Professional Documents
Culture Documents
Forecasting Demand
13 1
Forecasting
Forecasts are critical inputs to business plans, annual plans, and budgets Finance, human resources, marketing, operations, and supply chain managers need forecasts to plan: output levels, purchases of services and materials, workforce and output schedules, inventories, and long-term capacities
13 2
Demand Patterns
A time series is the repeated observations of demand for a service or product in their order of occurrence There are five basic time series patterns
Horizontal Trend Seasonal Cyclical Random
13 3
Demand Patterns
Figure 13.1 Patterns of Demand
Quantity
Demand Patterns
Figure 13.1 Patterns of Demand
Quantity
Demand Patterns
Figure 13.1 Patterns of Demand
Quantity
Year 1
Year 2 | J | F | M | A | M | | J J Months | A | S | O | N | D
13 6
Demand Patterns
Figure 13.1 Patterns of Demand
Quantity
| 1
| 2
| 3 Years
| 4
| 5
| 6
(d) Cyclical: Data reveal gradual increases and decreases over extended periods
Copyright 2010 Pearson Education, Inc.
13 7
Key Decisions
Deciding what to forecast
Level
Key factor in choosing the proper forecasting approach is the time horizon for the decision requiring forecasts
Copyright 2010 Pearson Education, Inc.
13 8
Judgment Methods
Other methods (casual and time-series) require an adequate history file, which might not be available Judgmental forecasts use contextual knowledge gained through experience Salesforce estimates
Executive opinion is a method in which opinions, experience, and technical knowledge of one or more managers are summarized to arrive at a single forecast Delphi method
13 9
Judgment Methods
Market research is a systematic approach to determine external customer interest through data-gathering surveys
Delphi method is a process of gaining consensus from a group of experts while maintaining their anonymity
Useful when no historical data are available Can be used to develop long-range forecasts and technological forecasting
13 10
Linear Regression
A dependent variable is related to one or more independent variables by a linear equation
The independent variables are assumed to cause the results observed in the past
Simple linear regression model is a straight line Y = a + bX
where Y X a b = dependent variable = independent variable = Y-intercept of the line = slope of the line
13 11
Linear Regression
Y Deviation, or error Estimate of Y from regression equation Actual value of Y
Regression equation: Y = a + bX
Dependent variable
Independent variable
Figure 13.2 Linear Regression Line Relative to Actual Data
Copyright 2010 Pearson Education, Inc.
13 12
Linear Regression
The sample correlation coefficient, r
Measures the direction and strength of the relationship between the independent variable and the dependent variable. The value of r can range from 1.00 r 1.00 Measures the amount of variation in the dependent variable about its mean that is explained by the regression line
Measures how closely the data on the dependent variable cluster around the regression line
13 13
The supply chain manager seeks a better way to forecast the demand for door hinges and believes that the demand is related to advertising expenditures. The following are sales and advertising data for the past 5 months:
Month 1 2 3 4 5 Sales (thousands of units) 264 116 165 101 209 Advertising (thousands of $) 2.5 1.3 1.4 1.0 2.0
The company will spend $1,750 next month on advertising for the product. Use linear regression to develop an equation and a forecast for this product.
Copyright 2010 Pearson Education, Inc.
13 14
We used POM for Windows to determine the best values of a, b, the correlation coefficient, the coefficient of determination, and the standard error of the estimate
a= b= r= r2 = syx = 8.135 109.229X 0.980 0.960 15.603
13 15
X X X X Forecasts X Data
1.0
Figure 13.3 Linear Regression Line for the Sales and Advertising Data
Copyright 2010 Pearson Education, Inc.
13 16
to estimate the average of a demand time series and thereby remove the effects of random fluctuation Most useful when demand has no pronounced trend or seasonal influences The stability of the demand series generally determines how many periods to include
13 17
10
15 Week
20
25
30
13 19
13 20
2
3
380
411
b. If the actual number of patient arrivals in week 4 is 415, what is the forecast error for week 4?
c. What is the forecast for week 5?
13 21
Application 13.1a
Estimating with Simple Moving Average using the following customer-arrival data
Month 1 2 3 4 Customer arrival 800 740 810 790
13 23
Application 13.1a
If the actual number of arrivals in month 5 is 805, what is the forecast for month 6?
