Demand Management and FORECASTING Operations Management Dr. Ron Lembke
Demand Management Coordinate sources of demand for supply chain to run efficiently, deliver on time Independent Demand ▫Things demanded by end users Dependent Demand ▫Demand known, once demand for end items is known
Affecting Demand Increasing demand ▫Marketing campaigns ▫Sales force efforts, cut prices Changing Timing of demand ▫Incentives for earlier or later delivery ▫At capacity, don’t actively pursue more
Predicting the Future We know the forecast will be wrong. Try to make the best forecast we can, ▫Given the time we want to invest ▫Given the available data The “Rules” of Forecasting: 1.The forecast will always be wrong 2.The farther out you are, the worse your forecast is likely to be. 3.Aggregate forecasts are more likely to accurate than individual item ones
Time Horizons Different decisions require projections about different time periods: Short-range: who works when, what to make each day (weeks to months) Medium-range: when to hire, lay off (months to years) Long-range: where to build plants, enter new markets, products (years to decades)
Forecast Impact Finance & Accounting: budget planning Human Resources: hiring, training, laying off employees Capacity: not enough, customers go away angry, too much, costs are too high Supply-Chain Management: bringing in new vendors takes time, and rushing it can lead to quality problems later
Qualitative Methods Sales force composite / Grass Roots Market Research / Consumer market surveys & interviews Jury of Executive Opinion / Panel Consensus Delphi Method Historical Analogy - DVDs like VCRs Naïve approach
Quantitative Methods Time Series Methods 0.All-Time Average 1. Simple Moving Average 2. Weighted Moving Average 3. Exponential Smoothing 4. Exponential smoothing with trend 5. Linear regression Causal Methods Linear Regression
Time Series Forecasting Assume patterns in data will continue, including: Trend (T) Seasonality (S) Cycles (C) Random Variations
All-Time Average To forecast next period, take the average of all previous periods Advantages: Simple to use Disadvantages: Ends up with a lot of data Gives equal importance to very old data
4/7/ Farm Angels: Ty: 1.000, Jacob 0.833, Noah (6 at bats)
End of 2008 season
Moving Average Compute forecast using n most recent periods Jan Feb MarAprMayJunJul 3 month Moving Avg: June forecast: F Jun = (A Mar + A Apr + A May )/3 If no seasonality, freedom to choose n If seasonality is N periods, must use N, 2N, 3N etc. number of periods
Moving Average Advantages: ▫ Ignores data that is “too” old ▫ Requires less data than simple average ▫ More responsive than simple average Disadvantages: ▫ Still lacks behind trend like simple average, (though not as badly) ▫ The larger n is, more smoothing, but the more it will lag ▫ The smaller n is, the more over-reaction
Simple and Moving Averages
Centered MA CMA smoothes out variability Plot the average of 5 periods: 2 previous, the current, and the next two Obviously, this is only in hindsight FRB Dalls graphs
Centered Moving Average Take average of n periods, Plot the average in the middle period Not useful for forecasting More stable than actuals If seasonality, n = season length (4wks, 12 mo, etc.)
CMA - # Periods to Average What if data has 12-month cycle? Ja F M Ap My Jn Jl Au S O N D Ja F M Avg of Jan-Dec gives average of month 6.5: ( )/12=6.5 Avg of Feb-Jan gives average of month 6.5: ( )/12=7.5 How get a July average? Average of other two averages
Stability vs. Responsiveness Responsive ▫Real-time accuracy ▫Market conditions Stable ▫Forecasts being used throughout the company ▫Long-term decisions based on forecasts ▫Don’t whipsaw those folks
Centered Moving Average To center even-number of periods 12: take half each of 1 and 13, plus sum of F14 = 0.5 A1 + A2 + A3 + A4 + A5 + A6 + A7 + A8 + A9 + A10 + A11 + A A13 This is exactly the same as what you get by taking the average of the averages from previous slide
Old Data Comparison of simple, moving averages clearly shows that getting rid of old data makes forecast respond to trends faster Moving average still lags the trend, but it suggests to us we give newer data more weight, older data less weight.
