Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 1
SYSTEM DYNAMICS:SYSTEMS THINKING AND MODELING FOR A COMPLEX WORLD
James Paine
System Dynamics Group
MIT Sloan School of Management
IAP 2020 SESSION – JANUARY 13, 2020
01 Welcome!
• Grab some food!
• Introduction to the SD Group at MIT
02 Overview of Systems Thinking
• What is ‘System Dynamics’?
• What is ‘Systems Thinking’?
• Tools of the trade and key concepts
03 Hands on! Fishbank Simulation
• Teams of 4 (+/- 1)
• One laptop per team needed
04 Debrief and wrapup
• Fishbanks debrief
• Tying it into Systems Thinking
• Other SD resources at MIT
Plan for Today
1
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 2
About Me
2
Background
• B.S. Chemical Engineering (UF)
• M.S. Mechanical Engineering (Ga Tech)
• MBA Operations Management and Marketing (WFU)
• Worked for ≈10 years in GE-Hitachi (nuclear engineering), Inmar (reverse logistics and continuous improvement), HanesBrands (product marketing)
James Paine
• MIT Sloan School of Management
• System Dynamics Group, emphasis on Behavioral Operations
Management
Research Interests
• Product development (and failure)
• Supply chain management and cost mitigation via behavioral modeling (BOM)
• Managerial decision making in non-optimal environments
jpaine.mit.edu
System Dynamics Group at MIT
3
https://mitsloan.mit.edu/faculty/academic-groups/system-dynamics/about-us
John Sterman
Jay W. Forrester
Professor of Management
Nelson Repenning
School of Management
Distinguished Professor of
System Dynamics and Organization Studies
Hazhir Rahmandad
Mitsubishi Career
Development Professor
and Associate Professor of System Dynamics
David Keith
Assistant Professor of System Dynamics
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 3
System Dynamics Group at MIT
4
https://mitsloan.mit.edu/phd/students/current-phd-students
Mahdi HashemianB.S. Electrical Engineering;
M.S. Management
James HoughtonS.B. Aeronautics and
Astronautics
Tianyi Li
B.S. Geophysics; B.S. Applied
Mathematics; M.A. Geosciences
Tse-Yang LimB.S. Biology; Master of
Environmental Management
Jose Luis LopezB.S. University of Costa Rica;
INCAE Business School, M.B.A.
Jad SassineM.S. Applied Mathematics
James PaineB.S. Chemical Engineering;
M.S. Mechanical Engineering; M.B.A.
System Dynamics Group at MIT
5
A (very) brief history
“Everything I have ever done has converged to become system dynamics.”
-Jay W. Forrester
at the 1989 International meeting of the System Dynamics Society
MIT-originated field
Created by Dr. Jay Forrester in the mid-1950’s while at MIT
First formalized in 1958 with “Industrial Dynamics - A Major Breakthrough for Decision Makers"
Origins in control theory
Dr. Forrester had background in EE and pioneer in early digital computers. Inventor of Random Access Memory while working on MIT’s WHIRLWIND I general purpose digital computer
Came to understand that social systems are much harder to control than physical systems, and often source of difficulties faced in projects
First major application was stock-flow-feedback structure of GE appliance plant three-year employment cycle, refined ideas of System Dynamics
Broadened beyond corporate management throughout 60’s and 70’s, including resource management such as WORLD2 simulation for Club of Rome
Evolved beyond methodology to thinking framework with applications in numerous fields
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 4
OVERVIEW OF
SYSTEM DYNAMICS
AND SYSTEMS
THINKING
*Portions of the following overview slides
are modified from source material by Drs.
John Sterman, Hazhir Rahmandad, and
Robert Nachtrieb
Open Loop Thinking
7
Identify Problem
Gather Data
Evaluate Alternative
Select Solution
Implement
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 5
Open Loop Thinking
8
Goals
Decisions
State of the
System
We are embedded in a larger system
9
Goals
Decisions
State of the
System
“Side
Effects”
“Side
Effects”
Actions of
Others
Goals of
Other Agents
DELAY DELAY
DELAY DELAY
DELAY
DELAY
DELAY
DELAY DELAY
DELAY
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 6
Systems Thinking Foundations
What is a System?
