conversational systems thinking the power of group engagement with the rigor of system...
TRANSCRIPT
Conversational Systems Thinking
The power of group engagement with the rigor of system dynamics…for
less!
Purpose
• To describe a set of skills and an approach that can improve the quality of thinking…about just about anything…without requiring computer simulation
Agenda
• Overview• Seven Skills• Value Add• Leverage Points• Q&A
Acknowledgements
• I have had the honor to work with several inspirational systems thinkers Barry Richmond
First articulated Seven Thinking Skills First coined term Conversational Systems Thinking
Dana Meadows Developer of Leverage Points concepts
Steve Peterson Continues to push these ideas forward Paper Barry Richmond, System Dynamics and Public Policy
is a great example
Continuum from Archetype/CLD ST to Computer-Facilitated
SD
• Easy to Learn • Nearly anyone can write it
• Some systemic insightsbut can be too generic
• Mostly right brain• Sexy
Archetype/CLD ST Computer-Facilitated SD
• Years to Learn• limited to few
• Most rigorous and uniquesystemic insights
• Heavy left brain• Geeky
• Possible to Learn• Somewhat limited
• More rigorous and more uniquesystemic insights
• Broad appeal - left/right brain
Barry Richmond’sValue per Effort graph
• A lot of clarity can be derived with little time and effort, simply by conversational application of (operational) systems thinking
Conversational use of skills
Simple stock/flow map
Simple model
Complex model
Foc
us o
f CS
T
Effort/TimeRequired
ValueDerived
Big honkin’ model!
Venn Diagram
• A majority of the population (e.g. in organization or policy discussion…dare I say in the population!) can learn and apply Conversational Systems Thinking
• They would become better participants (consumers and contributors) to strategy and policy processes
Apply ConversationalSystems Thinking Skills
95-100%
Apply ConversationalSystems Thinking Skills
95-100%
BuildSimple Maps
40-50%
BuildSimple Maps
40-50%
BuildSimpleModels15-20%
BuildSimpleModels15-20%
Build Complex Models2%
Build Complex Models2%
About the Current Age
"I wish it need not have happened in my time," said Frodo."So do I," said Gandalf, "and so do all who live to see such times. But that is not for them to decide. All we have to decide is what to do with the time that is given us.”
-J.R.R. TolkienOne crucially important thing we can do with the time given us is to increase our systems thinking capacity and of those around us.
Seven Thinking Skills Overview
Seven Thinking Skills of a Systems Thinker*
Systems Thinking skill Contrasts with… Dynamic Thinking – Focusing on patterns of behavior (trends) over time
Static Thinking – Focusing on specific events
System-as-Cause Thinking – Choosing to focus on the system within the organization’s control as responsible for performance issues
System-as-Effect Thinking – Choosing to focus on forces outside the organization’s control as generating the performance issues (creating “victimitis”)
Forest Thinking – Taking the 30,000 foot view of the system
Tree-by-tree Thinking – Focusing on the details, often getting lost in spreadsheets!
Operational Thinking – Looking for causality (How is this behavior generated?)
Factors Thinking – Developing a list of factors associated/correlated with the behavior
Closed-loop (Feedback) Thinking – Understanding the feedback and ongoing process responsible for behavior
Straight-line Thinking – Believing causality is a one-way, linear relationship
Quantitative Thinking – Understanding how to represent non-physical, immeasurable variables in analysis
Qualitative Thinking – Including only those variables believed measurable
Scientific Thinking – Building the most useful, entertainable theory of causality
Proving Truth Thinking – Looking for “The Answer”
Systems Thinking skill Contrasts with… Dynamic Thinking – Focusing on patterns of behavior (trends) over time
Static Thinking – Focusing on specific events
System-as-Cause Thinking – Choosing to focus on the system within the organization’s control as responsible for performance issues
System-as-Effect Thinking – Choosing to focus on forces outside the organization’s control as generating the performance issues (creating “victimitis”)
Forest Thinking – Taking the 30,000 foot view of the system
Tree-by-tree Thinking – Focusing on the details, often getting lost in spreadsheets!
Operational Thinking – Looking for causality (How is this behavior generated?)