Weighted Moving Average F Jun = (A Mar + A Apr + A May )/3 = (3A Mar + 3A Apr + 3A May )/9 Why not consider: F Jun = (2A Mar + 3A Apr + 4A May )/9 F Jun = 2/9 A Mar + 3/9 A Apr + 4/9 A May F t = w 1 A t-3 + w 2 A t-2 + w 3 A t-1 Complicated: Have to decide number of periods, and weights for each Weights have to add up to 1.0 Most recent probably most relevant, gets most weight Carry around n periods of data to make new forecast
Weighted Moving Average Wts = 0.5, 0.3, 0.2
Setting Parameters Weighted Moving Average ▫Number of Periods ▫Individual weights Trial and Error ▫Evaluate performance of forecast based on some metric
Exponential Smoothing A t-1 Actual demand in period t-1 F t-1 Forecast for period t-1 Smoothing constant >0, <1 Forecast is old forecast plus a portion of the error of the last forecast. Formulas are equivalent, give same answer F 10 = F (A 9 - F 9 ) F 10 = 0.8 F (A 9 - F 9 )
Exponential Smoothing Smoothing Constant between Easier to compute than moving average Most widely used forecasting method, because of its easy use F 1 = 1,050, = 0.05, A 1 = 1,000 F 2 = F1 + (A 1 - F 1 ) = 1, (1,000 – 1,050) = 1, (-50) = 1,047.5 units BTW, we have to make a starting forecast to get started. Often, use actual A1
Exponential Smoothing Alpha = 0.3
Exponential Smoothing Alpha = 0.5
Exponential Smoothing We take: And substitute in to get: and if we continue doing this, we get: Older demands get exponentially less weight
Choosing Low : if demand is stable, we don’t want to get thrown into a wild-goose chase, over-reacting to “trends” that are really just short-term variation High : If demand really is changing rapidly, we want to react as quickly as possible
Averaging Methods Simple Average Moving Average Weighted Moving Average Exponentially Weighted Moving Average (Exponential Smoothing) They ALL take an average of the past ▫With a trend, all do badly ▫Average must be in-between
Trend-Adjusted Ex. Smoothing
Forecast including trend for period 1 is Suppose actual demand is 115, A 1 =115
Trend-Adjusted Ex. Smoothing Forecast including trend for period 1 is Suppose actual demand is 120, A 2 =120
F5F5 FIT 5 =F 5 +T 5 A5A5 F6F6
Selecting and You could: ▫Try an initial value for each parameter. ▫Try lots of combinations and see what looks best. ▫But how do we decide “what looks best?” Let’s measure the amount of forecast error. Then, try lots of combinations of parameters in a methodical way. ▫Let = 0 to 1, increasing by 0.1 For each value, try = 0 to 1, increasing by 0.1
Evaluating Forecasts How far off is the forecast? What do we do with this information? Forecasts Demands
Measuring the Errors PeriodA-F Method 1 A-F Method RSFE0100 Method 1 forecasts are low, high, etc. Method 2 forecasts always too low. Running Sum of Forecast Errors, RSFE ▫Sum of all periods ▫Also known as the Bias
Evaluating Forecasts Mean Absolute Deviation Mean Squared Error Mean Absolute Percent Error
MAD of examples Period|A-F| Method 1 |A-F| Method MAD10010 MAD shows that method 1 is off by a larger amount Method 2 was biased However, overall, Method 2 seems preferable
Tracking Signal To monitor, compute tracking signal If >4 or <-4 something is wrong Top should sum to 0 over time. If not, forecast is biased.
Monitoring Forecast Accuracy Monitor forecast error each period, to see if it becomes too great Forecast Error Forecast Period Lower Limit Upper Limit
Updating MAD Simplified calculation avoids keeping running total of all errors and demands: Standard Deviation can be estimated from MAD:
Techniques for Trend Determine how demand increases as a function of time t = periods since beginning of data b = Slope of the line a = Value of y t at t = 0
Computing Values
Linear Regression Four methods 1.Type in formulas for trend, intercept 2.Tools | Data Analysis | Regression 3.Graph, and R click on data, add a trendline, and display the equation. 4.Use intercept(Y,X), slope(Y,X) and RSQ(Y,X) commands Fits a trend and intercept to the data. R 2 measures the percentage of change in y that can be explained by changes in x. Gives all data equal weight. Exp. smoothing with a trend gives more weight to recent, less to old.