A system is a set of interdependent parts sharing a common purpose.
The performance of the whole is affected by each and every one of its parts.
10
Social and Economic Systems
Are highly complex systems:
• Dynamic
• Tightly coupled
• Governed by feedback
• Nonlinear
• Limited Information
• Ambiguity and delays in cause and effect
…and are typically more complex than human-made, physical systems.
Systems Thinking and System Dynamics
...is not only tools and but rather framework to help ‘close the loops’ and:
Elicit and articulate mental models and impact of social and
organizational structure
Expand mental models by explicitly accounting for feedback
Test and improve mental models and structure via simulation
Develop shared mental models and more effective organizations
11
The simulation’s purpose is not to ‘be right’ but
rather to help improve mental models and
identify high leverage policy choices
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 7
Systems Thinking Foundations
Structure Generates BehaviorDynamics emerge from the interaction of:• Physics
• Information availability
• Decision rules
12
Mental Models Matter (a lot!)It’s not enough to change the physical structure, information, and incentives.
The Fundamental Attribution ErrorOur first instinct is to blame the people in the system. Almost always this is a low-leverage response
Breaking Away from the Fundamental Attribution Error
13
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 8
Structure Generates Behavior
14
EVENTS
STRUCTURE
PATTERNS OF BEHAVIOR
Structure Generates Behavior
15
EVENTS
STRUCTURE
PATTERNS OF BEHAVIOR
• “Drunk trader caused a spike in oil prices” (NY
Post, 2012)
• “Oil prices keep falling — this is why” (Washington
Post, 12/21/15)
• “OPEC Rumors Continue To Pull Oil Prices Higher”
(Oil Price, Aug 2016)
• “Trump slams OPEC for high oil prices” (Fortune,
4/20/18)
• “Another Sign of Economic Worry: Tumbling Oil
Prices” (NYT, 6/5/19)
• “Oil prices surge after tanker attack in Gulf of
Oman” (CNN, 6/13/19)
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 9
Structure Generates Behavior
16
EVENTS
STRUCTURE
PATTERNS OF BEHAVIOR
• Chronic boom and bust cycles
• Real Prices rising on average
$-
$20
$40
$60
$80
$100
$120
$140
$160
1860 1880 1900 1920 1940 1960 1980 2000 2020
20
18
$/b
bl
Structure Generates Behavior
17
EVENTS
STRUCTURE
PATTERNS OF BEHAVIOR
• Physical structure: o Stocks and flows
o Material delays
o Feedback processes
• Information availabilityo Delays, biases, error, gaps
o Access & transparency
• Mental Modelso Actor goals and incentives
o Time horizon, model boundary
o Misperceptions of feedback
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 10
(Some) Barriers to Learning in Dynamic Complexity
Dynamic Complexity
Limited Information
Confounding variables and Ambiguity
Bounded Rationality and Misperceptions of Feedback
Flawed Mental Models
Erroneous Inferences about Dynamics
Judgmental Errors and Biases
Defensive Routines and Interpersonal Impediments to Learning
Implementation Failure18
‘SYSTEM DYNAMICS’ IS REALLY APPLIED ‘SYSTEMS
THINKING’
19
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 11
(SOME) TOOLS OF
SYSTEM DYNAMICS
Tools and Methods
System Thinking and Modeling is Iterative
Spiral approach, and multiple tools available
System Dynamics is NOT just compartmental models
System Dynamics practitioners use many modeling and simulation toolsets test the implications of hypothesized causal relationships
All Models are Wrong: But some models are useful!
