Supply Chain Management

1
Supply Chain Management

• Lecture 10

2
Outline

• Today
• Finish Chapter 6 (Decision tree analysis)
• Start chapter 7
• Tomorrow
• Homework 2 due before 500pm
• Next week
• Chapter 7 (Forecasting)

3
Example Decision Tree Analysis

• New product with uncertain demand (85
  profit/unit)
• Annual demand expected to go up by 20 with
  probability 0.6
• Annual demand expected to go down by 20 with
  probability 0.4
• Use discount factor k 0.1

4
Example

1. Represent the tree, identifying all states as
   well as all transition probabilities

Period 2
P 12085(0.6122400.48160)/1.1 19844
P 12240
Period 1
D144
0.6
Period 0
D120
0.6
0.4
P 8160
D100
D96
0.6
0.4
P 10085(0.6198440.413229)/1.1 24135
D80
0.4
P 5440
D64
P 8085(0.681600.45440)/1.1 13229
5
Example

1. Represent the tree, identifying all states as
   well as all transition probabilities

Period 2
Period 1
D144
0.6
Period 0
D120
0.6
0.4
D100
D96
0.6
0.4
D80
0.4
D64
Calculate the NPV of each possible scenario
separately
6
Example

1. Represent the tree, identifying all states as
   well as all transition probabilities

Calculate the NPV of each possible scenario
separately
7
Decision Trees (Summary)

• A decision tree is a graphic device used to
  evaluate decisions under uncertainty
• Identify the duration of each period and the
  number of time periods T to be evaluated
• Identify the factors associated with the
  uncertainty
• Identify the representation of uncertainty
• Identify the periodic discount rate k
• Represent the tree, identifying all states and
  transition probabilities
• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• (Alternatively, calculate the NPV of each
  possible scenario separately)

8
Decision Trees

• Using decision trees to evaluate network design
  decisions
• Should the firm sign a long-term contract for
  warehousing space or get space from the spot
  market as needed
• What should the firms mix of long-term and spot
  market be in the portfolio of transportation
  capacity
• How much capacity should various facilities have?
  What fraction of this capacity should be flexible?

9
Example Decision Tree Analysis

• Three options for Trips Logistics
• Get all warehousing space from the spot market as
  needed
• Sign a three-year lease for a fixed amount of
  warehouse space and get additional requirements
  from the spot market
• Sign a flexible lease with a minimum change that
  allows variable usage of warehouse space up to a
  limit with additional requirement from the spot
  market

10
Example Decision Tree Analysis

• Trips Logistics input data
• Evaluate each option over a 3 year time horizon
  (1 period is 1 year)
• Demand D may go up or down each year by 20 with
  probability 0.5
• Warehouse spot price p may go up or down by 10
  with probability 0.5
• Discount rate k 0.1

11
Example

1. Represent the tree, identifying all states

0.25
0.25
0.25
0.25
0.25
Period 0
0.25
D100
0.25
p1.20
0.25
12
Example Option 1 (Spot)
Period 2

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• C(D 144,000, p 1.45, 2) 144,000 x 1.45
  208,800
• R(D 144,000, p 1.45, 2) 144,000 x 1.22
  175,680
• P(D 144,000, p 1.45, 2) R C
  175,680 208,800 33,120

D144
p1.45
D144
p1.19
D96
Cost
p1.45
D144
p0.97
Revenue
D96
p1.19
D96
Profit
p0.97
D64
p1.45
D64
p1.19
D64
p0.97
13
Example Option 1 (Spot)
Period 2

• Starting at period T, work back to period 0
  identify the expected cash flows at each step

D144
p1.45
D144
p1.19
D96
p1.45
D144
p0.97
D96
p1.19
D96
p0.97
D64
p1.45
D64
p1.19
D64
p0.97
14
Example Option 1 (Spot)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• EP(D 120, p 1.22, 1) 0.25xP(D 144, p
  1.45, 2) 0.25xP(D 144, p 1.19, 2)
  0.25xP(D 96 p 1.45, 2) 0.25xP(D 96, p
  1.19, 2) 12,000
• PVEP(D 120, p 1.22, 1) EP(D 120, p
  1.22, 1)/(1k) 12,000/1.1
  10,909

