Forecasting

1
Forecasting
2
Forecasting
3
Class Results Getting ready in the AM

• Most number of steps 21
• Longest snooze time 120 minutes
• Longest shower 45 mins
• Shortest time getting ready (w/shower) 20 mins
• Longest time getting ready 3 hrs
• Latest alarm setting 1130 AM
• sets or ties previous record

4
Tuesday mornings ready in 13 minutes
30 secs
1
2
shut off alarm / get up
5
Tuesday mornings ready in 13 minutes
3
Brush teeth
60 secs
30 secs
90 secs
4
1
2
start shower
shut off alarm / get up
start coffee
6
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
30 secs
90 secs
4
1
2
start shower
shut off alarm / get up
start coffee
7
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shower
30 secs
90 secs
6
4
1
2
180 secs
start shower
shut off alarm / get up
start coffee
8
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
start coffee
8 mins
9
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
10
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
9
11
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
dress
10
9
90 secs
12
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
make breakfast
dress
10
9
5
90 secs
60 secs
13
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
make breakfast
eat breakfast
dress
10
11
9
5
90 secs
60 secs
180 secs
gather stuff
30 secs
13
14
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
make breakfast
eat breakfast
dress
10
11
9
5
90 secs
60 secs
180 secs
gather stuff
30 secs
drive 30 mi
14
13
30 mins
15
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
make breakfast
eat breakfast
dress
10
11
9
5
90 secs
60 secs
180 secs
gather stuff
30 secs
Park n Ride
drive 30 mi
14
15
13
3 mins
30 mins
16
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
make breakfast
eat breakfast
dress
10
11
9
5
90 secs
60 secs
180 secs
gather stuff
30 secs
Park n Ride
shuttle/check-in
drive 30 mi
14
15
16
13
3 mins
30 mins
8 mins
17
Tuesday mornings ready in 13 minutes
3
Select clothes 30 secs
Brush teeth
60 secs
shave
shower
30 secs
90 secs
6
7
4
1
2
180 secs
120 secs
start shower
shut off alarm / get up
30 secs
towel dry
start coffee
8
8 mins
apply makeupcomb hair
0 secs
13 minutes
make breakfast
eat breakfast
dress
10
11
9
5
90 secs
60 secs
180 secs
gather stuff
30 secs
54 1/2 minutes
Park n Ride
shuttle/check-in
drive 30 mi
14
15
16
13
3 mins
30 mins
8 mins
18
Forecasting
19
Situation Analysis

• You are the recently promoted manager of CSUSMs
  Starbuck Coffee on campus. Its the week before
  finals, and you are ordering coffee for next
  week. What things should you consider into how
  much coffee to order???

20
Situation Analysis

• You are a senior forecasting agent for the
  Internal Revenue Service. Your boss would like
  you to forecast how much federal tax revenue the
  IRS can expect to collect in 2007 from
  California. What factors should you consider in
  your response?

21
Situation Analysis

• You are the Senior Production Manager for the
  recently announced Boeing 777 aircraft. How
  would you decide what production methods to use?

22
Situation Analysis

• You are the instructor for HTM302, and have
  access to homework grades for each student. Can
  you use these to forecast how well students will
  do on quizzes?

23
Forecast Quiz from Homework scores?
R2 0.57
24
Forecasting

• Predicting future events
• Qualitative methods
• Based on subjective methods
• Quantitative methods
• Based on mathematical formulas

25
Forecasting

• Example Here are the factors that have entered
  into a salespersons forecast for next quarter
  which are quantitive and which are qualitative?
• the past quarters sales
• marketings projections of market size growth
• opinions of fellow salespeople
• estimates of finding new accounts
• customers willingness to do repeat orders

26
Forecasting

• Example What are the causal underlying effects
  that influence whether or not a San Diego
  restaurant in the Gaslamp Quarter will fill to
  capacity this Friday night?
• Causal effects
• Is there a Padres game?
• Are there other special downtown events planned?
  
• Are there events elsewhere in the county that may
  draw people away from downtown?