21
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 12
Tools in the Spiral Approach to Model Formulation
• Reference Modes
• Causal Loop Diagrams
• Stock and Flows
• Equation Formulation
• Dimensional Analysis
• Simulation
• Sensitive Analysis
• Policy Testing
22
*Sterman 2000
Results of any step can yield insights affecting other steps
Systems Thinking Tools: Causal Links
23
Production Inventory Shipments
Orders
BookedPriceSalesforce
Population DeathsBirths
+
+
+
-
-
-
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 13
Systems Thinking Tools: Causal Links
24
Ice CreamSales
Murder Rate
+
Ice CreamSales
Murder Rate
AverageTemperature
++
Incorrect! Correct
https://www.tylervigen.com/spurious-correlationsJust for fun:
Systems Thinking Tools: Loops
25
EmplyeeSkill
CustomerSatisfaction
ComplaintsManager Time SpentResolving Customer
Issues
Manager Time SpentCoaching Employees
+
-
-
+
+Reinforcing Loop:
R
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 14
Attractivenessof Market
Number ofCompetitors
ProductPrice
Profits
Systems Thinking Tools: Loops
26
-
+
Balancing Loop:
+
+
B
Systems Thinking Tools: Loops
27
Balancing Loops also called Goal Seeking Loops
Performance
Desired
Performance
Need for
Additional EffortEffort
+
-
+
+
B
Loop Polarity
Right Way: Trace Effect of a Change Around the Loop
Quick Way: Count the ‘-’ connections
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 15
Systems Thinking Tools: Stock and Flows
28
Stock and Flow Diagram (Compartmental Model)
Integral Representation
StockInflow Outflow
Differential Representation
𝑆𝑡𝑜𝑐𝑘 𝑡 = න
𝑡0
𝑡
𝐼𝑛𝑓𝑙𝑜𝑤 𝑠 − 𝑂𝑢𝑡𝑓𝑙𝑜𝑤 𝑠 𝑑𝑠 + 𝑆𝑡𝑜𝑐𝑘 𝑡0
𝑑
𝑑𝑡𝑆𝑡𝑜𝑐𝑘 = 𝐼𝑛𝑓𝑙𝑜𝑤 𝑡 − 𝑂𝑢𝑡𝑓𝑙𝑜𝑤 𝑡 = 𝑁𝑒𝑡 𝐶ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑆𝑡𝑜𝑐𝑘 𝑡
Systems Thinking Tools: Stock and Flows
29
Hydraulic Metaphor
Stock
Inflows
Outflows
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 16
Systems Thinking Tools: Stock and Flows
30
Hydraulic Metaphor
Greenhouse Gasses in the Atmosphere
Greenhouse Gas Emissions
Net Removal
Systems Thinking Tools: Stock and Flows
31
Stocks Flows
Balance Sheet
Wealth
CO2 in Atmosphere
Integrals
Water in a bathtub
Accounts Payable
Income and Expenditures
Cash Flow Statement
CO2 Emissions
Derivatives
Flows through faucet and drain
Vehicle Production
Product preorders
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 17
HANDS-ON
MANAGEMENT
FLIGHT SIMULATOR
James Paine
System Dynamics Group
MIT Sloan School of Management
Fishbanks
*Briefing and debriefing borrowed heavily from Dr. Hazhir Rahmandad and Dr. John Sterman
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 18
Fishbanks• Intro (almost over)
• Fishbanks!• Short break
• Results and Discussion
Winslow Homer, The Herring Net
Fishbanks game by originally developed by Prof. Dennis Meadows, 1986. Web version developed by Prof. John Sterman, MIT Sloan School of Management, with help from Prof. Andrew King, Tuck School of Business, Dennis Meadows, Keith Eubanks, and Forio.com, 2010. Translations available: Chinese, Spanish, Portuguese.
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 19
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 20
Three oceans, Seven teams each
Teams of three to five (aim for 4)
A1
A2
A3 A4
A5
A6
A7
I1
I2I3
I4I5
I6I7
P1P2
P3
P4
P5P6
P7
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 21
Your GoalMaximize your team’s Net Worth at the end of the game.