Factors Thinking – Developing a list of factors associated/correlated with the behavior
Closed-loop (Feedback) Thinking – Understanding the feedback and ongoing process responsible for behavior
Straight-line Thinking – Believing causality is a one-way, linear relationship
Quantitative Thinking – Understanding how to represent non-physical, immeasurable variables in analysis
Qualitative Thinking – Including only those variables believed measurable
Scientific Thinking – Building the most useful, entertainable theory of causality
Proving Truth Thinking – Looking for “The Answer”
*Barry Richmond
7 ST SkillsContrasting CST with Simulation SD
Skill ConversationalComputer Supported
Dynamic Agreeing on issue, over time, scale of time concerned
Ditto
30,000 Foot (Forest)
Expanding boundary of focus, determining appropriate sectors
Ditto
System as Cause
Focusing on how system generates its own performance, not relying on exogenous causes for behavior
Ditto
OperationalBuilding a mental model that represents causality, using stock/flow maps to represent, may or may not need generic templates to get at essence of activities
Use computer simulation to test plumbing, will use generic templates to specify flows
Closed-loop
Closing feedback loops, removing free lunches. Can do so with connector arrows. In CST will link feedback loops through stock/flow structure to see implications on accumulations. Cannot determine loop dominance.
Use computer simulation to test loop validity and determine dominance.
Quantitative
Include non-physical determinants of behavior. Ensures mental model doesn’t exclude what’s unmeasurable but important.
Need to specify range and build structure to numerate non-physical. Often requires lots of selling to stakeholders!
Scientific Use a build confidence in analysis (map if done) approach. Not proving truth.
Ditto, but building confidence in simulations and results.
Fra
min
gIs
sues
Rep
rese
ntin
gA
ssum
ptio
nsM
inds
et &
App
roac
h
The Skills Part 1Framing the Issue
Dynamic Thinking
• Move from event to over-time perspective
• Power of group agreement on “what’s happening”
• I usually get individuals to sketch out…then have small groups discuss, integrate and prioritize
0
Profit (in Millions of $)
20
-20 Years
“Here & Now” Lens
“Short Time Horizon” Lens
“Longer Time Horizon” Lens
“Much Longer Time Horizon” Lens
Dynamic ThinkingDon’t only look back!
• Very interesting discussions can occur when asking to project how long it will take to achieve some objective.
Revenue
Years0
10
20Let’s double revenue!How will that occur?
Different curves represent different mental models that can be brought to the light of day!
Dynamic Thinking Malnutrition (Peru)
• Here the behavior of interest is…
• Why have improvements in reducing malnutrition stopped!? Understanding why it worked for a while and why it’s stopped can lead to more systemic conversations
Chronic Malnutrition in Peru 1975-2004
2004
20001996
1975
1992
1984
20
25
30
35
40
45
197519761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004
% malnourished, height (under age 5)
Chronic Malnutrition in Peru 1975-2004
2004
20001996
1975
1992
1984
20
25
30
35
40
45
197519761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004
% malnourished, height (under age 5)
More than one of every four children under the age of five suffer from chronic malnutrition in Peru. After steady advances, progress in addressing the problem essentially came to a halt – despite
roughly $300 million a year devoted to resolving the problem.
Unitsof Stuff
producing stufffrom raw materials
ProductionInfrastructure
removingold
utilization ofcapacity
s
s
30,000 Foot (Forest) Thinking
• Boundary expansion• One way to avoid
unintended consequences or poor projections
• Example: Oil Production Top map projects oil
production solely on production infrastructure
Second includes finite resource, price and the impact on demand (consumption infrastructure)
Unitsof Stuff
selling to consumers
producing stufffrom raw materials
months of inventoryremaining
$ perUnit ofStuff
adjusting$ per unit
Consumption Infrastructure
adding
removing
desiredgrowth
desired unitsper consumer
capacity
ProductionInfrastructure
investingin new
removingold
revenue fromstuff
revenue percapacity unit
desiredinvestment
utilization ofcapacity
PotentialRaw
Material
~
productivity ofcapacity
ss
o
oo
s
s
o
s
s
s
s
s
s
s
s
s
s
Production InfrastructureSupply
Consumption InfrastructureDemand
Price
Production InfrastructureSupply
30,000 Foot (Forest) Thinking
Tips
• Resist the temptation to “dive into” everything you know about a part of the issue in any one sector until you’ve captured the 30,000 foot view
• Use Sector Frames to help categorize
• Expand boundary by questioning clouds in a stock/flow map
The Skills Part 2Representing the Mental
Model
Operational Thinking
• The stock/flow language is the operational language of Conversational Systems Thinking
• It becomes a great framework for… Representing the main chain (core infrastructure)
Determining the essence of how activities are generated (stock vs. flow)
Identifying levers through generic templates
Main Chains (or Core Infrastructures)Examples of stock/flow maps
HealthyPopulation
At RiskPop
ChronicPop
births
becomingat risk
devlopingchronic
conditions
dying
losing riskystatus
Population Health
InexperiencedStaff
ExperiencedStaffbecoming
experiencedhiring
inexperiencedattrition of
experienced
CandidatePool
Workforce Development
Treatment CapacityCapacity in
DevelopmentTreatmentCapacity
beginningdevelopment
comingon line
removingfrom use
PersonalHabits
losing goodhabits
building goodhabits through medical
& public awareness
Personal Health Habits
Cash inAccount
revenue expenses
Financials
Main Chains (or Core Infrastructures) An example from the CDC
CDC was mandated by Congress with two measurable goals re: diabetes policy…1. Rate of diagnosing diabetes2. Reducing Prevalence3. Reducing deaths (dying)
What do you think of this strategy? Sufficient?
