1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives ...
-
Upload
doris-spencer -
Category
Documents
-
view
213 -
download
0
Transcript of 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives ...
![Page 1: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/1.jpg)
1
IES 371 Engineering Management
Chapter 13: Forecasting
Week 12August 24, 2005
Objectives Understand and practice on various forecasting method
![Page 2: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/2.jpg)
2
Dr. Karndee Prichanont IES371 1/2005
Strategic Role of Forecasting
Focus on supply chain management Short term role of product demand Long term role of new products, processes, and
technologies
Focus on Total Quality Management Satisfy customer demand Uninterrupted product flow with no defective items
Necessary for strategic planning
![Page 3: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/3.jpg)
3
Dr. Karndee Prichanont IES371 1/2005
Demand Forecast Applications
Time Horizon
Medium Term Long Term Short Term (3 months– (more than
Application (0–3 months) 2 years) 2 years)
Total salesGroups or familiesof products orservicesStaff planningProductionplanningMaster productionschedulingPurchasingDistribution
CausalJudgment
Forecast quantity Individualproducts orservices
Decision area InventorymanagementFinal assemblyschedulingWorkforceschedulingMaster productionscheduling
Forecasting Time seriestechnique Causal
Judgment
Total sales
Facility locationCapacityplanningProcessmanagement
CausalJudgment
![Page 4: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/4.jpg)
4
Dr. Karndee Prichanont IES371 1/2005
Components for Forecasting Demand
Time frame Short to mid-range Long-rangeDepend on each firm and industry type
Demand behavior Trend Cycle Seasonal
Forecasting method Forecasting process TimeTime
(c) Seasonal pattern(c) Seasonal pattern
Dem
and
TimeTime(c) Seasonal pattern(c) Seasonal pattern
TimeTime(d) Trend with seasonal pattern(d) Trend with seasonal pattern
Dem
and
Dem
and
TimeTime(d) Trend with seasonal pattern(d) Trend with seasonal pattern
Dem
and
Dem
and
Dem
and
TimeTime(b) Cycle(b) Cycle
Dem
and
Dem
and
TimeTime(b) Cycle(b) Cycle
Dem
and
Dem
and
TimeTime(a) Trend(a) Trend
Dem
and
Dem
and
Random Random movementmovement
TimeTime(a) Trend(a) Trend
Dem
and
Dem
and
Random Random movementmovement
![Page 5: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/5.jpg)
5
Dr. Karndee Prichanont IES371 1/2005
Forecasting Techniques Judgment method
Qualitative method that translates the opinions of managers, expert opinions, customer surveys, and sale-force estimate into quantitative estimates
Causal method
Use of past data on independent variables to derive mathematical relationship Demand = f(relevant factors)
Linear Regression
Time Series analysis
statistical technique using past data for short-term forecasting applications
Naïve forecast Moving average Weighted moving average
Exponential smoothing Adjusted exponential smoothing Linear trend line
![Page 6: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/6.jpg)
6
Dr. Karndee Prichanont IES371 1/2005
Causal Method – Linear Regression Method
Mathematical relationship between two or more variables
What causes to behave in a certain way?
Linear regression: y = a + bx Correlation
Strength of the relationship between dependent and independent variables
Range of [-1.00, 1.00] 2222
yynxxn
yxxynr
xbya
xnx
yxnxyb
22
![Page 7: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/7.jpg)
7
Dr. Karndee Prichanont IES371 1/2005
Causal Method – Linear Regression Method
Dep
end
ent
vari
able
Dep
end
ent
vari
able
Independent variableIndependent variableXX
YYEstimate ofEstimate ofY Y from fromregressionregressionequationequation
RegressionRegressionequation:equation:YY = = aa + + bXbX
ActualActualvaluevalueof of YY
Value of Value of X X used usedto estimate to estimate YY
Deviation,Deviation,or erroror error
{
![Page 8: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/8.jpg)
8
Dr. Karndee Prichanont IES371 1/2005
Ex 1: Causal Method – Linear Regression Method
Carpet City wants to develop a means to forecast its carpet sales. The carpet store manager believes that the store’s sales are directly related to the number of new housing starts in town. The manager has gathered data from county records of monthly house construction permits and from store records on monthly sales. These data are as follows:
MonthMonthly Carpet
Sales (1000s YD)
Montly Construction
Permits1 5 212 10 353 4 104 3 125 8 166 2 97 12 418 11 159 9 18
10 14 26
Required• Develop a linear regression model for this data and forecast sales if 30 construction permits for new home are filed
• Determine the strength of the casual relationship between monthly sales and new home construction using correlation
![Page 9: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/9.jpg)
9
Dr. Karndee Prichanont IES371 1/2005
Time Series Method- Simple Moving Average
Use several demand values in the recent past Smooth out randomness Suitable for stable demand Number of periods in the moving average smoother
forecast
n
DDDn
nF
nttt
t
...