Net Worth =
Bank Balance+
Value of Fleet
The winner is the team with the highest Net Worth at game end
$
Profit
Profit = Income - Expenses
Fish Sales
Ship Trade Sells
Interest Earning
Harbor Operation
Costs
Ship Purchases
New Ship Orders
Interest Charges
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 22
Income
Catch x Price ($20 per fish)Fish Sales
Ship Sales
Interest Earnings
2%/year if Minimum Bank Balance is greater than zero
Price set by auction
Expenses
Harbor: $ 50/year per shipCoastal Fishery: $150/year per shipDeep Sea Fishery: $250/year per ship
Harbor Operation
Costs
ConstructionNew Ships: $300 each. Charged at end of current year. Delivered the following year
Ship Purchases
Buy a ship at auction. Cost: your winning bid per ship * number bought
Interest Charges
5%/yr if Minimum Bank Balance is less than zero.
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 23
Sequence of Debits and Credits
Ban
k B
alan
ceStart year with bank balance that accumulated through all past years
If you buy ships at auction or from other teams, the cost is subtracted
START AUCTION
You are then charged for the operating costs of your fleet
Your fish catch is calculated and sales are credited as income
DECISION COSTS AND INCOME CALCULATED
The minimum balance is calculated and your account adjusted up or down based on the interest charged or earned
Finally, your account is charged for any new ships you ordered at the beginning of the year
Fishing Fleet
• Initial Fleet =3 Ships/team
• Fleet Growth- Purchase from other teams via auctions
- Order new ships
• Fleet Reduction- Sales to other teams via auctions
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 24
Ordering New Ships
Each year you may order the construction of new ships.
The maximum order is half of your current fleet (initial fleet + auction purchases).
If total fleet is an odd number, your maximum order is rounded up to the next whole number.
Catch
Catch influenced by:Number of Ships,Ship Effectiveness,
Weather
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 25
Recent History of the Fisheries
Ship Effectiveness
25
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 26
• Three oceans: Atlantic, Pacific, Indian
• 7 teams in each ocean, 3-5 people per team
• The oceans are separate
• Fish do not move between oceans
• Ships do not move between oceans
• Conditions identical except for your decisions
FishBanks
Let’s go fishing…
Winslow Homer, Fishing Boats, Key West (1903)
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 27
Login
• 1 Laptop per team (put all others away please)
• Go to: http://bit.ly/fishbanks
•STOP – wait for instructions
Alt link: http://forio.com/simulate/mit/fishbanks/simulation/login.html
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 28
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 29
Fishing Areas
Deep Sea
Maximum Population2000 - 4000 Fish
Annual Operating Cost$250 per Ship-Year
Productivity (Max Ship Effectiveness)
25 (Fish/year)/ship
Coast
Maximum Population1000 - 2000 Fish
Annual Operating Cost$150 per Ship-Year
Productivity (Max Ship Effectiveness)
15 (Fish/year)/Ship
Profit Example
FISH SALES = 25 X $20OPERATING COST
DEEP SEA SUBTOTAL
FISH SALES = 15 X $20 OPERATING COST
COASTAL SUBTOTAL
HARBOR COST
1 SHIP TO DEEP SEA
1 SHIP TO COASTAL
1 SHIP TO HARBOR
PROFIT
$250
$150
- $50
$350
$500- $250
$300- $150
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 30
Regeneration of Fish
New Fish
Per Year
Deep Sea
Coastal
Fish00
Max
Develop your strategy
1. Your goal is to end the game with the maximum possible assets.
2. Discuss within your team what strategies for boat acquisition and allocation you will follow to attain this.
3. Write your strategy down.
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 31
Step 1: Auctions
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 32
Step 2: Buy and Allocate ships
Any Questions?
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 33
Game is on!
Game over!You can take a 10-min break
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 34
Total Assets (Net worth) by team
The Winner is:Pacific 4 with 35,421
• What was your personal intention at the start?
• What was your team’s strategy at the start?
• What did you actually do?
• What did your team do?
• What could you have done differently?
