15 871 Recitation #4

15.871
M d ling P
Modeli Product
d t Ad
Adoption

ti  
& Diffusion

Fall 2013

 Agenda

• Adoption/Diffusion Models
– Modified Bass Diffusion Model (starting Vensim model)

• Delayys
– Response of a Delay (Ch 11.1‐11.2, more in H2)
• Stock Managgement 
– Beer Game Stock Management Example
• Learning Curve (Ch 10, BD pg 337‐8)
• Partial Model Tests
• Robustness under Extreme Conditions

10/4/2013 2
 Historical Data: Upload vdf

VCR Sales,
S l US

US

10/4/2013 3
 Bass Diffusion Model

Potential Adopters

Adopters P A

Adoption
Rate AR
B
+ + R
Market
Saturation Word of Mouth +

+ Probability of
Adoption from Adoption from Contact with
Advertising AA Word of Mouth AW
Adopters PCA
+

-
+ +
Advertising Adoption
Effectiveness a Fraction i

+
B
Social Contacts

Contacts SC Market +

Saturation with
Adopters CA Total Population N
+
+
+ +
Contact Rate c
Modified Bass Diffusion Model

 Total Population, N

Population Not Willing  Potential Adopters Adopters

to Adopt, N – PWA P A

=PNWA

Population Willing to Adopt

=PWA
PWA

N = INTEGRAL(Net Population Increase
Increase, N0)

Adopters = INTEGRAL(Adoption Rate, Initial Adopters)

Potential Adopters = MAX(0, Population Willing to Adopt – Adopters) 

 Stock Management Structure

10/4/2013 7
PART A INTERACTIVE : BEER GAME STOCK MANAGEMENT EXAMPLE

Draw the causal loop diagram (CLD) and write the equation for the “order rate” according to 
assumptions that will be presented. 

Please note that your CLD and equation for the “order rate” should be 
“dimensionally consistent”. 

Inventory
delivery shipments

order rate= ?
 

10/4/2013 8
PART A INTERACTIVE :

MAPPING OF A SIMPLE STOCK‐MANAGEMENT STRUCTURE
 

Draw the causal loop diagram (CLD) and write the equation for the “orders” according to the 
assumptions below. Please note that your CLD and equation for the “orders” should be 
“dimensionally
dimensionally consistent
consistent”. 

Inventory
deliveryy shipments

ASSUMPTIONS:

• No delays in deliveries
• No returns
• Shipments are exogenous
• Firm adjusts orders to bring inventory in line with “desired inventory”
• Seeks to close any inventory gaps over a 2-week period.

• HINT:
HINT What
Wh t should
h ld go iinto
t th
the ““orders”
d ” variable?
i bl ?

10/4/2013 9
 Response of a Delay

Product Retirement= DELAY N(Purchase Rate, Useful Life of Product, 0, 6)

10/4/2013 10
PART B INTERACTIVE : PULSE RESPONSE OF A DELAY

Please sketch your best estimate of “retirements” after a pulse of purchases: 

Installed Base
purchases retirements

How does the Retirements flow look like?
Purchase Rate

Units/Year

10/4/2013 0 0 Time 11
 Price & Learning 
Curve

Ch 10, BD pg 337‐8

10/4/2013 12
 Basic Learning Curve Formulation

• c0, E0 = initial unit costs and

cumulative experience
t
c t  f ( Et ) Et   es ds

ds

 Et 
0

c t  c0   •  = learning curve strength

 E0 
(1 f )c0  c0 (2E0 / E0 )  • Example: For a learning curve
or
or
with f=0.3, =0.5146

  ln(1 f ) / ln(2)  log 2 (1 f )

• f= 0.3 implies a 30% cost

reduction for every doubling of
cumulative experience ( ≈ 0.51)
MIT OpenCourseWare
http://ocw.mit.edu

15.871 Introduction to System Dynamics

Fall 2013

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.