Month 1 2 3 4 Customer arrival 800 740 810 790
13 24
Application 13.1a
Forecast error is simply the difference found by subtracting the forecast from actual demand for a given period, or Et = Dt Ft Given the three-month moving average forecast for month 5, and the number of patients that actually arrived (805), what is the forecast error?
E5 = 805 780 = 25
Forecast error for month 5 is 25
13 25
A three-period weighted moving average model with the most recent period weight of 0.50, the second most recent weight of 0.30, and the third most recent might be weight of 0.20
Ft+1 = 0.50Dt + 0.30Dt1 + 0.20Dt2
Copyright 2010 Pearson Education, Inc.
13 26
Application 13.1b
Revisiting the customer arrival data in Application 13.1a. Let W1 = 0.50, W2 = 0.30, and W3 = 0.20. Use the weighted moving average method to forecast arrivals for month 5. F5 = W1D4 + W2D3 + W3D2 = 0.50(790) + 0.30(810) + 0.20(740) = 786 Forecast for month 5 is 786 customer arrivals Given the number of patients that actually arrived (805), what is the forecast error? E5 = 805 786 = 19
Application 13.1b
If the actual number of arrivals in month 5 is 805, compute the forecast for month 6 F6 = W1D5 + W2D4 + W3D3 = 0.50(805) + 0.30(790) + 0.20(810) = 801.5
13 28
Exponential Smoothing
A sophisticated weighted moving average that calculates the average of a time series by giving recent demands more weight than earlier demands Requires only three items of data
The last periods forecast The demand for this period A smoothing parameter, alpha (), where 0 1.0
or the equivalent
Ft+1 = Ft + (Dt Ft)
Copyright 2010 Pearson Education, Inc.
13 29
Exponential Smoothing
The emphasis given to the most recent demand levels can be adjusted by changing the smoothing parameter Larger values emphasize recent levels of demand and result in forecasts more responsive to changes in the underlying average
Smaller values treat past demand more uniformly and result in more stable forecasts
Exponential smoothing is simple and requires minimal data When the underlying average is changing, results will lag actual changes
13 30
3-week MA forecast
6-week MA forecast
Patient arrivals
10
15 Week
20
25
30
13 31
3
4
411
415
13 32
a. The exponential smoothing method requires an initial forecast. Suppose that we take the demand data for the first two weeks and average them, obtaining (400 + 380)/2 = 390 as an initial forecast. (POM for Windows and OM Explorer simply use the actual demand for the first week as a default setting for the initial forecast for period 1, and do not begin tracking forecast errors until the second period). To obtain the forecast for week 4, using exponential smoothing with and the initial forecast of 390, we calculate the average at the end of week 3 as
F4 = 0.10(411) + 0.90(390) = 392.1 Thus, the forecast for week 4 would be 392 patients.
13 33
13 34
Application 13.1c
Suppose the value of the customer arrival series average in month 3 was 783 customers (let it be F4). Use exponential smoothing with = 0.20 to compute the forecast for month 5. Ft+1 = Ft + (Dt Ft) = 783 + 0.20(790 783) = 784.4
13 35
Application 13.1c
Given the actual number of arrivals in month 5, what is the forecast for month 6?
Ft+1 = Ft + (Dt Ft) = 784.4 + 0.20(805 784.4) = 788.52
13 36
Including a Trend
A trend in a time series is a systematic increase or decrease in the average of the series over time The forecast can be improved by calculating an estimate of the trend Trend-adjusted exponential smoothing requires two smoothing constants
13 37
Including a Trend
For each period, we calculate the average and the trend:
At = (Demand this period)
13 39
13 40
8
9 10 11
61
39 55 54
12
13 14 15
52
60 60 75
13 41
8
9 10 11
61
39 55 54
19.98
2.71 5.79
12
13 14 15
52
60 60 75
9.01
1.23 2.96 10.77
13 42
Trend-adjusted forecast
40
30
8 Week
9 10 11 12 13 14 15
13 43
Application 13.2
The forecaster for Canine Gourmet dog breath fresheners estimated (in March) the sales average to be 300,000 cases sold per month and the trend to be +8,000 per month. The actual sales for April were 330,000 cases. What is the forecast for May, assuming = 0.20 and = 0.10?