UndiagnosedPrevalence
DiagnosedPrevalence
diagnosing
dying
PopulationwithoutDisease
developingbut undiagnosed
developingand diagnosed
1
2
Turns out these goals are incongruent!
Prevalence & deaths (measurable)
must increase ifdiagnosing increases!
3
Main Chains (or Core Infrastructures)Homework for Session 3: Map the expenditure
An Example from working with GA Legislators
Proposed/ Current
Expenditure
Impacts flow (a-
d)
Why? How long till impact seen? Why?
Potential Unintended Consequences
School fitness programs
a & b Develops healthy behaviors in students; prevents the onset of obesity
Most visible after 20-40 years. Takes till that long until students reach “complications” ages
Funding from other programs
…. …. …. … …
Healthy& Safe
Population
At RiskPopulation
Afflicted &Chronic
Populationbecomingat risk
returning tohealthy & safe
becomingafflicted
dying fromchronic
complications
a.
b.
c. d.Map current or desired initiatives onto main chain
Main Chains
Tips
• Although you can map non-physical, it’s likely best to make core infrastructure physical
• A structure mapping the building Support for a program at Boeing becomes a chain of Advocates and Resistants
• Non-physical variables can then drive these more physical flows
Aware of& May
Use DLAdvocates
for DL
Resistantto DL
decidingagainst
decidingfor
becomingaware losing
advocates
gaining advocatesfrom resistant
Activity (Flow) Generation
• Understanding whether a flow is generated primarily from a stock or flow can lead to insights in where focus has been and what might be done
• Have conversations with stakeholders regarding the essence of how something is generated
Activity (Flow) GenerationGeneric Templates
Templates• Stock-based
Resource Compounding Draining Gap-
adjustment
• Flow-based Co-flow
criticizingCriticismDelivered
losing
SelfEsteem
self esteem per criticism
Co-flow
InfectedPopulation
becominginfected
infectionrate
Compounding
Charge onCapacitor
discharging
dischargerate
Draining
Oil Wells
pumping
well productivity
Barrelsof Oil
Resource
Perceived
Quality
Current Quality
changing
adjustment fraction
or adjustment time
perception
gap
Gap-adjustment
Using generic templates helps identifies levers
Example: Auto manufacturer
Strategy to reduce auto-related fatalities
• Manufacturer classified initiatives as active vs. passive
• Turns out using generic templates made it easier to classify
CumulativeFatalities
annualfatalities
accidentsper year
fatalities peraccidentmiles driven
per year
accidentsper mile
Vehiclesmiles pervehicle
per year
Co-flows
External resource
Features by automakers to reduce fatalities can reduce:
• Accident probability
• Fatality probability
But can also reduce:
• Autos (UIO)• Miles driven (VMT)
Closed-Loop Thinking
• Showing feedback loops that drive changes in dynamics Reinforcing Counteracting
(Balancing)• Can do so just as
well in stock/flow maps…
• And you don’t lose the main chain nor levers identified
Units ofStuff
Price
producing stufffrom raw materials
selling toconsumers
AvailableRaw
Material
ProductionInfrastructure
and its Utilization
units produced perunit of production
infrastructure
Demand
add 10% more!
Demand Balancing Loop
Supply Balancing Loop
s
oLimit to Growth Balancing Loop
s s
s
o
s
Can do with/without generic templates
Healthy and Chronic Population discussion
CST Exercise with GA Legislators
• Current treatment strategies might create a vicious cycle (Reinforcing Loop)
More demand & spending for treatments on the Chronic Population means less spending on the At Risk Population means more Chronic Population needing treatments)
• What might be the most effective way to limit this loop’s impact?
