Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Cars, Trains, and Kanban: How to Move Faster

Cars, Trains, and Kanban: How to Move Faster

Agile practitioners adopt Kanban with high expectations: decreased cycle time, increased throughput and a clearer look at where projects stand. But where do all these high hopes come from? And how do you make them come true? In this talk, we’ll explore the underlying principles behind Kanban—from systems thinking to lean manufacturing—and draw some fun parallels to everything from car traffic to passenger trains. Understanding these core concepts is the real key to moving faster with Kanban. This talk is for agile beginners, teams who are considering or have recently adopted Kanban, and teams who want to optimize their Kanban process. They’ll learn: What kanban really means—a brief history and how it applies to agile software development Why limiting work in progress (WIP) is essential to speeding up your development efforts + tips for setting WIP limits The right way to pull cards from one state to the next (and why it’s a lot like trains) How to spot bottlenecks in your process, fix them and maximize output.

Peter Kananen

May 11, 2016
Tweet

More Decks by Peter Kananen

Other Decks in Business

Transcript

  1. Cars, Trains and Kanban How to Move Faster Peter Kananen

    Partner, Gaslight teamgaslight.com @teamgaslight
  2. What Kanban is not SCRUM Alternative? Lack of Iterations and

    Cadence Not Planning Process for Building Software
  3. What Kanban really means Modeling the process Limiting work in

    progress Analyzing Flow Pulling, not pushing
  4. Our Issues Limiting work in progress Analyzing Flow Pulling not

    pushing • Poor economic decision making • Long cycle times • Large batch sizes • Over-utilization of capacity • Incurring and blind to the cost of queues • Optimizing for local efficiencies • High WIP • Reduced fast feedback • Limited insight into how to improve
  5. When you don’t model your process… Limiting work in progress

    Analyzing Flow Pulling not pushing • Poor economic decision making • Long cycle times • Large batch sizes
  6. When you don’t model your process… Limiting work in progress

    Analyzing Flow Pulling not pushing • Poor economic decision making • Long cycle times • Large batch sizes
  7. When you don’t limit WIP… Limiting work in progress Analyzing

    Flow Pulling not pushing • Over-utilization of capacity • Blind to the cost of queues • Optimizing for local efficiencies
  8. “In product development, our greatest waste is not unproductive engineers,

    but work products sitting idle in process queues.” - Donald G Reinertsen Which do you want to move?
  9. When you don’t limit WIP… Limiting work in progress Analyzing

    Flow Pulling not pushing • Over-utilization of capacity • Blind to the cost of queues • Optimizing for local efficiencies
  10. When you push instead of pull… Limiting work in progress

    Analyzing Flow Pulling not pushing • Over-utilization of capacity • Incur the cost of queues • High WIP • Reduce fast feedback
  11. When you push instead of pull… Limiting work in progress

    Analyzing Flow Pulling not pushing • Over-utilization of capacity • Incur the cost of queues • High WIP • Reduce fast feedback
  12. When you don’t analyze flow… Limiting work in progress Analyzing

    Flow Pulling not pushing • Increased cycle time • Incurring and blind to the cost of queues • Large batch sizes • Limited ability to improve • Further harm economic decision making
  13. When you don’t analyze flow… Limiting work in progress Analyzing

    Flow Pulling not pushing • Increased cycle time • Incurring and blind to the cost of queues • Large batch sizes • Limited ability to improve • Further harm economic decision making
  14. What Kanban really means Modeling the process Limiting work in

    progress Analyzing Flow Pulling not pushing
  15. Our Issues Limiting work in progress Analyzing Flow Pulling not

    pushing • Poor economic decision making • Long cycle times • Large batch sizes • Over-utilization of capacity • Incurring and blind to the cost of queues • Optimizing for local efficiencies • High WIP • Reduced fast feedback • Limited ability of improvement