AApr = Dt + (1 )(AMar + TMar) = 0.20(330,000) + 0.80(300,000 + 8,000) = 312,400 cases TApr = (AApr AMar) + (1 )TMar = 0.10(312,400 300,000) + 0.90(8,000) = 8,440 cases Forecast for May = AApr + pTApr = 312,400 + (1)(8,440) = 320,840 cases
13 44
Application 13.2
Suppose you also wanted the forecast for July, three months ahead. To make forecasts for periods beyond the next period, we multiply the trend estimate by the number of additional periods that we want in the forecast and add the results to the current average.
Forecast for July = AApr + pTApr = 312,400 + (3)(8,440) = 337,720 cases
13 45
Seasonal Patterns
Seasonal patterns are regularly repeated upward or downward movements in demand measured in periods of less than one year Account for seasonal effects by using one of the techniques already described but to limit the data in the time series to those periods in the same season This approach accounts for seasonal effects but discards considerable information on past demand
13 46
13 47
The manager wants to forecast customer demand for each quarter of year 5, based on an estimate of total year 5 demand of 2,600 customers
Copyright 2010 Pearson Education, Inc.
13 48
13 49
Figure 13.6 Demand Forecasts Using the Seasonal Forecast Solver of OM Explorer
Copyright 2010 Pearson Education, Inc.
13 50
Application 13.3
Suppose the multiplicative seasonal method is being used to forecast customer demand. The actual demand and seasonal indices are shown below.
Year 1 Quarter 1 Demand 100 Index 0.40 192 Year 2 Demand Index 0.64 Average Index 0.52
2
3 4 Average
400
300 200 250
1.60
1.20 0.80
408
384 216 300
1.36
1.28 0.72
1.48
1.24 0.76
13 51
Application 13.3
If the projected demand for Year 3 is 1320 units, what is the forecast for each quarter of that year? 1320 units 4 quarters = 330 units
Quarter 1 2 3 Average Index 0.52 1.48 1.24
0.76
Forecast for Quarter 1 = 0.52(330) 172 units Forecast for Quarter 2 = 1.48(330) 488 units Forecast for Quarter 3 = 1.24(330) 409 units Forecast for Quarter 4 = 0.76(330) 251 units
13 52
Seasonal Patterns
Demand
|
8 Period
10
12
14
16
13 53
Seasonal Patterns
Demand
|
8 Period
10
12
14
16
13 54
Forecast errors detect when something is going wrong with the forecasting system
Forecast errors can be classified as either bias errors or random errors Bias errors are the result of consistent mistakes Random error results from unpredictable factors that cause the forecast to deviate from the actual demand
13 55
CFE = Et
CFE E= n
MAD =
|Et |
n
MSE =
Et2
n
MAPE =
(|Et |/ Dt)(100)
n
13 56
3
4 5 6 7
300
270 230 260 210
285
290 250 240 250
15
20 20 20 40 400 400 1,600 20 20 40 8.7 7.7 19.0
275
240
Total
35
15
1,225
5,275
35
195
12.7
81.3%
13 57
3
4 5 6 7
300
270 230 260 210
285
290 250 240 250
15
20 20 20 40
225
400 400 400 1,600
15
20 20 20 40
5.0
7.4 8.7 7.7 19.0
275
240
Total
35
15
1,225
5,275
35
195
12.7
81.3%
13 58
13 59
[Et (1.875)]2
n1
= 27.4
13 60
13 61
Tracking Signals
A measure that indicates whether a method of forecasting is accurately predicting actual changes in demand Useful when forecast systems are computerized because it alerts analysts when forecast are getting far from desirable limits CFE Tracking signal = MAD Each period, the CFE and MAD are updated to reflect current error, and the tracking signal is compared to some predetermined limits
Copyright 2010 Pearson Education, Inc.
