1 Lecture 2 Decision Theory Chapter 5S
2 Certainty - Environment in which relevant parameters have known values Risk - Environment in which certain future events have probabilistic outcomes Uncertainty - Environment in which it is impossible to assess the likelihood of various future events Decision Environments
3 Maximin - Choose the alternative with the best of the worst possible payoffs Maximax - Choose the alternative with the best possible payoff Minimax Regret - Choose the alternative that has the least of the worst regrets Decision Making under Uncertainty
4 Payoff Table: An Example LowModerateHigh Small facility $10 Medium facility 712 Large facility Possible Future Demand Values represent payoffs (profits)
5 Maximax Solution Note: choose the “minimize the payoff” option if the numbers in the previous slide represent costs
6 Maximin Solution
7 Minimax Regret Solution
8 Decision Making Under Risk - Decision Trees State of nature 1 B Payoff 1 State of nature 2 Payoff 2 Payoff 3 2 Choose A’ 1 Choose A’ 2 Payoff 6 State of nature 2 2 Payoff 4 Payoff 5 Choose A’ 3 Choose A’ 4 State of nature 1 Choose A’ Choose A’ 2 1 Decision Point Chance Event
9 Decision Making with Probabilities Expected Value Approach Useful if probabilistic information regarding the states of nature is available Expected return for each decision is calculated by summing the products of the payoff under each state of nature and the probability of the respective state of nature occurring Decision yielding the best expected return is chosen.
10 Example: Burger Prince Burger Prince Restaurant is considering opening a new restaurant on Main Street. It has three different models, each with a different seating capacity. Burger Prince estimates that the average number of customers per hour will be 80, 100, or 120 with a probability of 0.4, 0.2, and 0.4 respectively The payoff (profit) table for the three models is as follows. s 1 = 80 s 2 = 100 s 3 = 120 Model A $10,000 $15,000 $14,000 Model B $ 8,000 $18,000 $12,000 Model C $ 6,000 $16,000 $21,000 Choose the alternative that maximizes expected payoff
11 Decision Tree d1d1d1d1 d2d2d2d2 d3d3d3d3 s1s1s1s1 s1s1s1s1 s1s1s1s1 s2s2s2s2 s3s3s3s3 s2s2s2s2 s2s2s2s2 s3s3s3s3 s3s3s3s3 Payoffs 10,000 15,000 14,000 8,000 18,000 12,000 6,000 16,000 21,
12 Management Scientist Solutions EVPI = Expected payoff under certainty – Expected payoff under risk
13 Lecture 2 Forecasting Chapter 3
14 A statement about the future value of a variable of interest such as demand. Forecasts affect decisions and activities throughout an organization Accounting, finance Human resources Marketing Operations Product / service design Forecast
15 AccountingCost/profit estimates FinanceCash flow and funding Human ResourcesHiring/recruiting/training MarketingPricing, promotion, strategy OperationsSchedules, MRP, workloads Product/service designNew products and services Uses of Forecasts
16 Elements of a Good Forecast Timely Accurate Reliable Meaningful Written Easy to use
17 Steps in the Forecasting Process Step 1 Determine purpose of forecast Step 2 Establish a time horizon Step 3 Select a forecasting technique Step 4 Gather and analyze data Step 5 Prepare the forecast Step 6 Monitor the forecast “The forecast”
18 Types of Forecasts Judgmental - uses subjective inputs Time series - uses historical data assuming the future will be like the past Associative models - uses explanatory variables to predict the future
19 Judgmental Forecasts Executive opinions Sales force opinions Consumer surveys Outside opinion Delphi method Opinions of managers and staff Achieves a consensus forecast
20 Time Series Forecasts Trend - long-term movement in data Seasonality - short-term regular variations in data Cycle – wavelike variations of more than one year’s duration Irregular variations - caused by unusual circumstances
21 Forecast Variations Trend Irregular variatio n Seasonal variations Figure 3.1 Cycles
22 Smoothing/Averaging Methods Used in cases in which the time series is fairly stable and has no significant trend, seasonal, or cyclical effects Purpose of averaging - to smooth out the irregular components of the time series. Four common smoothing/averaging methods are: Moving averages Weighted moving averages Exponential smoothing
23 n Sales of gasoline for the past 12 weeks at your local Chevron (in ‘000 gallons). If the dealer uses a 3- period moving average to forecast sales, what is the forecast for Week 13? Example of Moving Average n Past Sales Week Sales Week Sales Week Sales Week Sales
24 Management Scientist Solutions MA(3) for period 4 = ( )/3 = 19 Forecast error for period 3 = Actual – Forecast = 23 – 19 = 4
25 MA(5) versus MA(3)
26 Exponential Smoothing Premise - The most recent observations might have the highest predictive value. Therefore, we should give more weight to the more recent time periods when forecasting.