demands last of Sum
1
1 n = Total number of periods in the averageDt = Demand in period tFt+1 = Forecast for period t +1
![Page 10: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/10.jpg)
10
Dr. Karndee Prichanont IES371 1/2005
Ex 2: Time Series Method- Simple Moving Average
The manager forecast weekly demand for his specialty pizza so that he can order pizza shells weekly. Forecast the demand for pizza for June 23 to July 14 by using the simple moving average method with n = 3
Recently demand has been as follows:
Week of Pizzas Week of Pizzas
June 2June 9June 16
506552
June 23June 30July 7
565560
![Page 11: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/11.jpg)
11
Dr. Karndee Prichanont IES371 1/2005
Time Series Method- Weighted Moving Average
More closely reflect data fluctuations Recent data, more weight Weight: trial-and-error experiment
Wi = weight for period i, [0%, 100%] Wi = 1.00
t
iiit DWF
11
t
iiit DWF
11
Ex 3: With the demand data in Example 2, forecast the demand for pizza by using WMA method with n = 3 and weights of 0.50, 0.30, and 0.20, with 0.50 applying to the most recent demand
![Page 12: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/12.jpg)
12
Dr. Karndee Prichanont IES371 1/2005
Time Series Method- Exponential Smoothing
Weights the most recent data more strongly React to recent changes in demand Seasonal pattern of demand Widely used in many businesses Requires minimal data
ttt FDF 11
Ft+1 = the forecast for the next periodDt = actual demand in the current periodFt = the previously determined forecast for the current period = smoothing constant (weighted factor)
![Page 13: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/13.jpg)
13
Dr. Karndee Prichanont IES371 1/2005
Ex 4: Time Series Method- Exponential Smoothing
Use the exponential smoothing method to forecast the number of units for June to January. The initial forecast for May was 105 units; = 0.2
The monthly demand for units manufactured of a company has been as follows:
Month Units Month Units
MayJuneJulyAugust
10080110115
SeptemberOctoberNovemberDecember
105110125120
![Page 14: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/14.jpg)
14
Dr. Karndee Prichanont IES371 1/2005
Time Series Method- Linear Trend Line
For demand with obvious trend over time A least squares regression line
y = a + bx Dependent variable (y) = Forecasted
demand Independent variable (x) = Time a = intercept at period 0 b = Slope of the line
xbya
xnx
yxnxyb
22
![Page 15: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/15.jpg)
15
Dr. Karndee Prichanont IES371 1/2005
Choosing a Time-Series MethodForecast Error
Forecast error
Cumulative sum of forecast errors (CFE)
Mean squared error (MSE)
Standard deviation
Mean absolute deviation (MAD)
Mean absolute percent error (MAPE)
![Page 16: 1 IES 371 Engineering Management Chapter 13: Forecasting Week 12 August 24, 2005 Objectives Understand and practice on various forecasting method.](https://reader036.fdocuments.in/reader036/viewer/2022082819/56649e165503460f94b01bb1/html5/thumbnails/16.jpg)
16
Dr. Karndee Prichanont IES371 1/2005
Tracking Signal
Tracking signal = Tracking signal = CFECFE
MADMAD+2.0 +2.0 —
+1.5 +1.5 —
+1.0 +1.0 —
+0.5 +0.5 —
0 0 —
––0.5 0.5 —
––1.0 1.0 —
––1.5 1.5 —| | | | |
00 55 1010 1515 2020 2525 Observation numberObservation number
Tra
ckin
g s
ign
alT
rack
ing
sig
nal
Control limitControl limit
Control limitControl limit
Out of controlOut of control