13 62
Tracking Signals
The MAD can be calculated as a weighted average determined by the exponential smoothing method MADt = |Et| + (1 )MADt-1
If forecast errors are normally distributed with a mean of 0, the relationship between and MAD is simple
= ( /2)(MAD) 1.25(MAD) MAD = 0.7978 0.8
13 63
Tracking Signals
+2.0 +1.5 Tracking signal +1.0 +0.5 Out of control Control limit
0
0.5 1.0 1.5
|
Control limit
| | | |
10 15 20 Observation number
25
13 64
13 65
13 66
Forecasting as a Process
A typical forecasting process
Step 1: Adjust history file Step 2: Prepare initial forecasts
13 67
Forecasting as a Process
Revise forecasts 4
13 68
Forecasting Principles
TABLE 13.2 | SOME PRINCIPLES FOR THE FORECASTING PROCESS Better processes yield better forecasts Demand forecasting is being done in virtually every company, either formally or informally. The challenge is to do it wellbetter than the competition Better forecasts result in better customer service and lower costs, as well as better relationships with suppliers and customers The forecast can and must make sense based on the big picture, economic outlook, market share, and so on The best way to improve forecast accuracy is to focus on reducing forecast error Bias is the worst kind of forecast error; strive for zero bias
Whenever possible, forecast at more aggregate levels. Forecast in detail only where necessary
Far more can be gained by people collaborating and communicating well than by using the most advanced forecasting technique or model
13 69
Solved Problem 1
The monthly demand for units manufactured by the Acme Rocket Company has been as follows:
Month Units Month Units
100 80 110
115
a. Use the exponential smoothing method to forecast June to January. The initial forecast for May was 105 units; = 0.2.
b. Calculate the absolute percentage error for each month from June through December and the MAD and MAPE of forecast error as of the end of December.
c. Calculate the tracking signal as of the end of December. What can you say about the performance of your forecasting method?
Copyright 2010 Pearson Education, Inc.
13 70
Solved Problem 1
SOLUTION a.
Current Month, t May June July Calculating Forecast for Next Month Ft+1 = Dt + (1 )Ft 0.2(100) + 0.8(105) 0.2(110) + 0.8(99.2) = 104.0 or 104 99.2 or 99 = 101.4 or 101 0.2(80) + 0.8(104.0) = Forecast for Month t + 1 June July August
August
September October November December
September
October November December January
13 71
Solved Problem 1
b.
Actual Demand, Dt 80 Absolute Percent Error, (|Et|/Dt)(100) 30.0% 10.0 12.0 1.0 5.5 16.0 9.2 83.7%
Month, t June
Forecast, Ft 104
Error, Et = Dt Ft 24 11 14 1 6 20 11 39
July
August September October November December Total
110
115 105 110 125 120 765
99
101 104 104 105 109
Solved Problem 1
c. As of the end of December, the cumulative sum of forecast errors (CFE) is 39. Using the mean absolute deviation calculated in part (b), we calculate the tracking signal: CFE 39 = = 3.14 MAD 12.4
Tracking signal =
The probability that a tracking signal value of 3.14 could be generated completely by chance is small. Consequently, we should revise our approach. The long string of forecasts lower than actual demand suggests use of a trend method.
13 73
Solved Problem 2
The Polish Generals Pizza Parlor is a small restaurant catering to patrons with a taste for European pizza. One of its specialties is Polish Prize pizza. The manager must forecast weekly demand for these special pizzas so that he can order pizza shells weekly. Recently, demand has been as follows:
Week June 2 Pizzas 50 Week June 23 Pizzas 56
June 9
June 16
65
52
June 30
July 7
55
60
a. Forecast the demand for pizza for June 23 to July 14 by using the simple moving average method with n = 3 then using the weighted moving average method with and weights of 0.50, 0.30, and 0.20, with 0.50. b. Calculate the MAD for each method.
Copyright 2010 Pearson Education, Inc.
