Operation Management
Operation Management
Lecture 2: Forecasting
Learning Objectives
What you will learn in this lecture:
Demand Management Qualitative Forecasting Methods Quantitative Forecasting Methods Forecasting Error Estimation
Slide 1 of 46
School of Business
Slide 2 of 46
School of Business
Demand Management
Independent Demand: Finished Goods
A
B(4)
C(2)
D(2)
E(1)
D(3)
F(2)
Slide 3 of 46
School of Business
Slide 4 of 46
School of Business
2/13/2014
Forecast
A statement about the future Used to help managers
Plan the system Plan the use of the system
Uses of Forecasts
Accounting Finance Human Resources Marketing Operations Product/service design Cost/profit estimates Cash flow and funding Hiring/recruiting/training Pricing, promotion, strategy Schedules, workloads, inventory planning Modification of current features, design new products and services
Slide 5 of 46
School of Business
Slide 6 of 46
School of Business
Timely
Reliable
Accurate
Written
Slide 7 of 46
School of Business
Slide 8 of 46
School of Business
2/13/2014
Types of Forecasts
Qualitative (Judgmental)
The forecast
Quantitative
Step 6 Monitor the forecast Step 5 Prepare the forecast Step 4 Gather and analyze data Step 3 Select a forecasting technique Step 2 Establish a time horizon Step 1 Determine purpose of forecast
Slide 9 of 46
School of Business
Slide 10 of 46
School of Business
Qualitative Methods
Delphi Method
l. Choose the experts to participate representing a
variety of knowledgeable people in different areas 2. Through a questionnaire (or E-mail), obtain forecasts (and any premises or qualifications for the forecasts) from all participants 3. Summarize the results and redistribute them to the participants along with appropriate new questions 4. Summarize again, refining forecasts and conditions, and again develop new questions 5. Repeat Step 4 as necessary and distribute the final results to all participants
Executive Judgment
Grass Roots
Historical analogy
Qualitative Methods
Market Research
Delphi Method
Panel Consensus
Slide 11 of 46
School of Business
Slide 12 of 46
School of Business
2/13/2014
Slide 13 of 46
School of Business
Slide 14 of 46
School of Business
Components of Demand
Analysis of time series data requires the
analyst to identify the underlying behavior of the series
x x x x x xx x x xx x x x x x x x x x x x x xx x x x x x x x x x x x
Linear
x x x
Trend
x x x x x 1
x x
Year
School of Business North South University
Slide 16 of 46
Slide 15 of 46
School of Business
2/13/2014
Ft =
Question: What are the 3g week and 6-week moving average forecasts for demand? Assume you only have 3 weeks and 6 weeks of actual demand data for the respective forecasts
Slide 18 of 46
School of Business
School of Business
Note how the 3Week is smoother than the Demand, and 6-Week is even smoother
School of Business
Slide 20 of 46
School of Business
2/13/2014
Week 1 2 3 4
Weights: t1 .5 t2 .3 t3 .2
w
i=1
=1
Note that the weights place more emphasis on the most recent data, that is time period t-1
School of Business
Slide 22 of 46
School of Business
Week 1 2 3 4
Forecast
693.4
WMA is better than SMA because of the ability to vary the weights!
F4 =0.5(720)+0.3(678)+0.2(650)=693.4
Slide 23 of 46
School of Business
Slide 24 of 46
School of Business
2/13/2014
Where : Ft Forcast value for the coming t time period Ft - 1 Forecast value in 1 past time period At - 1 Actual occurance in the past 1 time period Alpha smoothing constant
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
Question: Given the weekly demand data, what are the exponential smoothing forecasts for periods 2-10 using =0.10 and =0.60? Assume F1=D1
Slide 25 of 46
School of Business
Slide 26 of 46
School of Business
Week 1 2 3 4 5 6 7 8 9 10
North South University
Demand 820 775 680 655 750 802 798 689 775
0.1 820.00 820.00 815.50 801.95 787.26 783.53 785.38 786.64 776.88 776.69
0.6 820.00 820.00 793.00 725.20 683.08 723.23 770.50 787.00 728.20 756.30
School of Business North South University
Slide 28 of 46
Slide 27 of 46
School of Business
2/13/2014
Commonly used values for : 0.050 05 0.5 05 represents a percentage of forecast error The closer the value of is to zero, the slower the forecast will be to adjust the forecast errors the greater
the smoothing
The closer the value of is to 1, the greater the responsiveness the less the smoothing
North South University
Slide 29 of 46
School of Business
Slide 30 of 46
School of Business
What do you think will happen to a moving average or exponential smoothing model when there is a trend in the data?
The simple linear regression model seeks to fit a line through various data over time
012345x (Time)
Yt =a +bx
i yi - Yi
North South University
Slide 31 of 46
School of Business
School of Business
2/13/2014
a = y - bx
x y - n ( y )( x ) x
2
b =
- n(x )2
Assumptions Variations around the line are random Deviations around the line should be normally distributed Predictions are being made only within the range of observed values Limitations Can be applied only to linear relationships with one independent variable Needs considerable amount of data Equal weightage on each data
Week 1 2 3 4 5
North South University
Slide 33 of 46
School of Business
School of Business
Yt =143.5+6.3x
Nowifweplottheregressiongeneratedforecastsagainsttheactualsalesweobtain thefollowingchart: 180 175 170 165 160 155 150 145 140 135 1 2 3 Period 4 5
Sales
Sales Forecast
b=
School of Business
Slide 36 of 46
School of Business
2/13/2014
x = Number of housing
start permits
13,000 12 000 12,000 11,000 10,000 14,000 16,000 19,000 17,000 13,000
y = Sales of carpeting p g
School of Business
Slide 38 of 46
School of Business
Y a b1 x1 b2 x2 b3 x3 ...
'
a intercept, the value of y when all the x' s are zero b j amount by which y changes when the particular x j increases by one unit with all other values held the same
North South University
Slide 39 of 46
1 2 3 4 5 6 7 8 9 10 11 12 13
School of Business
Slide 40 of 46
School of Business
10
2/13/2014
Forecast Accuracy
It is a significant factor if the task is to decide among
available forecasting alternatives
Forecast t n
Actual
Forecastt
n 1
MAPE=
Slide 41 of 46
School of Business
Slide 42 of 46
School of Business
Forecast Accuracy
MAD weights all errors evenly MSE weights errors according to their squared values MAPE weights according to relative error
Period 1 2 3 4 5 6 7 8 Actual 217 213 216 210 213 219 216 212 Forecast 215 216 215 214 211 214 217 216
Slide 43 of 46
School of Business
Slide 44 of 46
School of Business
11
2/13/2014
Assignment-2
Divide data into initiation set and evaluation set Use U the th first fi t set t to t develop d l the th models d l Use the second set to evaluate Compare the MADs, MSEs, and MAPEs of each
model
Problems: P bl 2 3, 2, 3 5, 5 7, 7 26
Slide 45 of 46
School of Business
Slide 46 of 46
School of Business
12