15
Example Option 1 (Spot)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• P(D 120, p 1.32, 1) R(D 120, p 1.22,
  1) C(D 120, p 1.32, 1) PVEP(D 120, p
  1.22, 1) 146,400 - 158,400
  (10,909) 22,909

16
Example Option 1 (Spot)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step

17
Example Option 1 (Spot)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step

NPV(Spot) 5,471
18
Example Decision Tree Analysis

• Three options for Target.com
• Get all warehousing space from the spot market as
  needed
• Sign a three-year lease for a fixed amount of
  warehouse space and get additional requirements
  from the spot market
• Get 100,000 sq ft. of warehouse space at 1 per
  square foot
• Additional space purchased from spot market
• Sign a flexible lease with a minimum change that
  allows variable usage of warehouse space up to a
  limit with additional requirement from the spot
  market

19
Example Option 2 (Fixed lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step

Period 2
D144
p1.45
0.25
Period 1
D144
0.25
p1.19
D120
0.25
D96
p1.32
0.25
p1.45
0.25
D144
Period 0
D120
0.25
p0.97
p1. 08
D96
D100
0.25
p1.19
p1.20
D96
D80
p0.97
p1.32
D64
0.25
p1.45
D80
D64
p1.32
p1.19
D64
p0.97
20
Example Option 2 (Fixed lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• P(D , p , 2) R(D , p , 2) C(D , p , 2)
• P(D , p , 2) Dx1.22 (100,000x1.00 Sxp)

8
21
Example Option 2 (Fixed lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• P(D , p , 1) R(D , p , 1) C(D , p , 1)
  PVEP(D , p , 1)
• P(D , p , 1) Dx1.22 (100,000x1.00 Sxp)
  EP(D , p , 1)/(1k)

22
Example Option 2 (Fixed lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• P(D , p , 0) R(D , p , 0) C(D , p , 0)
  PVEP(D , p , 0)
• P(D , p , 0) 100,000x1.22 100,000x1.00
  16,364/1.1

NPV(Fixed lease) 38,364
23
Example Decision Tree Analysis

• Three options for Target.com
• Get all warehousing space from the spot market as
  needed
• Sign a three-year lease for a fixed amount of
  warehouse space and get additional requirements
  from the spot market
• Sign a flexible lease with a minimum change that
  allows variable usage of warehouse space up to a
  limit with additional requirement from the spot
  market
• 10,000 upfront payment
• Use anywhere between 60,000 and 100,000 sq ft. at
  1 per sq ft.
• Additional space purchased from spot market

24
Example Option 3 (Flexible lease)

• Flexible lease rules
• Up-front payment of 10,000
• Flexibility of using between 60,000 and 100,000
  sq.ft. at 1.00 per sq.ft. per year
• Additional space requirements from spot market

25
Example Option 3 (Flexible lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step

Period 2
D144
p1.45
0.25
Period 1
D144
0.25
p1.19
D120
0.25
D96
p1.32
0.25
p1.45
0.25
D144
Period 0
D120
0.25
p0.97
p1. 08
D96
D100
0.25
p1.19
p1.20
D96
D80
p0.97
p1.32
D64
0.25
p1.45
D80
D64
p1.32
p1.19
D64
p0.97
26
Example Option 3 (Flexible lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• P(D , p , 2) R(D , p , 2) C(D , p , 2)
• P(D , p , 2) Dx1.22 (Wx1.00 Sxp)

27
Example Option 3 (Flexible lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• P(D , p , 1) R(D , p , 1) C(D , p , 1)
  PVEP(D , p , 1)
• P(D , p , 1) Dx1.22 (Wx1.00 Sxp) EP(D
  , p , 1)/(1k)