13 74
Solved Problem 2
SOLUTION a. The simple moving average method and the weighted moving average method give the following results:
Current Week June 16 Simple Moving Average Forecast for Next Week Weighted Moving Average Forecast for Next Week [(0.5 52) + (0.3 65) + (0.2 50)] = 55.5 or 56 [(0.5 56) + (0.3 52) + (0.2 65)] = 56.6 or 57 [(0.5 55) + (0.3 56) + (0.2 52)] = 54.7 or 55 [(0.5 60) + (0.3 55) + (0.2 56)] = 57.7 or 58
52 + 65 + 50 = 55.7 or 56 3
56 + 52 + 65 = 57.7 or 58 3 55 + 56 + 52 = 54.3 or 54 3 60 + 55 + 56 = 57.0 or 57 3
June 23
June 30 July 7
13 75
Solved Problem 2
b. The mean absolute deviation is calculated as follows:
Simple Moving Average Week June 23 June 30 July 7 Actual Demand 56 55 60 Forecast for This Week 56 58 54 Absolute Errors |Et| |56 56| = 0 |55 58| = 3 |60 54| = 6 MAD = 0+3+6 =3 3 Weighted Moving Average Forecast for This Week 56 57 55 Absolute Errors |Et| |56 56| = 0 |55 57| = 2 |60 55| = 5 MAD = 0+2+2 = 2.3 3
For this limited set of data, the weighted moving average method resulted in a slightly lower mean absolute deviation. However, final conclusions can be made only after analyzing much more data.
13 76
Solved Problem 3
Chicken Palace periodically offers carryout five-piece chicken dinners at special prices. Let Y be the number of dinners sold and X be the price. Based on the historical observations and calculations in the following table, determine the regression equation, correlation coefficient, and coefficient of determination. How many dinners can Chicken Palace expect to sell at $3.00 each?
Observation 1 Price (X) $2.70 Dinners Sold (Y) 760
2
3
$3.50
$2.00
510
980
4
5 6 Total Average
Copyright 2010 Pearson Education, Inc.
$4.20
$3.10 $4.05 $19.55 $3.258
250
320 480 3,300 550
13 77
Solved Problem 3
SOLUTION We use the computer to calculate the best values of a, b, the correlation coefficient, and the coefficient of determination a= b= r= r2= The regression line is Y = a + bX = 1,454.60 277.63X For an estimated sales price of $3.00 per dinner 1,454.60 277.63 0.84 0.71
Y = a + bX = 1,454.60 277.63(3.00)
= 621.71 or 622 dinners
Copyright 2010 Pearson Education, Inc.
13 78
Solved Problem 4
The Northville Post Office experiences a seasonal pattern of daily mail volume every week. The following data for two representative weeks are expressed in thousands of pieces of mail:
Day Sunday Monday Tuesday Wednesday Thursday Friday Saturday Total Week 1 5 20 30 35 49 70 15 224 Week 2 8 15 32 30 45 70 10 210
a. Calculate a seasonal factor for each day of the week. b. If the postmaster estimates 230,000 pieces of mail to be sorted next week, forecast the volume for each day.
Copyright 2010 Pearson Education, Inc.
13 79
Solved Problem 4
SOLUTION a. Calculate the average daily mail volume for each week. Then for each day of the week divide the mail volume by the weeks average to get the seasonal factor. Finally, for each day, add the two seasonal factors and divide by 2 to obtain the average seasonal factor to use in the forecast.
13 80
Solved Problem 4
Week 1 Mail Volume 5 20 30 35 49 70 15 224 224/7 = 32 Seasonal Factor (1) 5/32 = 0.15625 20/32 = 0.62500 30/32 = 0.93750 35/32 = 1.09375 49/32 = 1.53125 70/32 = 2.18750 15/32 = 0.46875 Mail Volume 8 15 32 30 45 70 10 210 210/7 = 30 Week 2 Seasonal Factor (2) 8/30 = 0.26667 15/30 = 0.50000 32/30 = 1.06667 30/30 = 1.00000 45/30 = 1.50000 70/30 = 2.33333 10/30 = 0.33333 Average Seasonal Factor [(1) + (2)]/2 0.21146 0.56250 1.00209 1.04688 1.51563 2.26042 0.40104
Day
Sunday Monday Tuesday Wednesday Thursday Friday Saturday Total Average
13 81
Solved Problem 4
b. The average daily mail volume is expected to be 230,000/7 = 32,857 pieces of mail. Using the average seasonal factors calculated in part (a), we obtain the following forecasts:
Day Sunday Monday Tuesday Wednesday Thursday Friday Saturday Calculations 0.21146(32,857) = 0.56250(32,857) = Forecast 6,948 18,482 32,926 34,397
1.00209(32,857) =
1.04688(32,857) = 1.51563(32,857) = 2.26042(32,857) = 0.40104(32,857) = Total
49,799
74,271
13,177
230,000
13 82
13 83