Kanban from the Inside – Summary – Part 2

Kanban from the Inside  by Mike Burrows

Part II – Models

In Part II, the values take on a supporting role and the centre stage goes to models.

CP6- Improve collaboratively, evolve experimentally (using models and scientific methods)

Models here mean

  • replicating the example of others
  • following some set pattern template or routine that gives structure to one’s action or thinking
  • understanding the world based on a defined set of assumptions
  • expecting certain kinds of outcomes as consequences of those assumptions

The various models represent an important paradigm shift

  • systems thinking – looking at holistic view of systems
  • Theory of Constraints – aligned to system goals
  • Lean – pursuit of flow than maximal utilization of resources
  • Agile using a evolutionary and collaborative approach

Chapter 11 –   Systems Thinking, Complexity and the Learning Organization

Systems Thinking

Broadly Systems thinking is concerned with how systems behave as a whole – taking a holistic view, emphasising the relationships, interactions and influences among components and the behaviours and outcomes that emerge from them. The better we understand our systems relationships with its environment, the more likely it is that we will be able to identify and implement effective interventions.

Donella Meadows in her book on Systems Thinking talks of 12 leverage points to intervene in a system

  1. Numbers – constants and parameters
  2. Buffers – the sizes of stabilizing stocks relative to their flows
  3. Stock and Flow structures – Physical systems and their nodes of intersection
  4. Delays – Lengths of time relative to rates of system changes
  5. Balancing feedback loops – the strength of feedbacks relative to the impacts they are trying to correct
  6. Reinforcing feedback loops – the strength of the gain of driving loops
  7. Information flows – the structure of who does and does not have information
  8. Rules – incentives, punishments and constraints
  9. Self organization – the power to add change or evolve system structure
  10. Goals – the purpose of the system
  11. Paradigms – the mind set out of which the system – its goals, structure rules, delays etc arises
  12. Transcending paradigms


Systems that contain delayed signals and feedback loops can exhibit behaviour that is very hard to predict even when they remain fully deterministic, immune to randomness.  (Beer Game).

The Bullwhip effect or the Forester effect (after Jay Forreseter) explains violent swings observed in supply chain inventories.

Another way to magnify the impact of amplification is via feedback loops – poorly designed feedback loops can be devastating to a system.

Causality and Cynefin Framework

Dave Snowden describes in the Cynefin framework five domains that characterize causality within systems

  • Obvious – where one can easily categorize what we see and respond accordingly perhaps with a best practice whose outcome will be readily apparent
  • Complicated – unobvious to a casual observer, where an expert could understand what is going on sufficiently well – to select an appropriate good practice with a predictable outcome
  • Complex – the domain of emergent behaviour only in retrospect can we understand the impact of our interventions
  • Chaos – where causal relationships disappear entirely even in retrospect
  • Disorder – the state of not knowing which of the other four domains applies

Cynefin allows us to see some interventions as moving aspects of the system from one domain to another – pulling them back from chaos, simplifying the complicated and so on.

Complex Adaptive Systems (CAS)

These are multiple levels of loosely coupled self organizing systems. Adaptive systems are more likely to keep finding configurations that maintain their edge.  Their well designed feedback loops are strategies for evolutionary change.

Knowledge, Learning and the Learning Organization

Deming’s System of Profound Knowledge

  • Appreciation of a system – understanding business systems together with their context of suppliers and customers
  • Knowledge of variation – understand the causes and impacts of variation in quality and the proper use of statistical methods
  • Theory of Knowledge – challenging management to learn by systematically developing, testing and applying theories
  • Knowledge of psychology – understanding people and what motivates them. – also psychology of change

Argyris and Double loop learning

How do organizations learn? Chris Argyris, a noted contributor in this field talks about

  • The Ladder of Inference – which helps explain and deal with very different conclusions that colleagues can reach
  • Espoused theory vs. theory in use – a way to seek out learning opportunities by looking at differences between what people and organizations profess versus what they actually do
  • double loop learning – a simple model that elegantly describes a process of deep learning

Double loop learning – Most learning is single loop – we adjust our action strategies, our immediate goals, plans and moves according to the results we are observing. Single loop learning is very efficient when the goal is to keep a system under good control in a predictable environment.

