-Hall, Inc. S2 - 28 You’re organizing a Kwanza meeting. You want to forecast attendance for 1998 using exponential smoothing ( =.10). The 1993 forecast was 175. 1993180 1994 168 1995159 1996175 1997190 Exponential Smoothing Example © 1995 Corel Corp.

29
© 1997 Prentice-Hall, Inc. S2 - 29 Exponential Smoothing Solution F t = F t-1 + · (A t-1 - F t-1 ) TimeActual Forecast,F t ( =.10) 1993180 175.00 (Given) 1994168 175.00 +.10(180 - 175.00) = 175.50 1995159 1996175 1997190 1998NA

30
© 1997 Prentice-Hall, Inc. S2 - 30 Exponential Smoothing Solution F t = F t-1 + · (A t-1 - F t-1 ) TimeActual Forecast,F t ( =.10) 1993180 175.00 (Given) 1994168 175.00 +.10(180 - 175.00) = 175.50 1995159 175.50 +.10(168 - 175.50) = 174.75 1996175 174.75 +.10(159 - 174.75) = 173.18 1997190 173.18 +.10(175 - 173.18) = 173.36 1998NA 173.36 +.10(190 - 173.36) = 175.02

31
© 1997 Prentice-Hall, Inc. S2 - 31 Exponential Smoothing Graph Year Sales 140 150 160 170 180 190 939495969798 Actual Forecast

32
© 1997 Prentice-Hall, Inc. S2 - 32 Exponential Smoothing Thinking Challenge You’re an economist for GM. You want to forecast next year’s car sales. You decide to use exponential smoothing with =.25. Yearly sales (million units) in order are 2, 4, 1, 3. Assume that the first year’s forecast was 1. © 1995 Corel Corp. AloneGroupClass

33
© 1997 Prentice-Hall, Inc. S2 - 33 1.F 1 = 1.00 2.F 2 = 3.F 3 = 4.F 4 = 5.F 5 = Exponential Smoothing Solution*

34
© 1997 Prentice-Hall, Inc. S2 - 34 Linear Trend Projection n Used for forecasting linear trend line n Assumes relationship between response variable Y & time X is a linear function n Estimated by least squares method l Minimizes sum of squared errors

35
© 1997 Prentice-Hall, Inc. S2 - 35 Y X Linear Regression Model Observed value YabX ii YabX ii Error Error Regression line

36
© 1997 Prentice-Hall, Inc. S2 - 36 CorrelationCorrelation n Answers ‘how strong is the linear relationship between 2 variables?’ n Coefficient of correlation used l Sample correlation coefficient denoted r l Values range from -1 to +1 l Measures degree of association n Used mainly for understanding

37
© 1997 Prentice-Hall, Inc. S2 - 37 Coefficient of Correlation Values +1.00 Perfect Positive Correlation Increasing degree of negative correlation -.5+.5 Perfect Negative Correlation No Correlation Increasing degree of positive correlation

38
© 1997 Prentice-Hall, Inc. S2 - 38 Guidelines for Selecting Forecasting Model n No pattern or direction in forecast error l Error = (Y i - Y i ) = (Actual - Forecast) l Seen in plots of errors over time n Smallest forecast error l Mean square error (MSE) l Mean absolute deviation (MAD) ^^

39
© 1997 Prentice-Hall, Inc. S2 - 39 Pattern of Forecast Error Trend Not Fully Accounted for Desired Pattern Time (Years) ErrorError 00 Error 0

40
© 1997 Prentice-Hall, Inc. S2 - 40 Forecast Error Equations n Mean Square Error (MSE) n Mean Absolute Deviation (MAD)

41
© 1997 Prentice-Hall, Inc. S2 - 41 Selecting Forecasting Model Example You’re a marketing analyst for Hasbro Toys. You’ve forecast sales with a linear model & expo. smoothing. Which model do you use? ActualLinear ModelExpo Smooth YearSalesForecastForecast (.9) 199210.61.0 199311.31.0 199422.01.9 199522.72.0 199643.43.8

42
© 1997 Prentice-Hall, Inc. S2 - 42 Year ^ Y i Y i ^ 199210.6 0.4 0.40.160.4 199311.3-0.30.090.3 199422.0 0.0 0.00.000.0 199522.7-0.70.490.7 199643.4 0.6 0.60.360.6 Total0.01.102.0 Linear Model Evaluation MSE = Error 2 / n = 1.10 / 5 =.220 MAD = |Error| / n = 2.0 / 5 =.400 Error Error 2 |Error|

43
© 1997 Prentice-Hall, Inc. S2 - 43 Exponential Smoothing Model Evaluation Year Y i Y i 199211.00.00.000.0 199311.00.00.000.0 199421.90.10.010.1 199522.00.00.000.0 199643.80.20.040.2 Total0.30.050.3 ^ MSE = Error 2 / n = 0.05 / 5 = 0.01 MAD = |Error| / n = 0.3 / 5 = 0.06 Error Error 2 |Error|

44
© 1997 Prentice-Hall, Inc. S2 - 44 Tracking Signal n Measures how well forecast is predicting actual values n Ratio of running sum of forecast errors (RSFE) to mean absolute deviation (MAD) l Good tracking signal has low values n Should be within upper & lower control limits

45
© 1997 Prentice-Hall, Inc. S2 - 45 Tracking Signal Equation

46
© 1997 Prentice-Hall, Inc. S2 - 46 Tracking Signal Computation*

47
© 1997 Prentice-Hall, Inc. S2 - 47 Tracking Signal Plot

48
© 1997 Prentice-Hall, Inc. S2 - 48 ConclusionConclusion n Defined forecasting n Described types of forecasts n Described time series n Used time series forecasting methods n Used causal forecasting methods n Explained how to monitor & control forecasts

Similar presentations

Presentation is loading. Please wait....

OK

Operations Management Forecasting Chapter 4

Operations Management Forecasting Chapter 4

© 2018 SlidePlayer.com Inc.

All rights reserved.

Ads by Google

Ppt on forward rate agreement quote Ppt on data collection methods for quantitative research Ppt on international quality certification Ppt on disk formatting in windows Ppt on point contact diode symbol Animated ppt on magnetism worksheets Ppt on low level language pdf Download ppt on pulse code modulation math Ppt on astronomy and astrophysics encyclopedia Ppt on two point perspective examples