This presentation looks at aspects of Theory of Constraints that have worked for me in creative knowledge work activities such as software development. It also looks at others that haven't and features 6 suggestions for how this experience affects the body of knowledge of the Theory of Constraints
FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756
TOCPA 2013 - Towards a Framework for Managing Knowledge Work
1. Towards a
framework for
managing
knowledge work
How TOC influenced my
thinking …
Presenter
David J. Anderson
TOCPA
Utrecht
Netherlands
November 2013
Release 1.0
dja@leankanban.com @lkuceo
And where it didn’t!
Copyright Lean Kanban Inc.
2. How I got into this…
• Published in 2003
• Synthesized
• Feature-Driven
Development
• Lean Product
Development
• TOC
•
•
•
•
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
Five Focusing Steps
DBR
Throughput Accounting
TDD
3. Futility
• I was writing the wrong book
• Lots of (I believed at the time) logically correct,
valid guidance but likely to be almost impossible
to implement
• Humans resist change!
• Installing methodologies (defined or designed
processes) meets with resistance
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
4. My Evolutionary Epiphany
• What was needed was a “start with what you do
now” approach – an evolutionary approach
• An approach where consensus could be formed
around the no.1 problem and what to do about it
• Theory of Constraints and its Five Focusing
Steps provided such a method
• Identify the bottleneck (with consensus)
• Agree (collaboratively) what to do about it
• Resistance (if any) is minimized!
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
6. Motivation for the Kanban Method
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
7. Traditional Change is an A to B process
Designed
Current
Process
Defined
transition
Future
Process
• A is where you are now. B is a destination.
• B is either defined (from a methodology definition)
• or designed (by tailoring a framework or using a model
based approach such as VSM* or TOC TP**)
• To get from A to B, a change agency*** will guide a
transition initiative to install B into the organization
* Value stream mapping, ** Theory of Constraints Thinking Processes
***either an internal process group or external consultants
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
8. Daniel Kahneman has given us a simple model for
how we process information
Learning from
theory
Learning by
Experience
SLOW
FAST
But fast to learn
But slow to learn
System 1
Sensory Perception
Pattern Matching
dja@leankanban.com @lkuceo
Daniel Kahneman
Copyright Lean Kanban Inc.
System 2
Logical Inference
Engine
9. How we process change…
I logically evaluate
change using System 2
I adapt quickly
I feel change emotionally
using System 1
Silicon-based
life form
I adapt slowly
Daniel Kahneman
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
Carbon-based
life form
10. Adopting new processes challenges people
psychologically & sociologically
• New roles attack identity
• New responsibilities using new
techniques & practices threaten
self-esteem & social status
• Most people resist most change
because individually they have
more to lose than gain
• It is safer to be conservative and
stick to current practices and
avoid shaking up the current
social hierarchy
• Only the brave, the reckless
or the desperate will pursue
grand changes
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
11. TOC Observation #1
• Thinking Processes represent managed change
with a transition initiative
• Despite all the counter-measure considerations
with negative branches the “designed
destination” process changes are likely to invoke
(passive-aggressive) resistance amongst a
knowledge worker workforce
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
13. The Kanban Method…
• Rejects the traditional
approach to change
• Believes, it is better to avoid
resistance than to push
harder against it
• Don’t install new processes
• Don’t reorganize
• Is designed for carbon-based
life forms
• Evolutionary change that is
humane
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
14. The Kanban Method…
• Catalyzes improvement
through use of kanban
systems and visual boards*
• Takes its name from the use
of kanban but it is just a name
• Anyone who thinks Kanban is
just about kanban (boards &
systems) is truly mistaken
*also known as "kanban" in Chinese and in Japanese when written with Chinese characters