Causal Forecasting Linear regression seeks a linear relationship between the input variable and the output quantity. For example, furniture sales correlates to housing sales Not easy, multiple sources of error: ▫Understand and quantify relationship ▫Someone else has to forecast the x values for you
Video sales of Shrek 2? Shrek did $500m at the box office, and sold almost 50 million DVDs & videos Shrek2 did $920m at the box office
Video sales of Shrek 2? Assume 1-1 ratio: ▫920/500 = 1.84 ▫1.84 * 50 million = 92 million videos? ▫Fortunately, not that dumb. January 3, 2005: 37 million sold! March analyst call: 40m by end Q1 March SEC filing: 33.7 million sold. Oops. May 10 Announcement: ▫In 2 nd public Q, missed earnings targets by 25%. ▫May 9, word started leaking ▫Stock dropped 16.7%
Lessons Learned Flooded market with DVDs Guaranteed Sales ▫Promised the retailer they would sell them, or else the retailer could return them ▫Didn’t know how many would come back 5 years ago ▫Typical movie 30% of sales in first week ▫Animated movies even lower than that 2004/ % in first week ▫ Shrek 2: 12.1m in first 3 days ▫American Idol ending, had to vote in first week
The Human Element Colbert says you have more nerve endings in your gut than in your brain Limited ability to include factors ▫Can’t include everything If it feels really wrong to your gut, maybe your gut is right
Washoe Gaming Win, What did they mean when they said it was down three quarters in a row?
Seasonality Seasonality is regular up or down movements in the data Can be hourly, daily, weekly, yearly Naïve method ▫N1: Assume January sales will be same as December ▫N2: Assume this Friday’s ticket sales will be same as last
Seasonal Factors Seasonal factor for May is 1.20, means May sales are typically 20% above the average Factor for July is 0.90, meaning July sales are typically 10% below the average
Seasonality & No Trend SalesFactor Spring200200/250 = 0.8 Summer350350/250 = 1.4 Fall300300/250 = 1.2 Winter150150/250 = 0.6 Total1,000 Avg1,000/4=250
Seasonality & No Trend If we expected total demand for the next year to be 1,100, the average per quarter would be 1,100/4=275 Forecast Spring275 * 0.8 = 220 Summer275 * 1.4 = 385 Fall275 * 1.2 = 330 Winter275 * 0.6 = 165 Total1,100
Trend & Seasonality Deseasonalize to find the trend 1.Calculate seasonal factors 2.Deseasonalize the demand 3.Find trend of deseasonalized line Project trend into the future 4.Project trend line into future 5.Multiply trend line by seasonal component.
Washoe Gaming Win, Looks like a downhill slide -Silver Legacy opened 95Q3 -Otherwise, upward trend Source: Comstock Bank, Survey of Nevada Business & Economics
Washoe Win Definitely a general upward trend, slowed 93-94
Cache Creek Thunder Valley CC Expands 9/11
DateQuarterWin , , , , , , , , , , , , , , , , ,734 QAvgIndex 1 240, , , , Total Avg. 262,382 For each Q: Compute Indexes Deseasonalize: Divide Win by Index 276,371 / = 250,755 Compute Avg Win for each Q Divide Avg by Total Avg to get Index: 240,562/262,382 =
periodWinDeseasonalized , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,343 Do LR on deseasonalized data intercept 185, slope 1, rsq Create Linear Forecasts Int + slope * period Linear 251, , , , , , , , , , , , , , , , , , , , ,011 A LR on the original, seasonal data gives an R 2 value of 0.045!
Seasonal Forecast ,062 DeseasonalizedLinearForecast , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,596 Multiply Linear forecast by indexes 251,613 * = 277, ,291 * = 245,063 QIndex
Q Win
How Good Was It?
Summary 1.Calculate indexes 2.Deseasonalize 1.Divide actual demands by seasonal indexes 3.Do a LR 4.Project the LR into the future 5.Seasonalize 1.Multiply straight-line forecast by indexes