Is Critical Chain Project Management (CCPM) An All or None Deal?

1
Is Critical Chain Project Management (CCPM)An
All or None Deal?
Dr. Russ Johnson, Jonah President, Improvement
Quest, Inc. Loveland, CO 80538 970-581-0075 Imp
rovementquest_at_aol.com
2
What is a project?

• A temporary endeavor undertaken to create a
  unique product or service (PMI-BOK 1996)
• Consists of three major efforts
• Planning
• Execution
• Management

3
What is project management?

• The application of knowledge, skills, tools,
  and techniques to project activities in order to
  meet or exceed stakeholder needs and
  expectations.
• (PMI -BOK, 1996)
• A balancing act between the three project
  commitments
• Scope
• Time
• Budget
• Common elements
• A scheduling mechanism (software and rules)
• Existing management paradigms
• Human behavior

4
Project characteristics

• All projects have many things in common
• They involve high uncertainty.
• They involve three different and perceptually
  opposing commitments Due date, budget, and
  content
• They require different levels of a variety of
  expertise and resources at different times and
  for different amounts of time
• They are impacted by variability within and
  between events

5
Establishing some current reality

• Do we go back to resources and pressure them to
  reduce their time and/or cost estimates?
• Do we hold resources to the scheduled start and
  finish dates for activities?
• Do we sometimes miss entire activities or at
  least dependencies in the planning stages of the
  project?
• Do our projects quickly evolve to having multiple
  critical paths?
• Do our projects seem to always be behind?

6
Are these are fairly typical?

• Existing project work is not complete before new
  projects require a shifting in priorities.
• The organization is too slow responding to
  important opportunities.
• Management feels constant pressure to
  increaseresources to handle peak project loads.
• Promised lead times are longer than desired.
• There are difficulties completing projects on
  time.
• There is too much rework activity.

7
Are these are fairly typical?

• There are difficulties completing projects within
  budget.
• Project scope/content is too often compromised to
  meet dates and/or budget.
• Some projects are abandoned or completed without
  the organization gaining the promised benefit.
• Project Managers and resource managers have
  frequent conflicts about priorities and resource
  commitments.
• Problems in one part of a project cascade into
  other parts of the project and/or into other
  projects.

8
If

• Our reputation is important to us, and
• Our past experience has been a mixed bag that has
  often left our customers less than elated with
  our performance and shaken their confidence in
  us, and
• We need to be able to communicate status to a
  variety of people and be able to quickly and
  accurately predict the impact of change and
  resource availability and assignment issues, and
• Conditions and expectations today are
  fundamentally different than they were when the
  current formal project management approach was
  developed,
• Resources are scarce and heavily shared
• We are moving from competitive bid to
  design-build and negotiated contracts
• It is becoming more and more critical that we
  collaborate rather than combat

9
Then

• Do we need to approach project management
  differently in terms of a formal planning,
  execution and monitoring/reporting process?
• Is there an alternative to the primary formal
  project management style or methodology we use
  today?
• If so, do we have the knowledge and experience we
  would need relative to these alternatives to be
  able to determine which one would be best for our
  needs and situation?

10
A formal project management tool should

• Help us with
• Planning the project
• Executing the plan
• Managing the process
• Set us up for success in both the current
  projects and future projects
• Facilitate collaborative efforts and picking the
  best partners

11
Even more fundamentally, the system we put in
place should

• Improve flow (order to cash in hand cycle time)
  as the primary objective
• Be translatable into practical mechanisms that
  guides the operation when not to produce
• Enable the need to abolish local efficiencies
• Include a focusing process to balance flow

12
Two basic approaches

• Traditional two names same basic concepts
• PERT (Program Evaluation Review Technique)
• CPM (Critical Path Method)
• Contemporary
• Critical Chain
• Lets examine the traditional

13
Traditional

• PERT/CPM
• Developed in parallel in the 1950s
• Needed a way to organize highly complex project
  (Polaris Missile)
• Money and resources were not a problem
• Many modifications over the years to try to
  accommodate the fact that money and resources are
  now a big issue
• Constantly increasing sophistication of software
  and hardware has perpetuated the idea that
  problems with the process can be solved if we can
  just get enough data and process it fast enough
• Tremendous inertia (50 years of common practice)
  and investment causes considerable resistance to
  different approaches