Reflection
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 35
Fishbanks Debrief
The Iceberg: A Metaphor for Systems Thinking
More
Leve
rage
Events
Patterns of Behavior
Systemic Structure
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 36
Loopholes found
In fishing rules
Fishing banned
at Georges BankLocal fishermen fear overcrowding
Lobstermen
Snag record
38m pounds
Codfish
depleted
off MaineRestrictions could
Hurt local fishermen
N.E. lawmakers seek boat buyback ideas
Hearing casts fishery as sinking ship
Canada’s
Gunboat
DiplomacyChrétien to protect
Atlantic fish stocks
Limits may follow
as cod diminishes
in Gulf of Maine
Feds approve boat buyback programHope to thin fishing fleet with $2m in incentives
Event level: the Headlines
Pattern #1: Overshoot and Collapse
Atlantic Swordfish
Catch
0
1
2
3
4
5
1950 1960 1970 1980 1990 2000
Thousand M
etr
ic T
ons/y
ear
Pacific Bluefin Tuna
Catch
0
4
8
12
16
1950 1960 1970 1980 1990 2000
Thousand M
etr
ic T
ons/y
ear
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 37
North Sea Herring Catch
Consider the Cod
• Northern or Atlantic Cod• Long-lived, slow to mature
• Once immensely abundant• Early fishers (e.g., Basque) claimed fish so dense you could walk
from Spain to the New World on their backs.
• John Cabot, exploring Newfoundland in 1497, noted fish so thick they practically blocked his ship.
• Harvest ≈ 250,000 metric tons/yr through 1950s
• Vital in feeding the Old World, in the development of the New World,
…and of Massachusetts:
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 38
The Sacred CodMassachusetts State House
Prevailing Mental Model: Unlimited Abundance
“Probably all the great fisheries are inexhaustible; that is to say that nothing we do seriously affects the number of fish.”
• Thomas Henry Huxley, 1883
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 39
Source: US National Marine Fisheries Service
Source: https://foss.nmfs.noaa.gov/
0
10000
20000
30000
40000
50000
60000
1950 1970 1990 2010
(Updated) US Atlantic Cod Commercial Landings
(Metric Tons/Year)
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 40
Estimated Biomass in 1852
Estimated Carrying Capacity
(Myers et al. 2001)
1200
800
400
0
1850 1870 1890 1910 1930 1950 1970 1980
Total Cod Biomass
Total Cod Biomass Age 5+
Estimated Cod Stocks, Scotian Shelf
(1000 Metric Tons)
Rosenberg et al., Frontiers in Ecology, 2005
Rebranding “Trash” Fish
• Slimehead * “Orange Roughy”
• Patagonian Toothfish * “Chilean Sea Bass”
• Whore’s eggs * “Maine Sea Urchin”
• Mud Crabs * “Peekytoe Crab”
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 41
Aquaculture expanding to feed the world
Source: http://www.fao.org/state-of-fisheries-aquaculture
Overshoot and CollapseWhy the pervasive pattern of overshoot and collapse of fisheries?
Time
An
nu
al fish
ca
tch
Where are the leverage
points for creating a
sustainable fishery?
Where are they not?
What could you
have done
differently?
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 42
1. Renewable resources can be used no faster than they regenerate.
2. Pollution and wastes can be emitted no faster than natural systems can absorb them, recycle them, or render them harmless.
3. Nonrenewable resources can be used no faster than renewable substitutes can be introduced.
Source: Herman Daly (e.g., H. Daly (1990) Ecological Economics 2, 1).
Hu
man
Activity
Eco
syst
em S
ervi
ces
The “Daly Rules”
Feedback structure Governing Renewable Resources
Fish Stock
Net Recruitment Catch
+
Fish Density
Maximum FishStock
+
-
Fractional NetRecruitment
+
+
Catch per Ship
Fleet Size
+
+
+R1
Population
Growth
B1
Limits to
Growth
B2
Fishing
Effectiveness
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 43
Fractional Net Recruitment vs. Fish Density
0
0 1 Fish Density, S/Smax (Dimensionless)
Fra
cti
on
al N
et
Rec
ruit
men
t (1/year)
Impact of Technology on Ship Effectiveness
0
0 1Fish Density, S/Smax
(Dimensionless)
Catc
h p
er
Sh
ip
(Fis
h/S
hip
/ye
ar)
Low Technology
High Technology
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 44
0
0 1 Fish Density, S/Smax (Dimensionless)
Ne
t R
ecru
itm
en
t (fish/year)
Net Recruitment vs. Fish Density
Net Recruitment vs. Fish Density
0
0 1 Fish Density, S/Smax (Dimensionless)
Ne
t R
ecru
itm
en
t (fish/year)
Reinforcing Feedback Dominant (R1)
Balancing Feedback Dominant (B1)
Unstable Equilibrium
Stable Equilibrium
MSY
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 45
Overshoot and Collapse
Simulation:
Fleet (potential
catch) grows
2%/year
Overshoot and Collapse: Examples
• Carrying capacity of the Earth:- climate change, - ozone layer, - ground water, - agricultural soils, - forests, - etc.