In Double loop learning, instead of just quickly recalibrating when results don’t match expectations, the learner digs deeper.  Assumptions are challenged and mental models get discarded and new ones come up.

Though it is very difficult to engage in double loop learning, the conditions for frequent double loop learning must be created.

The Learning Organization

Peter Senge   focussed more on individuals and teams – comes out with 5 characteristics of a learning organization

  1. Systems thinking – understanding and managing the organization as a whole
  2. Personal mastery – individuals committed to the process of learning
  3. Mental models – assumptions and theories made explicit, open to enquiring and challenge in an environment of trust
  4. Shared vision – the challenge that unifies and energises the organization – higher purpose
  5. Team Learning – social mechanism that accelerate individual learning

Senge described a learning organization as a group of people working together collectively to enhance their capacities to create results they really care about

Chapter 12 – Theory of Constraints

The Theory of Constraints has these components

  1. Process of Ongoing improvement
  2. Drum Buffer Rope
  3. The Logical Thinking Process
  4. Critical Chain Project Management
  5. Throughput accounting

The Five Focussing Steps and the Process of Ongoing improvement (POOGI)

  • Identify the system’s constraint – the factor most responsible for the systems failure to achieve higher performance
  • Decide how to exploit the systems constraint
  • Subordinate everything else to the previous decisions
  • Elevate the systems constraint
  • Repeat


  1. There is only one constraint that matters – working on other constraints would be wasteful
  2. The constraint typically takes the form of a bottleneck in the process – a step that limits the pace of others in the process
  3. Improving throughput takes precedence over inventory and lead time
  4. Eventually after removing internal constraints, the constraint will reside externally. when this happens the system will typically be constrained by a supply side bottleneck or by a shortfall of market demand relative to capacity.

Drum Buffer Rope

Drum Buffer Rope (DBR) is the production scheduling system of the Theory of Constraints.  DBR plays the role in TOC that Kanban systems do in the Kanban method.

  • The Drum is a work schedule for the constraint, planned ahead of time.
  • The Buffer is the time allowed for each kind of work items, component or subassembly to reach the constraint via the upstream process. Buffer management effectively subordinates the upstream process to the needs of the constraint.
  • The Rope ties the buffer’s input to its output. Work is released into the system according to a schedule constructed relative to the drum – the work schedule for the constraint activity.

The thinking processes

TOC recognizes that improvement implies change and that people find change difficult.  TOC includes a suite of thinking tools designed to address what it calls resistance to change.TOC describes a number of layers of resistance – Efrat Goldratt has the 9 layer model which talks of the 9 layers along with the tools to address the resistance.  The goal of the thinking process is to define a target state that can be delivered through a series of transformations.

Critical Chain Project Management

Critical Chain applies the thinking of Drum Buffer Rope to the problem of planning and controlling projects.

  1. Start with a network of project tasks, their estimates and dependencies
  2. Schedule tasks so that
  • no task is started before its dependencies are fully satisfied.
  • multi tasking is eliminated
  • project duration is minimized
    1. Identify the critical chain – the sequence of tasks that determine the project’s overall duration
    2. Separate task estimates into two components, the expected duration and the remaining safety margin

Move all safety margins to a buffer. These live either at the end of the project (the project completion buffer) or protecting each dependency. (feeding buffers)

Throughput accounting

TOC has its own accounting model – Throughput accounting which aims to reverse these supposed priorities of traditional management that is driven by conventional cost accounting model.