27 Exponential Smoothing Weighted averaging method based on previous forecast plus a percentage of the forecast error A-F is the error term, is the % feedback F t+1 = F t + ( A t - F t )
28 Picking a Smoothing Constant .1 .4 Actual
29 Linear Trend Equation F t = Forecast for period t t = Specified number of time periods a = Value of F t at t = 0 b = Slope of the line F t = a + bt t FtFt a Suitable for time series data that exhibit a long term linear trend
30 Linear Trend Example F11 = (11) = 32.5 Linear trend equation Sale increases every time 1.1 units
31 Actual vs Forecast Linear Trend Example Week Actual/Forecasted sales Actual Forecast F(t) = t
32 Forecasting with Trends and Seasonal Components – An Example Business at Terry's Tie Shop can be viewed as falling into three distinct seasons: (1) Christmas (November-December); (2) Father's Day (late May - mid-June); and (3) all other times. Average weekly sales ($) during each of the three seasons during the past four years are known and given below. Determine a forecast for the average weekly sales in year 5 for each of the three seasons. Year Season
33 Management Scientist Solutions
34 Interpretation of Seasonal Indices Seasonal index for season 2 (Father’s Day) = Means that the sale value of ties during season 2 is 23.6% higher than the average sale value over the year Seasonal index for season 3 (all other times) = Means that the sale value of ties during season 3 is 41.4% lower than the average sale value over the year
35 Associative Forecasting Predictor variables - used to predict values of variable interest Regression - technique for fitting a line to a set of points Least squares line - minimizes sum of squared deviations around the line
36 Regression Analysis – An Example Home-Size (Square feet)Price 600$72, $116, $152, $80, $141, $124, $117, $165, $153, $126, $122, $140, $223, $99, $211, $121, $169, $156, $123, $136, $194, $128, $302, $142, $146, $180, $126, $139,500 Linear model seems reasonable A straight line is fitted to a set of sample points
37 Regression Results Use MS-Excel macro Template posted at class website y = x Price = (Square footage) Forecast price of a 2000 square feet house y = (2000) = $157,272.78
38 Forecast Accuracy Error - difference between actual value and predicted value Mean Absolute Deviation (MAD) Average absolute error Mean Squared Error (MSE) Average of squared error
39 MAD and MSE MAD = Actualforecast n MSE = Actualforecast ) 2 n (
40 Measure of Forecast Accuracy MSE = Mean Squared Error
41 Forecasting Accuracy Estimates Example 10 of textbook
42 Sources of Forecast errors Model may be inadequate Irregular variations Incorrect use of forecasting technique
43 Characteristics of Forecasts They are usually wrong A good forecast is more than a single number Aggregate forecasts are more accurate The longer the forecast horizon, the less accurate the forecast will be Forecasts should not be used to the exclusion of known information
44 Choosing a Forecasting Technique No single technique works in every situation Two most important factors Cost Accuracy Other factors include the availability of: Historical data Computers Time needed to gather and analyze the data Forecast horizon