20,000
20,000
28
Example Option 3 (Flexible lease)

• Starting at period T, work back to period 0
  identify the expected cash flows at each step
• P(D , p , 0) R(D , p , 0) C(D , p , 0)
  PVEP(D , p , 0)
• P(D , p , 0) 100,000x1.22 100,000x1.00
  38,198/1.1

NPV(Flexible lease) 56,725 10,000 46,725
29
From Design to Planning

• Network design
• C4 ? Designing Distribution Networks
• C5 ? Network Design in the Supply Chain
• C6 ? Network Design in an Uncertain Environment
• Planning in a supply chain
• C7 ? Demand Forecasting in a Supply Chain
• C8 ? Aggregate Planning in a Supply Chain
• C9 ? Planning Supply and Demand

30
Demand Forecasting

• How does BMW know how many Mini Coopers it will
  sell in North America?
• How many Prius cars should Toyota build to meet
  demand in the U.S. this year? Worldwide?
• When is it time to tweak production, upward or
  downward, to reflect a change in the market?

What factors influence customer demand?
31
Factors that Affect Forecasts

• Past demand
• Time of year/month/week
• Planned advertising or marketing efforts
• Planned price discounts
• State of the economy
• Market conditions
• Actions competitors have taken

32
Example Demand Forecast for Milk

• A supermarket has experienced the following
  weekly demand (in gallons) over the last ten
  weeks
• 109, 116, 108, 103, 97, 118, 120, 127, 114, and
  122

What is a reasonable demand forecast for milk for
the upcoming week?
When could using average demand as a forecast
lead to an inaccurate forecast?
If demand turned out to be 125 what can you say
about the demand forecast?
33
1) Characteristics of Forecasts

• Forecasts are always wrong!
• Forecasts should include an expected value and a
  measure of error (or demand uncertainty)
• Forecast 1 sales are expected to range between
  100 and 1,900 units
• Forecast 2 sales are expected to range between
  900 and 1,100 units

34
2) Characteristics of Forecasts

• Long-term forecasts are less accurate than
  short-term forecasts
• Less easy to consider other variables
• Hard to include the effects of weather in a
  forecast
• Forecast horizon is important, long-term forecast
  have larger standard deviation of error relative
  to the mean

35
3) Characteristics of Forecasts

• Aggregate forecasts are more accurate than
  disaggregate forecasts

36
3) Characteristics of Forecasts

• Aggregate forecasts are more accurate than
  disaggregate forecasts
• They tend to have a smaller standard deviation of
  error relative to the mean

Monthly sales SKU
Monthly sales product line
37
4) Characteristics of Forecasts

• Information gets distorted when moving away from
  the customer
• Bullwhip effect

38
Characteristics of Forecasts

1. Forecasts are always wrong!
2. Long-term forecasts are less accurate than
   short-term forecasts
3. Aggregate forecasts are more accurate than
   disaggregate forecasts
4. Information gets distorted when moving away from
   the customer

39
Role of Forecasting
Manufacturer
Distributor
Retailer
Customer
Supplier
Push
Push
Push
Pull
Push
Push
Pull
Push
Pull
Is demand forecasting more important for a push
or pull system?
40
Types of Forecasts

• Qualitative
• Primarily subjective, rely on judgment and
  opinion
• Time series
• Use historical demand only
• Causal
• Use the relationship between demand and some
  other factor to develop forecast
• Simulation
• Imitate consumer choices that give rise to demand

41
Components of an Observation

• Quarterly demand at Tahoe Salt

Actual demand (D)
42
Components of an Observation

• Quarterly demand at Tahoe Salt

Level (L) and Trend (T)
43
Components of an Observation

• Quarterly demand at Tahoe Salt

Seasonality (S)
44
Components of an Observation
Observed demand Systematic component Random
component
L Level (current deseasonalized demand)
T Trend (growth or decline in demand)
S Seasonality (predictable seasonal fluctuation)
45
Time Series Forecasting
Forecast demand for the next four quarters.