  1. Reduce cost
  2. Reduce required investment or inventory
  3. Increase throughput

The logic of throughput accounting is that reductions in cost, capital investment or inventory that look good on paper in the cost accounting model may be damaging to throughput and therefore detrimental to the interests of the organization.

Chapter 13  Agile   

Chapter 14  TPS and Lean

In 1978, Taiichi Ohno published a book describing the Toyota Production System.  Some of these concepts were hardly known outside Japan

  • Just in time – the radical idea that the right materials, parts and assemblies should arrive when they are needed only as they are needed and in small quantities
  • Autonomation – automation with a human touch – the production line’s early warning system
  • The Andon system – visual indication of trouble combined with the means for ordinary shop floor workers to “stop the line”
  • The Five Whys – a technique for root cause analysis
  • Kanban – card based system by which just in time production is managed

Two books much later – “The Machine that changed the world” by James Womack, Daniel Jones and Daniel Roos  and “Lean Thinking” by Womack and Jones – a whole new set of Japanese terms entered the lexicon

  • Kaizen – continuous improvement through incremental change
  • Kaikaku – radical change
  • Heijunka – production levelling – that is deliberately mixing work on the production line rather than producing similar work in batches
  • Poka Yoke – mistake proofing
  • Gemba Gembatsu Genjitsu – the three reals – the real place where work is done, the real thing and real facts respectively
  • Honshin Kanri – Strategic planning and policy deployment

TPS and Lean in perspective

TPS is a great example of systems thinking.  It starts with a vision – a true north that gives the direction for change.

– Single piece flow, in sequence, on demand, with zero defects, 100% value adding activities and security for the people performing them

Two pillars to support the purpose of Toyota – taking into account the conditions prevailing in post war Japan where land, factory space, plant and materials were in short supply

  • Just in time
  • Respect for people

Lean Improvement

Much of the Lean thinking is built into and around these 5 improvement steps

  1. Identify value – from the customer’s view point
  2. Identify the value stream – the value creating steps in teh process and seek to eliminate the non value adding ones
  3. Create flow – removing delays between those value creating steps seeking smoothness
  4. Establish pull – where work is taken upstream only in response to downstream demand ultimately from the customer
  5. Identify waste – removing impediments to smooth flow reducing delays reducing inventories and eliminating defects at source etc

Ohno identified 7 wastes or non value adding activities

  1. Transportation – source of delay cost or risk of loss or damage
  2. Inventory – materials work in progress finished by undelivered
  3. Motion – damage to people and equipment caused by the production process
  4. Waiting – time spent by work items in inactive states
  5. Over processing – doing more work than is necessary to meet specifications
  6. Over Production – producing work in excess of immediate demand
  7. Defects – effective capacity wasted on bringing inferior work upto the required standard

Lean Startup

Eric Ries’s  Lean Startup model takes product development into areas of extreme uncertainty – where basic things like customers, business models or the basic shape of the product are largely unknown.

Its continuous incremental model is organized around an experimental improvement loop called “Build – Measure –  Learn” – which is highly suited to web based services where concepts like continuous delivery and A/B testing allow products to evolve rapidly.

Kanban and Lean

From a Lean perspective, Kanban has

  • Visual pull systems represented by the values of transparency and balance
  • Respect for people in the form of collaboration, leadership, understanding agreement and respect

Chapter 15 Economic Approaches to flow

Some economic concepts important in Kanban are

  • Cost of Delay – Don Reinertsen
  • Cost of carry – tool used in many traditional industries
  • Real Options – Chris Matts and Olav Maasen

CoD is a way to understand the time dependence of value and a good guide to scheduling decisions.

Cost of Carry measures the cost to the organization of the inventory that it holds.

Options or Real Options takes from the world of banking the idea of an option instrument and applies it in the field of project evaluation.