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
15. Water flows around the rock
“be like water”
the rock represents resistance
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
16. Kanban should be like water*
In change
management,
resistance is from
the people involved
and it is always
emotional (system 1)
To flow around the
rock, we must learn
how to avoid
emotional resistance
* http://joecampbell.wordpress.com/2009/05/13/be-like-water/
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
17. Principles behind the Kanban Method
• Start with what you do now
• Agree to pursue evolutionary change
• Initially, respect roles, responsibilities and job
titles
• Encourage acts of leadership at all levels
The first 3 principles were specifically chosen to
address System 1 objections, to flow around the
rock of emotional resistance in humans
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
18. The Kanban Lens
Kanban asks us to view the world of work through a
new lens
• Creative work is service-oriented
• Service delivery involves workflow
• Workflow involves a series of knowledge discovery
activities
Kanban would be less applicable if a serviceorientated view of work were difficult to conceive or
the work was sufficiently new that a definable series of
knowledge discovery activities had not emerged
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
19. 6 Practices Enable Process Evolution
The Kanban Method
Visualize
Limit Work-in-progress
Manage Flow
Make Policies Explicit
Implement Feedback Loops
Improve Collaboratively, Evolve Experimentally
(using models & the scientific method)
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
20. Any WIP Limited Pull System will work
Bottleneck
(a) D-B-R
(b) CONWIP
(c) D-B-R + CONWIP
(“CapWIP”)
(d) Kanban
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
21. TOC Observation #2
• Arguing about which pull system is better is futile
for knowledge work
• Limiting WIP catalyzes conversations about
problems and how to improve (evolve)
• Limiting WIP also forces conversations about
what to…
• Work on now
• Leave until later
• Abandon altogether
• Limiting WIP forces real options theory thinking
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
23. Flow Efficiency
Dev
Ready
Ideas
Flow efficiencyDevelopment
measures the
Test
Ready
Testing
UAT
5
3
5
3
∞
percentage of total lead time
Ongoing
is spent actually adding Done
value
(or knowledge) versus waiting
Flow efficiency = Work Time
Flow efficiencies of 1-5% are
F commonly reported. *, ** P1
D
> 40% is good!
G
PB
I
GY
DE
Waiting Working
x 100%
Local
Multitasking means time spent
E in working columns is often
Cycle
waiting time
Working
Time
Waiting
Lead Time
* Zsolt Fabok, Lean Agile Scotland, Sep 2012, Lean Kanban France, Oct 2012
** Hakan Forss, Lean Kanban France, Oct 2013
dja@leankanban.com @lkuceo
∞
Lead Time
MN
AB
Waiting
Release
Ready
Copyright Lean Kanban Inc.
Waiting
24. Focus on Delay
• Eliminating sources of delay is the highest
leverage on improving the system
• Delay comes in both special cause (external)
forms such as vendor dependencies and
common cause forms such as queues and
buffers
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
25. Upstream Kanban Prepares Options
Pool
of
Ideas
Biz
Case
Dev
Requirements
Analysis
Ready
for
Engineering
∞
24 - 48
12 - 24
4 - 12
K
L
Min & Max limits
insure sufficient
options are always
available
Committed
4
Development
Testing
3
3
Ongoing
Done
Verification
J
I
D
F
Options
$$$ spent acquiring options
Committed Work
Reject
Abandoned
O
P
Q
Commitment point
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
26. Bottleneck should always be downstream of the
commitment point
Pool
Enginof
Ideas
eering
Ready
Ongoing
2
Pull
F
Development
Analysis
3
Done
Testing
3
Verification Acceptance
Ongoing
Done
Bottleneck workers should never be asked to
work on something that is optional and may be
discarded. This includes any risk analysis (or
estimation in legacy processes) that may be
D
P1
required to assess viability of an option
G
E
PB
GY
Reject
DE
MN
AB
Abandoned
Bottleneck should be here
I
Commitment point
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
3
27. Competing pressure of Last Responsible Moment
versus Upstream Bottleneck
There is a natural boundary
Pool
Enginbetween “the business” and the
eering
Testing
Analysis
Development
Ready
3
3 delivery/engineering function.