27
Forecasting

• You forecast all the time, though you may not
  often think of it as such. Examples
• How long will it take me to complete the
  homework?
• How good a job will my teammates do on their
  portions of our group project?
• How long will it take me to drive to work?
• How long will it take me to drive to Mammoth?
• Will this class end early or on time?

28
Forecasting

• You need to find the underlying variables that
  cause or effect the underlying pattern, monitor
  those, and then construct your forecast as best
  you can
• Example What are the causal underlying effects
  that influence whether or not it will rain here
  Saturday?
• Casual effects air movement patterns, air
  pressure changes, weather patterns in surrounding
  regions

29
Forecasting

• Situation analysis You are planning for next
  semesters classes, and
• ABC 102 is on the required list for graduation.
  You want to know,
• will it be a good class? What do you do for your
  forecast???
• Qualitative methods, based on subjective methods
• you ask the COBA advisors
• You ask your friends
• Quantitative methods
• You look at past enrollment statistics
• You look at ratemyprofessor.com (score out of 5
  points)
• 1 point out of 5 A bitter professor that needs
  a major attitude adjustment. Her dark reign of
  terror continues. I wouldn't wish this class
  upon my worst enemy. If I was in HELL, I'm pretty
  sure this would be part of the torture
  curriculum. For a more exciting two hours a week,
  watch a colony of ants slowly move pieces of food
  from one place to another. 

30
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

31
Components of Forecasting Demand

• Time Frames
• Short-range
• medium-range
• long-range
• Demand Behavior
• Trends, cycles, seasonal patterns, random

32
Time Frame

• Short-range to medium-range
• Daily, weekly, monthly, quarterly forecasts of
  sales data
• Up to 2 years into the future
• Long-range
• Strategic planning of goals, products, markets
• Planning beyond 2 years

33
Demand Behavior

• Trend
• gradual, long-term up or down movement
• Cycle
• up down movement repeating over long time frame
• Seasonal pattern
• periodic oscillation in demand which repeats
• Random movements follow no pattern

34
Demand Behavior HTM 302 Textbooks

• Trend
• More rapid revisions of textbooks to keep up with
  pace of change of industry
• Long term industry trend toward using
  unconventional materials (online materials,
  customized textbooks, etc.)
• Seasonal pattern
• Demand high at start of fall and spring
  semesters, very low otherwise

35
Demand Behavior Snowboard sales

• Trend
• Snowboard sales have gradually displaced/replaced
  ski sales as snowboarding grew in popularity
• Seasonal pattern
• Sales higher in fall/winter, lower in
  spring/summer

36
Demand Behavior Restaurant sales

• Trend
• People are eating out more as population has
  become more mobile, demand convenience time
  savings, multiple wage earners/family,
  multiplicity of fast food restaurants
• Cycle
• Eating-out follows general economic trends
• Food selections follow diet trends

37
Forms of Forecast Movement Trend
Idealized
Demand
Time
38
Forms of Forecast Movement Cyclic
Ideal
Demand
Time
39
Forms of Forecast Movement Seasonal
Idealized
Demand
Time
40
Trend with Seasonal Pattern
Demand
Time
41
Demand Behavior HTM 302 Textbooks
What demand Behavior factors are apparent from
the data at right?
42
Demand Behavior Class attendance

• What is the most important causal factor that
  affects class attendance next week in HTM 302?
• What other causal factors do you think might be
  in play?

Class quiz
Holiday nearby, difficulty of material,other
class work and tests, etc.
43
Forecasting Methods

• Qualitative methods
• Management judgment, expertise, opinion
• Use management, marketing, purchasing,
  engineering
• Quantitative methods
• Time series analysis, regression, or causal
  modeling
• Delphi method
• Solicit forecasts from experts

44
Forecasting Process
45
Time Series Methods

• Assume patterns will repeat
• Naive forecasts
• Forecast data from last period
• Statistical methods using historical data
• Moving average
• Exponential smoothing
• Linear trend line

46
Moving Average

• Average several periods of data
• Dampen, smooth out changes
• Use when demand is stable with no trend or
  seasonal pattern