14
Typical Work Breakdown Structure (WBS)
Spring home repair
Paint Inside
Replace Roof
Landscape Yard
Move Furniture
Cover Floors
Mask Windows
Paint
Remove Dead Tree
Plant Shrubs
Install Sprinkler System
Select Paint
Mix Paint
Apply Paint
Activity to schedule
Summary activity
15
How a project looks graphically
Network (PERT View)
Bar Chart (Gantt View)
What do all of these things represent?
16
Traditional (PERT/CPM)

• Mechanisms for building the schedule
• Assumption of infinite resource availability
  before identifying the Critical Path

Initial Critical Path equals longest sequence of
task and path dependencies exclusive of resource
dependencies

• Projected Lead Time

17
Traditional (PERT/CPM)

• Resolving contentions (the real CP)
• Assumption of infinite resource availability
  before identifying the critical path
• Resolving resource contentions from start of
  project to completion giving priority to Critical
  Path tasks

• Increase in Projected Lead Time
• with resource contentions resolved

• Projected Lead Time

18
Assumptions that variation in task times follow a
normal distribution
10
Total series variation square root of thesum
of the variances squared
20
20
20
20
20
/-5
/-2
/-2
/-2
/-2
/-2
50
Reality the potential impact of bad things is
much greater than the potential impact of good
7 10 13 25
10 50 90 90
19
10
20
20
20
20
20
/-5
/-2
/-2
/-2
/-2
/-2
50
10
10
10
10
10
What happens in each situation if a task is
finished late? or early ?
20
Assumptions that variations of actual task times
will cancel each other out
10
20
20
20
20
20
-2/?
-2/?
-2/?
-2/?
-2/?
-5/?
50
10
10
10
10
10
?
52
Late
10
10
10
10
12
Early/late
48
6
12
10
10
10
52
No report
6
10
10
12
10
21
Assumptions that variations of actual task times
will cancel each other out
Total path/integration variation Probability
of event 1 Probability of event 2 Probability
of event 3 Probability of event 4
10
Probability of orange integration task starting
on time if all four feeding task time estimates
are 90 is 66(.9.9.9.9)
10
10
10
22
No consistent method for determining when tasks
with float should start

• Do we start these tasks as soon as possible
  (ASAP), as late as possible (ALAP), or somewhere
  in between?

Float

• Critical Path

23
No mechanism for decoupling the overall project
from individual task and path variations
Today with schedule updated for future
24
Traditional (PERT/CPM)

• Managing the schedule
• Panic sets in with the first late task
• Focus switches from the global perspective of the
  original project goals to a more local
  perspective of task completion
• At the task level, the focus switches from
  content, dependencies, and durations to start
  and end dates
• We hold resources to the original schedule dates
  in place of the necessary conditions that define
  what is needed to start a task and the
  deliverables that define when a task is finished

25
Traditional Project control
No reliable mechanisms to determine when a
project is in trouble or to determine which
activities can afford to wait awhile - No
Visibility of the Impact of Decisions or
Variability!
Importance is placed on achieving task or
milestone conformance to scheduled start and
completion dates rather than deliverables in an
effort to insure or improve project on time
performance.
When faced with conflicting task priorities and
no clear way to determine how much safety remains
in each task, resources and Resource Managers
multi-task to try to minimize the harm to either
task.
26
How does this fit with our fundamental needs of
the system?

• Improve flow (order to cash in hand cycle time)
  as the primary objective
• Be translatable into practical mechanisms that
  guide the operation when not to produce
• Enable the need to abolish local efficiencies
• Include a focusing process to balance flow

27
Contemporary

• Critical Chain
• Developed in the early 1990s
• Utilized the Thought Process tools of the Theory
  of Constraints to analyze the situation and test
  the solution
• Started from scratch taking into account current
  realities of limited resources, money and time.
• Focused not only on the mechanical/software
  aspects of organizing the project but also looked
  heavily at the psychological and human behavioral
  issues of projects
• This includes the negative behaviors of years of
  experience and
• The desired behaviors
• Used aspects of traditional approaches where
  applicable

28
TOCs View of Project Management

• Improving flow
• Minimize bad multi-tasking both within and
  between projects
• Multi-project environment Freeze about 25 of
  the projects to switch focus to finishing work
  rather than starting
• Individual projects
• Plan project from a necessity point of view
  starting with desired outcome and working to
  beginning (Handoff)
• Separate safety from task time to get an
  aggressive but possible time to create a sense of
  urgency