3
of
Ideas
Ongoing
Done
2
Ongoing
Done
Verification Acceptance
One generates options, the other
F
Keeping the bottleneck early in the workflow
converts options / delivers
means all downstream functions have slack
commitments
capacity. Reduces our need to manage WIP and
reduces negative effects of variability in demand
D
P1
G
GY
Different governance is required
E
PB
for each
DE
MN
AB
Bottleneck early in workflow
Last Responsible Moment
Commitment point
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
28. TOC Observation #3
• Our delivery/engineering function must be
capacity constrained, in comparison to…
• Our business function must generate excess
options. Options are discarded in proportion to
the risk & uncertainty in the business domain
• Upstream business function must have “slack”
capacity. This is used to generate options that
will be discarded
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
30. Variability in Delivery Rate (velocity)
• It is important to understand the role that delivery
rate plays in long term planning. However, it is not
useful for short-term goal setting due to extreme
variation
• Often velocity exhibits a +/-2x spread of variation
• As a result velocity cannot be used as a short-term
planning tool
See following examples
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
31. Velocity Variation
South African Team from 2011plotted per Sprint (2 weekly)
Mean 29, UCL (+1 sigma) 43 (+1.5x), LCL (-1 sigma) 15 (- 2x)
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
32. DBA Team Velocity
90
80
Trend
70
60
50
Total Velocity
Small support tasks
40
30
Trend
20
10
(not included
in total velocity)
Week of Christmas
0
Mattias Skarin client based in Paris in 2009/2010, plotted weekly
Mean 42, +1 sigma = 55, -1 sigma = 29 (+/- 1.4x)
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
33. Investment Bank, London, Extreme Programming
Weekly Mean 10, Max = 16, Min = 6
Spread (+/- 1.6x)
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
34. TOC Observation #4
• In as many of 90% of cases the bottleneck is
moving about on a day-to-day basis
• Playing “stop the bottleneck” whack-a-mole isn’t a
good use of managerial time and is frustrating for
workers
• Kanban systems solve this problem
• All we need to know is that the bottleneck is
downstream of the commitment point
• Selecting the commitment point is the valuable lever
rather than managing a bottleneck in a narrowly
defined activity
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
36. Evolutionary change has no defined end point
Initial
Process
Evaluate
Fitness
Roll
back
Evaluate
Fitness
We don’t know the
end-point but we do
know our emergent
process is fitter!
Evaluate
Fitness
Roll
forward
Evolving
Process
Evaluate
Fitness
Evalua
Fitnes
Future process is
emergent
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
37. Institutionalize feedback systems to enable
evolutionary change
Operations
Review
System
Capability
Review
Standup
Meeting
manager to subordinate(s)
(both 1-1 and 1-team)
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
38. Fitness criteria are metrics that evaluate
capabilities external stakeholders care about
• If we have a service-oriented view of the
world, and want to evaluate service delivery then
we already know what customers care about
•
•
•
•
Lead time
Quality
Predictability
Safety (or conformance to regulatory reqs)
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
39. If we order a pizza we know what we care about…
• Fast delivery
• lead time from order to
delivery
• Accuracy and quality
• Pepperoni not Hawaiian
• Still warm on delivery
• Predictable Delivery
• If they say “ready in 30
minutes”, we want delivery in
25-35 minutes
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
40. Disintermediate!
Risks, fitness criteria & classes of service should
be explicit & transparent
Operations
Review
Lead time
Quality
Predictability
Expose risk, classes of
service & fitness criteria
at all 3 levels of
Lead time
Quality
feedback
System
Capability
Review
manager to subordinate(s)
(both 1-1 and 1-team)
dja@leankanban.com @lkuceo
Predictability
Standup
Meeting
Copyright Lean Kanban Inc.