• Borrowing to maintain lifestyle
• Speculative bubbles
(housing, tech stocks, art, etc.)
• Abusing trust and good will - Misleading accounting - Telephone Marketing- False online reviews
• Drinking
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 46
“The Tragedy of The Commons”Garrett Hardin. Science 1968; 162:1243-8.
G. Hardin,
1915-2003
Photo: 1986
The Tragedy of the Commons
“No technical solution can rescue us.…”
“Each man is locked into a system that compels him to
increase his herd without limit—in a world that is limited.
Ruin is the destination toward which all men rush, each pursuing his own best interest…”
“We may well call it ‘the tragedy of the commons,’ using
the word ‘tragedy’ as the philosopher Whitehead used it:
‘The essence of dramatic tragedy is not unhappiness. It
resides in the solemnity of the remorseless working of things.’
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 47
“Common Pool Resources”
• Limited Stock and Rate of Renewal• Easily Appropriable (Low barriers to access)• Rival (What you use, I can’t use)
• EXAMPLES:
• Pastures
• Fish
• Forests
• Irrigation
• Clean Air & Water
• Climate
• Roads and Highways
• Parking Spaces
• Views
• Server Resources
• Trust among
consumers
Causes of Collapse (1)
Collapse of the carrying capacity can occur when underlying resources are
Nonrenewable or
Renewable but Consumable or Degradable
Collapse is worse with
• Common pool resources (Tragedy of the Commons)
• Slow or limited regeneration potential
• Tipping points created by positive feedbacks
• Irreversibilities due to e.g.
➢ Trophic cascade
➢ Evolutionary impacts
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 48
Causes of Overshoot (2)
• Long Delays in
Changes in Resource level (Physical/Biological delay)
Measuring resource level (Perception delay)
Understanding causes (Research delay)
Recommending action (Political/social delay)
Implementing policies (Political/social delay)
Policy impact (Physical/Biological delay)
Note: Delays are partly physical and partly political and social. Those with vested interests in the status quo often misrepresent the situation to delay action (e.g. tobacco, lead in gasoline, toxics in food, climate change).
Privatization can help, but has Limits• Enforcement
• Requires effective rule of law to enforce property rights.
• Conflict around equity, fairness.
• High discount rate• Owners may find it in their interest to harvest unsustainably
• Irreducible externalities• You can pen up your sheep, but your fish? Air? Water? Climate?
• Dynamic complexity• Difficulty of measuring resource stocks, consumption, regeneration
• Long delays in detecting and responding to overexploited resource
• Long recovery delays
• Nonlinearities, thresholds, ‘side effects’
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 49
Limits to Privatization• Moxnes (1998): Experiment similar to Fishbanks
but with perfect private property rights: no Tragedy of Commons
• 74% of subjects overbuilt fleets; average fleet 60% above optimal
• Average fish stocks 15% below optimal
• Average subject wealth 46% below optimal
• Subjects: Fishers, Fishing Resource Managers, Researchers
Typical subject behavior:
No uncertainty case
Moxnes, E. (1998) Not Only the Tragedy of the Commons: Misperceptions of Bioeconomics. Management Science, 44(9):1234-1246.