Reinforcing Loop
Healthy& Safe
Population
At RiskPopulation
Afflicted &Chronic
Population
becomingat risk
returning tohealthy & safe
becomingafflicted
CumeSpending
spending onafflicted & chronic
$ per afflicted& chronic
dying fromchronic
complications
demand for spendingon afflicted & chronic
mortality %
The Skills Part 3Framing the Issue
Scientific Thinking Process
• In applying CST use same confidence building process you would with a model Building confidence not proving
truth
• Iterate through applying skills• Revisit reference behavior• When building maps start small -
core infrastructure• Best if core infrastructure is
physical• Add a piece of structure, discuss
implications, revise and add from there
Observe/IdentifyIssues
Build/ReviseCausal Theory
Develop/TestStrategies
Communicate &Disseminate Solutions
& Insights
Value Add in Practice
Pulls out concept of systemic orchestration
• Operational strategy maps identify where systemic orchestration will determine strategic effectiveness
• Knowing timing and magnitude is just as important as knowing levers!
CumulativeSales
SF InTraining
Info Systemin Development
Large AccountSales Force
Info System
hiring newsf
becomingfully
productive
beginning coming online
productivity
Large Account Sales
6 month delay
3 year delay
Identify Unintended Consequences Rigorous approach to improving mental models
Policy QuestionHow can we reduce the number of people dying with
AIDs?
If our primary strategy is to develop interventions that increase the lifespan of the Infected Population, what will happen to the rate of “dying” over time?
Annual deaths
Years
InfectedPopulationbecoming
infected dying
average timeinfected before
dying
becoming infected %
Expanding boundaries leads to…
InfectedPopulationbecoming
infected dying
average timeinfected before
dying
becoming infected %
RiskyBehavior
Populationbeginningrisky behavior
HealthyPopulation
births
beginning riskybehavior %
Able to Identify Leverage Points
• Conversational Systems Thinking – because it applies stocks and flows – allows better analysis using Donella Meadow’s “places to intervene in a system” framework
• Her list of leverage points (next page) works from constants (e.g. productivity terms) down to rules and mindsets that determine the system
• CST allows you to examine the gamut of her list
Identifying PoliciesPlaces to Intervene in a system*
12.Constants, parameters, numbers (such as subsidies, taxes, standards)
11.The sizes of buffers and other stabilizing stocks, relative to their flows
10.The structure of material stocks and flows (such as transport networks, population age structures)
9. The length of delays, relative to the rate of system change
8. The strength of negative feedback loops, relative to the impacts they are trying to correct against
7. The gain around driving positive feedback loops
6. The structure of information flows (who does and does not have access to what kinds of information)
5. The rules of the system (such as incentives, punishments, constraints)
4. The power to add, change, evolve, or self-organize system structure
3. The goals of the system2. The mindset or paradigm out of
which the system - its goals, structure, rules, delays, parameters - arises
1. The power to transcend paradigms*Donella Meadows, Sustainability Institute and author of The Global Citizenhttp://www.sustainer.org
In order of increasing effectiveness
By using CST, you can more easily explore implications of policies with regard to leverage
Summary
Dana Meadows said:
We have just enough time…
starting now!
A Brief Bibliography
Resource Author/Publisher NotesSeven Thinking Skills Barry Richmond, Pegasus
Communications (2000)Introduced concept of sevent thinking skills. Excellent resource guide.
Introduction to Systems Thinking
Barry Richmond, iseesystems (2002)
Extremely accessible work produced by one of the leading lights of the field.
Strategic Modelling and Business Dynamics
John Morecroft, Wiley Press (2007)
One of the most accessible modeling books. Perhaps too business-y for the layperson.
The Strategic Forum Chris Soderquist, Berrett-Koehler (2000)
Description of using modeling for strategic planning
The Systems Thinker Newsletter
Pegasus Communicationswww.pegasuscom.com
Newsletter format. Broad range of topics covered. Rigor of treatment varies.Pegasus website is a clearinghouse for a host of dynamic modeling/system dynamics/systems thinking resources.
Operational Strategy Mapping
Soderquist, C. and Shimada, M., Pegasus Communications
Feature article on using strategy mapping at Boeing
The Fifth Discipline Peter Senge, Doubleday (1994)
Qualitative treatment of broad array of dynamic modeling skills and topics. Popularized “systems thinking” approach. Many follow-on “field books” published over the years.
Industrial Dynamics Jay Forrester, Pegasus Communications, (1961)
Classic book on system dynamics, written by the founder of the field.