29
TOCs View of Project Management

• When not to work
• Aggregate ½ of the freed safety to create buffers
  
• Buffers correctly set tasks to the as late as
  possible position while still protecting the
  project due date form 95 of the uncertainty
• Relative status of buffers tells which task a
  resource should work on when there is more than
  one open task for that resource this minimizes
  bad multi-tasking
• Full Kit concept delays start of task until all
  necessary inputs are available to minimize
  ineffective workarounds

30
TOCs View of Project Management

• Abolish local efficiencies
• Focus is on completing work as fast and
  accurately as possible and getting it handed off
• What constitutes done is clearly defined so
  that tasks cant be drawn out to fill available
  time (Parkinsons law and 3 minute egg rule)
• The only dates that are important are necessary
  milestones and the project completion promise
  focus is on content not dates
• Review progress by asking how long to finish not
  what percent is complete

31
TOCs View of Project Management

• Focusing process to balance flow
• Buffer Management
• Buffer status (Green, Yellow, Red) directs
  management as to when, and how, to react to
  individual disruptions to flow
• Causes for delays that result in buffer
  consumption are recorded and analyzed to target
  common offender disruptions to flow. (Pareto)
  This can be process/activity focused and/or
  resource focused
• Lean, Six-Sigma and other process improvement
  tools are utilized to systematically and
  continuously improve flow

32
Flow killers

• The assumption that the earlier we start a
  project/task, the earlier it will be finished
• How we estimate durations
• bad multi-tasking
• Missing tasks and/or dependencies in the planning
  stage

33
Multi-project fast lane

• Tactic Freeze 25 of open projects

Level of Management involvement
Rate of completion
Safe Zone
X
Number of open projects or tasks
34
What is estimated task time really composed of?

• Productive work time
• Nonproductive time
• Safety (insurance against uncertainty)

35
10
20
20
20
20
20
/-5
/-2
/-2
/-2
/-2
/-2
50
As risk or non-validaccountability increases so
will safety
Duration 7 10 13
45
Probability 10 50 90
90
?
36
The Sixes Game
You work in my organization You are rewarded
according to your performance versus a
standard The standard is the maximum number of
rolls it should take to get a 6 Your performance
is based on your roll of the dice You will be
measured on the number of rolls it actually takes
you to get a 6
37
How we waste safety

• Parkinsons law - Work expands to fill the time
  available (poor definition of DONE)
• Three minute egg rule - There is an implication
  of poor quality if done too soon as well as
  changed expectations regarding future estimates
• Student syndrome Argue for extension of time
  estimate for all kinds of reasons then, Why do
  today what you can put off to tomorrow
• Multi-tasking - Increases lead time for any
  individual activity as there is unplanned time
  spent starting and stopping

38
Understanding the problem further Task
completion times
The time to complete each task consists of
the time to perform the task, the time
to set up to work on the task (finding
everything and remembering where you left
off) again and again. the time to shut down
or set down time. the time the task had
to wait for the resource while the
resource worked on other tasks.
39
Multi-tasking
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
C
C
C
C
C
C
C
How work was planned
C
How work was performed
Actual resource time
40
The multi-tasking Game

• Our job is to complete three tasks
• Write a column of numbers from 1-26
• Write the letters of the alphabet from A to Z
• Draw a repeating sequence of Square, circle,
  triangle until you have 26 objects
• You must alternate columns as you complete your
  task (number, letter, shape)
• We will time to see how long it takes

41
Is the buffer time already in the estimate?
Have we validated that uncertainty exists and
that we need to protect ourselves from it? Have
we discovered that everyone protects themselves
by adding significant amounts of safety time? Is
it true that the more project experience the more
safety included? Who has control of the safety?
Who should? Have we discovered that a
significant amount of the safety that is built in
to the tasks is, in the end, wasted?
42
Separating the work from the safety

• Traditional Distributed safety time remains
  within the tasks and in the control of each
  resource. Everyone must protect themselves as we
  know Murphy will strike we just dont know when
  and where.
• Critical Chain - Aggregated safety time is
  gathered and placed strategically in the control
  of the project manager but available to the
  resources when needed. We still dont know when
  and where Murphy will strike but we control the
  insurance.