Population
ProductiveCapacity
NetBirths
NetInvestment
Human Activity
Net FractionalGrowth Rate
+
Net Increase inHuman Activity
+
R1
Population andEconomic Growth
Adequacy ofResources
+
-B1
Involuntary Limitsto Growth
Global CarryingCapacity
+
Resource Prices,Social Concern,
Gov't Policy-
Innovation+
Technology
+
Technological"Side Effects"
+
Consumption and
Degradation
Regeneration andRestoration
-+ +
B2Resource
Consumption
B4
TechnologicalSolution
R3
TechnologicalNightmare
RegenerationCapacity
-
+
+
Voluntary Limits-
-
B5Voluntary Limits
to Growth
B3
Regeneration
R2
EnvironmentalTipping Points
DELAY
DELAY
DELAY
DELAY
DELAYDELAY
DELAY
Beyond Fishbanks
Source: John Sterman, MIT Sloan
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 50
Collective Action
Elinor Ostrom (1933-2012): Winner, 2009 Nobel Memorial Prize
in Economic Sciences
Elinor Ostrom’s Optimism
“I would rather address the
question of how to enhance the
capabilities of those involved to
change the constraining rules of
the game to lead to outcomes
other than remorseless
tragedies…”
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 51
• Individuals know the boundaries and limits
• Of the resource (“Common Pool Resource”)
• Of the community of users (“Appropriators”)
• Rules are locally made and adapted to context
• Decisions are made together
• Active measurement and monitoring
• Effective, graduated sanctions
• Accessible mechanisms for conflict resolution
• Latitude from higher authorities to act locally
Design Principles for “Governing the Commons”
Leadership question: how do we enroll and mobilize people to
create these conditions?
OTHER SYSTEM
DYNAMICS
RESOURCES
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 52
Want more?
Classes at MIT
15.871- Introduction to System Dynamics
15.873 - System Dynamics for Business and Policy
102
Population
Susceptible to
Ebola
PopulationInfected with
EbolaInfection Rate
Infectivity
Contacts BetweenInfected and Uninfected
Persons
+
+
Susceptible
ContactsProbability of Contact
with Infected Person
Contact
Frequency Total Population
+
+
-
+
++
R
Contagion
B
Depletion
+
CumulativeReported
CasesNew Reported
Cases
<Infection Rate>
+
Assignment 1 - Section B1
Submitted by: James Paine
Want more?
Books
103All are in the MIT Library!
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 53
Want more?
Articles (personal favorites):
104
All are free when accessing from MIT’s network!
System Dynamics at Sixty: The Path Forward
Selling System Dynamics to (other) Social Scientists
Making the Numbers? “Short Termism” and the Puzzle of Only Occasional Disaster
Nobody ever gets credit for fixing problems that never happened: Creating and sustaining process improvement
Capability Traps and Self-Confirming Attribution Errors in the Dynamics of Process Improvement.
https://sdjournalclub.mit.edu/sites/default/files/documents/Sys%20Dyn%20Reading%20List.xls
Want more?
Websites (personal favorites):
105
Creative Learning Exchange http://www.clexchange.org/
Tom Fiddaman’s MetaSDhttps://metasd.com/model-library/
MIT OCW System Dynamics Self Studyhttps://ocw.mit.edu/courses/sloan-school-of-management/15-988-system-dynamics-self-study-fall-1998-spring-1999/
MIT System Dynamics Journal Clubhttp://sdjournalclub.mit.edu/
The System Dynamics Societyhttps://www.systemdynamics.org/what-is-sd
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 54
(Some) Software
106
(Some more) Software
107
System Dynamics: Systems Thinking and Modeling for a Complex World
1/13/2020
DO NOT DISTRIBUTE: Intended for registrants of the MIT System Dynamics Group IAP 2020 Session only 55
System Dynamics in Action
En-Roads Climate Policy Simulator https://en-roads.climateinteractive.org/scenario.html?v=2.7.6
108
THANK YOU!PLEASE PROVIDE FEEDBACK
SYSTEM DYNAMICS:
SYSTEMS THINKING AND MODELING FOR A COMPLEX WORLD
https://tinyurl.com/SDIAP20Feedback