People as Bottlenecks
10 likes5,956 views
The document discusses the concept of bottlenecks in various systems, including physical and non-physical forms that hinder progress. It emphasizes the economic impact of bottlenecks, strategies to identify and manage them, and the importance of flow and variability. Additionally, the document suggests approaches like reducing batch sizes and fostering a learning culture to minimize the effects of bottlenecks.
People as Bottlenecks
- 2. BOTTLENECK: DEFINITIONS 1 A : A NARROW ROUTE B : A POINT OF TRAFFIC CONGESTION 2 A : SOMEONE OR SOMETHING THAT RETARDS OR HALTS FREE MOVEMENT AND PROGRESS B : IMPASSE Merriam-Webster www.m-w.com
- 4. TRAFFIC JAMS IS SLOW TRAFFIC CAUSED BY BAD DRIVERS? Deming 95/5
- 5. TRAFFIC JAMS WHAT ABOUT THESE? Deming 95/5
- 6. BOTTLENECKS PHYSICAL NONPHYSICAL LET’S FOCUS ON THESE UNFORTUNATELY THEY’RE NOT SO EASY TO SPOT
- 8. PEOPLE AND BOTTLENECKS BEING A BOTTLENECK VS INDUCING/CREATING BOTTLENECKS
- 9. BEING A BOTTLENECK NOT AVAILABLE (OVERLOAD, TRAVEL, HIDING) DEFERRING/NOT MAKING DECISIONS (OVERBURDEN, INCOMPETENCE, CYA) CENTRALIZING DECISIONS/NEVER DELEGATE (CONTROL FREAK) LACK OF COMMUNICATION SKILLS SPECIALIST* (MULTIPLE CLIENTS) *formally not a bottleneck but effects are the same
- 10. INDUCING/CREATING BOTTLENECKS OFTEN DUE TO OBSESSION WITH EFFICIENCY AND CONTROL
- 11. INDUCING/CREATING BOTTLENECKS OVERLOADING PEOPLE ASSIGNING MULTIPLE TASKS NO FILTERING (“I WANT EVERYTHING”) HIDING INFORMATION, MAKING IT AVAILABLE TOO LATE (GENERATING EMERGENCIES) CREATING/MAINTAINING SILOS AND CEILINGS MICROMANAGING PUSH
- 12. WEAPONS OF FLOW DESTRUCTION HOW DO YOU SPOT BOTTLENECKS? WARNING: THE MAIN ISSUE IS NOT BOTTLENECKS BUT THEIR ECONOMIC IMPACT
- 13. FOCUS ON FLOW VISUALIZE AND MEASURE
- 14. VISUALIZE FLOW ideas elaborate do delivered validate B C D F E A
- 15. VISUALIZE FLOW B C D F E A ideas elaborate do delivered validate
- 16. MEASURE FLOW B C D F E A ideas elaborate do delivered validate G H I sub-cycle times
- 17. MEASURE # OF ITEMS IN SYSTEM (WIP) B C D F E A ideas elaborate do delivered validate G H I # of items in a specific state = 2 # of items in specific area = 4 total # of items = 9
- 18. SPEED AT BOTTLENECK FLOW = VELOCITY x AREA VELOCITY IS LOWER BEFORE BOTTLENECK
- 19. BOTTLENECKS B C D F E A ideas elaborate do delivered validate G H I gate J K
- 20. BOTTLENECKS B C D F E A ideas elaborate do delivered validate G H I gate J K L
- 21. BOTTLENECKS B C D F E A ideas elaborate do delivered validate G H I J K L gate ITEMS QUEUEING UP EMPTY AREA
- 22. CHARTING CC (CONTROL CHART) CFD (CUMULATIVE FLOW DIAGRAM)
- 23. CONTROL CHART 4 6 8 10 12 14 16 average cycleLme UCL LCL cycle Lme KNOWLEDGE WORK HAS HIGH VARIABILITY
- 24. CONTROL CHART EFFECT OF BOTTLENECK 0 5 10 15 20 25 average cycleLme UCL LCL cycle Lme
- 25. CFD Lme cumulaLve quanLty arrivals departures WIP (avg. queue size) cycle Lme (avg. Lme in queue) arrivals departures AIRPORT CHECK-IN
- 26. CFD Lme cumulaLve quanLty arrivals departures WIP (avg. queue size) cycle Lme (avg. Lme in queue) arrivals departures AIRPORT CHECK-IN
- 27. CFD Lme cumulaLve quanLty arrivals departures WIP (avg. queue size) cycle Lme (avg. Lme in queue) arrivals departures AIRPORT CHECK-IN