Lead time
Quality
Predictability
41. Kanban viewed through a Cynefin*work types
Multiple Lens
Ordered
Kanban systems alone
Unordered
aren’t enough in the
unordered domain
domains
Pragmatic
Holistic
Philosophy
Probabilitic
Quantum
Mechanics
Multiple classes of
domains
service
Complex
Kanban
Method
Emergent Practices
Complicated
Deep
Kanban
Good Practices
System Single work type
Enlightenment
Reductionist
Philosophy
Single class of service
disorder
Chaotic
Kanban
Method
Novel Practices
Not
Applicable
*http://en.wikipedia.org/wiki/Cynefin
dja@leankanban.com @lkuceo
Simple
Simple
Kanban
Best Practice
System
Copyright Lean Kanban Inc.
Deterministic
Newtonian
Physics
42. TOC Observation #5
• TOC’s Five Focusing Steps are a Complex
Domain approach
• Thinking Processes are an unordered
domains, reductionist approach
• Reviving a deep interest in Five Focusing Steps
and an evolutionary approach is vital if TOC is to
be relevant for managing knowledge work
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
43. Other TOC concepts I’ve tried and
discarded
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
44. Critical Chain
• In software development chains were often
arbitrary
• I was serving the method, not it me
• Estimation is speculative
• Decomposition is reductionist
• Hedging for common cause variability meant
buffers were too large!
• Stakeholders only willing to accept 10-15% project
buffer
• Mathematics suggested ~50-67% buffer required
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
45. Throughput accounting
• Time on the bottleneck requires speculative
estimation
• Reductionist
• Not pragmatic
• Bottleneck workers have to estimate future
optional work
• Disruptive
• Doesn’t exploit the bottleneck
• Core conflict!
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
46. TOC Observation #6
• Developing the Kanban Method has been an
evolutionary, experimental process
• Much of what has survived & thrived as
concepts is counter-intuitive
• Analogous mapping of practices from other
domains such as manufacturing has proven
dangerous or fruitless
• Dogmatic attachment to practices leads to
dissonance and denial of actual observations
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
48. Differentiators of the Kanban Method
•
•
•
•
Pragmatic
Probabilistic
Evolutionary/Emergent
Service-oriented
• Not project oriented
• Kanban Method is designed for complex domain
problems
• If humans are involved the domain is complex
• System 1 always wins a cognitive discipline dispute
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
49. A future for TOC in knowledge worker industries
• Too much of TOC assumes silicon-based life
forms in the workplace
• Sort the pragmatic, probabilistic and
evolutionary techniques from the reductionist
and speculative techniques
• Amplify the Five Focusing Steps and other work
that applies to the complex domain
• Ray Immelman’s work on tribalism embraces the
complex nature of humans
• It embraces that System 1 is in charge
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
51. About
David Anderson is a thought
leader in managing effective
software teams. He leads a
training, consulting, publishing
and event planning business
dedicated to
developing, promoting and
implementing sustainable
evolutionary…
He has 30 years experience in the high technology industry
starting with computer games in the early 1980’s. He has led
software teams delivering superior productivity and
quality using innovative agile methods at large companies
such as Sprint and Motorola.
David is the pioneer of the Kanban Method an agile and
evolutionary approach to change. His latest
book, published in June 2012, is, Lessons in Agile Management
– On the Road to Kanban.
David is a founder of the Lean Kanban Inc., a business
dedicated to assuring quality of training in Lean and Kanban
for knowledge workers throughout the world.
dja@leankanban.com @lkuceo
Copyright Lean Kanban Inc.
Traditional change is an A to B process. A is where you are now. B is a destination. B is either defined (from a methodology definition) or designed (by tailoring a framework).To get from A to B, a change agency* will guide a transition initiative to install destination B into the organization.*either an internal SEPG or external consultants
The reason is people resist change. The traditional change model would work perfectly well with silicon-based life forms because the benefits could be argued and agreed with logical. But carbon-based life forms resist change because they don't process it logically but with their sensory perception, their emotional intelligence, the older brain function Daniel Kahneman calls "system 1".