47
Moving Average

• Average several periods of data
• Dampen, smooth out changes
• Use when demand is stable with no trend or
  seasonal pattern

48
Simple Moving Average
Its October 31st, how much Product should we
order for Sale in November?
49
Smoothing Effects
50
Simple Moving Average
Previous 3 periods
110 orders for Nov
51
Simple Moving Average
52
Simple Moving Average
Previous 5 periods
91 orders for Nov
53
Simple Moving Average
54
Smoothing Effects
55
Smoothing Effects
56
Smoothing Effects
57
HTM 302 Quiz Scores
You have a student in the class that has
achieved the following quiz scores is he/she in
trouble or not? What conclusions can you draw?
What would be a good forecast for the next quiz
grade?
58
(No Transcript)
59
Forecasting at home
60
(No Transcript)
61
Weighted Moving Average

• Adjusts moving average method to more closely
  reflect data fluctuations, often those most
  recent in time

62
Weighted Moving Average

• Adjusts moving average method to more closely
  reflect data fluctuations

63
Weighted Moving Average Example
64
Weighted Moving Average Example
65
Exponential Smoothing

• Averaging method
• Weights most recent data more strongly
• Reacts more to recent changes
• Widely used, accurate method

66
Linear Trend Line
y a bx where a intercept (at period
0) b slope of the line x the time
period y forecast for demand for period x
67
Least Squares Example Raw Data
Example 8.5
68
Linear Trend Line
Example 8.5
69
Linear Trend Line
y 35.2 1.72x
Example 8.5
70
Whats going on?
The next slide displays historical data on number
of service calls to Geeks on Call.
What would be an appropriate method of
forecasting based on your observation of the data?
71
(No Transcript)
72
(No Transcript)
73
Forecast Accuracy

• Error Actual - Forecast
• Find a forecast method which minimizes the error
• Mean Absolute Deviation (MAD)

74
Mean Absolute Deviation (MAD)
where t the period number Dt demand
in period t Ft the forecast for period t
n the total number of periods ???? the
absolute value
75
Forecast Control

• Reasons for out-of-control forecasts
• Change in trend
• Promotions
• Competition
• Politics
• Monitor forecast accuracy and modify accordingly

76
Correlation and Coefficient of Determination

• Correlation, r
• Measure of strength of relationship
• Varies between -1.00 and 1.00
• Coefficient of determination, r2
• Percentage of variation in dependent variable
  resulting from changes in the independent variable

77
Computing Correlation
78
Correlation Coefficient Examples
79
Multiple Regression
Study the relationship of demand (y) to two or
more independent variables (x1, x2, x3.) y
?0 ?1x1 ?2x2 ?kxk where ?0 the
intercept ?1, , ?k parameters for
the independent variables x1, ,
xk independent variables
80
Application Gaming Quality of Service
Sensitivity of online gamers to network quality
Empirical data
Communications of the ACM, vol. 49 no. 11
November 2006
81
Modeling Gaming Quality of Service
prob(player leaving)
Communications of the ACM, vol. 49 no. 11
November 2006
82
In class Example 1

• You are the manager of a local Pizza Hut. Here
  are the Sales figures (number of pizzas sold) for
  the last 3 weeks

• Questions
• What are the general trends apparent from the
  data?
• What would you forecast for demand for Saturday
  night of the 4th week?

83
In class Example 2

• You are the manager of the campus Starbucks.
  Here are the Sales figures (number of customers)
  for the last 3 weeks

• Questions
• What are the general trends apparent from the
  data?
• What is a possible explanation for week 2?

84
In class Example 3

• You are the buyer for Dals Surf Shop in
  Oceanside here are the per quarter sales for the
  past three years

• Question
• What are the general trends apparent from the
  data?

85
In class Example 4

• You are the professor of HTM 302 here are the
• class average scores on the quizzes to date

What would be a good forecast for the next quiz
grade?
86
(No Transcript)
87
(No Transcript)
88
equally-weighted average over course
weighted average over recent three classes
89
Informal feedback

• Write a 2 minute journal to be handed in
  immediately
• The journal should briefly summarize
• Major points learned
• Areas not understood or requiring clarification