- 28. CFD Lme cumulaLve quanLty faster arrivals AIRPORT CHECK-IN
- 29. Lme cumulaLve quanLty QUEUE LENGTH AND WAITING TIME INCREASE WIP cycle Lme AIRPORT CHECK-IN FASTER ARRIVALS CFD
- 30. Lme cumulaLve quanLty WIP cycle Lme CFD AIRPORT CHECK-IN FASTER ARRIVALS arrivals departures
- 31. Lme cumulaLve quanLty WIP cycle Lme CFD AIRPORT CHECK-IN FASTER ARRIVALS arrivals departures
- 32. BACK TO OUR BOTTLENECK EXAMPLE CFD
- 33. BACK TO OUR BOTTLENECK EXAMPLE CFD FLAT LINE, HEIGHT GOES TO ZERO INCREASING HEIGHT (QUEUE)
- 34. CONSTRAIN WIP HIGH WIP -> LONG CYCLE TIMES HIGH UTILIZATION LEADS TO QUEUES (USE SLACK) THE BIGGER THE QUEUES THE MORE THEY COST
- 35. 0" 2" 4" 6" 8" 10" 12" 14" 16" 18" 1" 2" 3" 4" 5" 6" ideas" elaborate" do" validate" delivered" BOTTLENECKS AND WIP CONSTRAINTS QUEUE MOVES HERE TWO FLAT LINES ARTIFICIAL BOTTLENECK
- 36. FOCUS ON THE REAL PROBLEM
- 37. REDUCE BATCH SIZE SMALLER BATCHES REDUCE QUEUE SIZE CYCLE TIME FLOW VARIABILITY FEEDBACK DELAY RISK OVERHEAD BEFORE ATTACKING BOTTLENECKS
- 38. IT IS EASIER TO START THAN IT IS TO FINISH MULTITASKING
- 39. MULTITASKING SUCKS MULTITASKING MAKES YOU LOOK BUSY AND SMART MULTI-TASKERS MAKE VERY POOR MULTI-TASKERS (Stanford study, 2009) “they felt productive because they touched so many different tasks – but when tested against people who focused on one thing at a time, multi-taskers lost and lost big”. Eyal Ophir
- 40. MULTITASKING SUCKS visualphotos.com 1 2 3 4 5 6 YOUR MULTITASKING IS MY BOTTLENECK Jim Benson 1 2 3 4 5 6
- 41. T.O.C. APPROACH 1. IDENTIFY THE CONSTRAINT (CCR - Capacity Constrained Resource) 2. DECIDE HOW BEST TO EXPLOIT THE CONSTRAINT 3. SUBORDINATE EVERYTHING ELSE TO THE ABOVE DECISIONS 4. ELEVATE THE CONSTRAINT 5. RE-EVALUATE E.GoldraA
- 42. LIMITS OF T.O.C. FLOW ISSUES DO NOT ALWAYS IMPLY A BOTTLENECK KNOWLEDGE WORK HAS HIGH VARIABILITY CCR NOT EASY TO FIND IN THIS CONTEXT ADDING CAPACITY AT BOTTLENECK NOT ALWAYS CONVENIENT BOTTLENECK NEVER RUNNING OUT OF WORK NOT ALWAYS CONVENIENT
- 43. LOOKING AT THE WHOLE BOTTLENECKS BEFORE OR AFTER DEVELOPMENT CONCEPT PHASE TOO SLOW? PRODUCING INCOHERENT, CONFUSED INFORMATION? LOTS OF REWORK? STAGING AND DEPLOYMENT ISSUES? LAUNCHING PRODUCTS TOO EARLY? LOTS OF REWORK? SUPPORT
- 44. AGILE HAS INTRODUCED A NEW BOTTLENECK BACKLOG TECH BUSINESS UNFORTUNATELY POTENTIAL v
- 45. AGILE HAS INTRODUCED A NEW BOTTLENECK BACKLOG COMMITTED OR NOT? STATIC OR DYNAMIC? PUSH OR PULL? COST OF DELAY? UNFORTUNATELY POTENTIAL v
- 46. THE PRODUCT OWNER NOT AVAILABLE NOT KNOWLEDGEABLE NOT AUTHORITATIVE NOT COMMUNICATIVE TOO MUCH FOR ONE PERSON?
- 47. YES BUT YET ANOTHER KIND OF BOTTLENECK RESISTANCE TO CHANGE EMBRACING JUST SOME OF IT NEVER CHANGING BOARD DESIGN, POLICIES, WIP LIMITS, ETC. KANBUT
- 48. WRAPPING UP FOCUS ON FLOW BEWARE OF VARIABILITY REDUCE BATCH SIZE CONSTRAIN WIP
- 49. “the greatest waste… is failure to use the abilities of people… to learn about their frustrations…” E.Deming BUT WAIT, THERE’S MORE
- 50. TRUST LEARN KANBAN BUILDS TRUST AND FOSTERS A LEARNING CULTURE
- 51. TRUST != HERO CULTURE TRUST CAN SCALE HEROES DON’T C.Moody
- 52. AM I THE BOTTLENECK? NEVER STOP ASKING YOURSELF