The reason is people resist change. The traditional change model would work perfectly well with silicon-based life forms because the benefits could be argued and agreed with logical. But carbon-based life forms resist change because they don't process it logically but with their sensory perception, their emotional intelligence, the older brain function Daniel Kahneman calls "system 1".
New roles (defined in the methodology) attack their identityNew responsibilities using new techniques & practices attack their self-esteem and put their social status at riskStatistically, most people resist most change because individually they have more to lose than to gain. Probabilistically, it is safer to be conservative and stick to current practices and avoid shaking up the current social hierarchy. Only the brave or the reckless will pursue grand changes.
The Kanban Method rejects the traditional change management method and rejects the installation of a new style of working - a new methodology. It does this because it is better to avoid resistance than to push harder against it.The Kanban Method introduces an evolutionary approach to change that is humane. It is designed to work with carbon-based life forms processing change with system 1. The Kanban Method catalyzes improvement through the use of kanban systems and visual boards (also known as "kanban" in Chinese and in Japanese when written with Chinese characters). It is from the use of kanban that the method takes its name, but it is just a name. Anyone who thinks Kanban is just about kanban (boards & systems) is truly mistaken. The Kanban Method is an example of a new approach to improvement. It is a method without methodology.
The Kanban Method rejects the traditional change management method and rejects the installation of a new style of working - a new methodology. It does this because it is better to avoid resistance than to push harder against it.The Kanban Method introduces an evolutionary approach to change that is humane. It is designed to work with carbon-based life forms processing change with system 1. The Kanban Method catalyzes improvement through the use of kanban systems and visual boards (also known as "kanban" in Chinese and in Japanese when written with Chinese characters). It is from the use of kanban that the method takes its name, but it is just a name. Anyone who thinks Kanban is just about kanban (boards & systems) is truly mistaken. The Kanban Method is an example of a new approach to improvement. It is a method without methodology.
Bruce Lee was a philosopher. He majored in philosophy at the University of Washington, Seattle. His own personal philosophy was heavily influenced by Taoism and Buddhism. He brought this philosophy to his interpretation of Kung Fu and the heart of JeetKune Do.One of his key teachings was "to be like water". Water flows around the rock. The rock represents resistance - in fighting, the resistance is from the opponent.
In change management, resistance is from the people involved and it is always emotional.To flow around the rock, we must learn how to avoid emotional resistance.
Kanban, like JKD, _is_ based on simple principles. As already described, these are: service-orientation service delivery involves workflowand work flows through a series of information discovery activitiesThese principles give us a lens through which to view knowledge work activities and some clues as to the applicability of Kanban. Kanban would be less applicable if a service-orientated view of work were difficult to conceive or the work was without a definable workflow.
Traditional change is an A to B process. A is where you are now. B is a destination. B is either defined (from a methodology definition) or designed (by tailoring a framework).To get from A to B, a change agency* will guide a transition initiative to install destination B into the organization.*either an internal SEPG or external consultants
Kanban closes the learning loop using 3 feedback mechanisms:the standup meeting in front of the kanban boardthe manager to subordinate meetings (both 1-1 and 1-team)the operations review meetingIronically, these have come to known as the Kanban Kata. Ironic because Lee was opposed to Kata as they normally represent an open loop system without learning.
If we order a pizza we want it quickly. We want it to be accurate – if we order a pepperoni, we don’t want a hawaiian. And we want predictability of delivery. If they say they’ll be there in 30 minutes, we expect delivery in 25-35 minutes. And we want the pizza to be still warm.
Kanban closes the learning loop using 3 feedback mechanisms:the standup meeting in front of the kanban boardthe manager to subordinate meetings (both 1-1 and 1-team)the operations review meetingIronically, these have come to known as the Kanban Kata. Ironic because Lee was opposed to Kata as they normally represent an open loop system without learning.