Chris Matts and Olav Maasen have distilled options into 3 principles

  • Options have value
  • Options expire
  • Never commit early unless you know why

Putting it altogether

  • Put your project portfolio on a diet – aggressively reduce batch sizes towards the likely current ideal
  • Don’t treat all work as alike. At all levels understand and classify work by its urgency profile and control the overall mix of work
  • Within each urgency classification, learn to identify work that carries a high cost of delay – Implement queuing discipline that maximizes this throughput.
  • make your options visible
  • Whether through Cost of Delay or Cost of Carry – understand the cost of WIP. At each level find ways to control it and expect it to keep on reducing
  • Continually seek to reduce transaction costs

Chapter 16 – The Kanban Method

Personal Kanban

In “Personal Kanban – Mapping Work Navigating Life”, Jim Benson and Tonianne DeMaria Barry describe how Kanban can be applied to one’s personal workload.

Jim and Tonianne distill Kanban down to the two practices most relevant to “choosing the right work at the right time”

  1. Visualize your work
  2. Limit your wIP

These correspond to the four values of Kanban – transparency, balance, flow and collaboration.


Scrumban is a name coined by Corey Ladas for what happens when “what you do now” is Scrum and you apply Kanban.  Here

  • The team organizes work according to its “done-ness” – which covers acceptance, deployment and customer validation states
  • Stand up meetings are organized around the board
  • Team pays more attention to the amount of work started but not yet finished and hence finish tasks sooner – explicit WIP limits may be introduced.
  • Greater attention is given to later stages of the process such as continuous delivery
  • Having decoupled releases from sprint planning, the system now accommodates work of different types and speeds. Mid sprint changes become much easier to accommodate, classes of service may be offered
  • Identify riskiest items first and have them broken down as necessary
  • With need for customer validation made more visible, new feedback loops begin to emerge.

Chapter 17  Smaller Models

This chapter covers a number of models that support the concepts such as

  • Little’s Law
  • Satir Change Model
  • Two coaching models – GROW and Toyota’s A3
  • Jeff Anderson’s Lean Change canvas
  • various models of facilitation including games
  • two models of leadership and collaboration – T shaped leadership and triads

Little’s Law

WIP = Delivery rate * Lead Time or Delivery rate = WIP / Lead time

Satir Change Model

Virginia Satir describes changes in 5 stages

  • late status quo – the period before the introduction of foreign element
  • a period of resistance – the foreign element has been introduced but we try to cling onto the status quo
  • Chaos – not dealing adequately with the foreign element and status quo is no longer tenable
  • Integration – the penny drops the pieces fall into place the transforming idea allows the foreign element to be taken on board
  • New Status quo – have a new status quo when the process of integration is complete

Thinking Tools and Coaching Models

  1. GROW – John Whitmore – “Coaching for Performance”  – Goals, Reality Options Will
  2. A3 – to fit into a A3 size paper


  • some context perhaps a visualization of the current process, qualitative or quantitative analysis of the current condition
  • the target condition
  • list of possible counter measures – ideas that could mitigate things when we dont like the current condition
  • a plan that outlines how the chosen counter measure will be implemented
  1. Lean Change Canvas – Jeff Andersen
  • Urgency
  • Target options
  • success criteria
  • vision
  • communication
  • action
  • change participants
  • commitment
  • wins / benefits
  1. Group facilitation and games
  2. Kaner’s Facilitation model – In Facilitator’s guide to Participatory decision making, Samuel Kaner describes the role of the facilitator guiding through 3 stages
  • divergence – ensures plentiful supply of raw material, generating ideas identifying problems and so on
  • then comes the uncomfortable part – the groan zone in which teh raw material is sifted and analyzed. after the energy and creativity of the divergence phase, the discomfort of this phase stems from the lack of a clear way forward
  • finally comes convergence where thoughts become increasingly well organized and agreed upon
  1. Serious Games – In her book “Reality is Broken – Why games make us better and how they can change the world” , Jane McGonigal describes 4 essential properties of games
  • a clear goal
  • a set of rules or constraints
  • feedback or a way to keep score
  • optional participation

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s