50 for work
25 freed
25 for insurance/buffer
Buffer
Freed safety
43
Negotiations

• Traditional
• Give plans and specs and ask for a price and
  duration
• Try to get their numbers to match your needs
  after the fact
• Each variable is now played against the
  otherLower costlonger timeLess timereduced
  scope
• The resource is in control of the negotiations

• TOC Critical Chain
• Confirm capability to perform scope and
  deliverables
• Determine prerequisites
• Get estimate of duration and first
  availability(90 skewed time likely)
• Split time 5050
• Check fit to schedule
• Ask for cost reminding that others are bidding
  under same circumstances

44
Critical Chain Task Definition and Building of
the Project Network

• Task Definition and Building of the Project
  Network (begin with the end in mind)
• Clear identification of deliverables needed to
  accomplish project goals stated in terms of
  expected outcomes for, or impacts to, the
  organization
• Clear identification of expectations of project
  plan and management is included in the project
  goals definition
• The tasks are defined from the end (future) of
  the project to the beginning (current time)
• Task definition is complete when all starting
  tasks have either already begun or their required
  inputs are already available

45
Defining tasks and relationships simultaneously
Task thataccomplishesthe need or overcomesthe
obstacle
Need orobstacle
Clearly statedobjectives of theproject and
theproject plan
Need orobstacle
Tasks are defined by starting with the project
goal(s) and then working earlier in time until
currently occurring activities are reached by
asking In order to I must immediately have
completed
Need orobstacle
When branches occur, one branch is completed
before starting another
46
Determining the who, what, and how long
?
Tasks definition is complete when the starting
tasks are either already in process or all inputs
needed to begin them are available. The resources
needed to perform the tasks can now be brought in
to help verify that no tasks or needed inputs or
requirements have been missed. This includes
adding detail where what is needed is uncertain
(outsourced activity). Once the work is defined,
the times to complete the work can be determined.
47
Converting to the Gantt view

• There is no need to convert WBS scheduled
  activities to PERT view (precedence diagram) to
  determine task dependencies (task to task and
  paths/integrations) as this was done
  simultaneously with task identification
• This information does still need to be entered
  into our scheduling software
• The network view must still be converted to the
  Gantt view to determine time relationships to
  allow us to identify the Critical Chain and
  immunize the schedule from variability
• The software does this for us

48
How a project looks graphically
Network View
?
Bar Chart (Gantt View)
The need and obstacle elements of the network
view disappear in the Bar Chart (Gantt View) as
they do not contain elements of work or time
49
Mechanisms for immunizing the schedule

• TOC - Critical Chain
• Task, path, and resource dependencies are all
  considered prior to identifying the Critical
  Chain
• Resource contentions are resolved from project
  completion toward start
• Strategically sized and placed buffers allow
  decoupling of the overall project from individual
  task and path variations
• Buffer 1/3 of total path time (task Buffer)

50
Identifying the Critical Chain
Move all tasks to as late as possible and all
dependencies are taken into account (task, path
and resource)
51
Identifying the Critical Chain
Begin resolving resource contentions by moving
competing task to earlier time
52
Identifying the Critical Chain
Continue resolving resource contentions from the
end of the project to the beginning.
(CC)
The Critical Chain (CC)is the longest path of
continuous dependent events including resource,
task and path dependencies.
53
Immunizing the project from variation along the
Critical Chain Sizing and placing the project
completion buffer (PCB)
(CC)
PCB
The project completion buffer (PCB) acts as a
variation absorber and is equal to 50 of the
total task time along the Critical Chain making
it 1/3 of the total project lead time.
54
Immunizing the Critical Chain Sizing and
placing the feeding buffers (FB)


PCB
(CC)


Identify locations where activities feed into
(integrate with) the Critical Chain. The
feeding may be due to task dependencies, path
dependencies or resource dependencies
55
Immunizing the Critical Chain Sizing and
placing the feeding buffers (FB)
FB


FB
PCB
(CC)
FB
FB


The feeding buffers (FB) act as variation
isolators between non Critical Chain activities
and the Critical Chain. They are also equal to
50 of the total task time along the chain of
tasks they are isolating. This may result in
gaps in the Critical Chain and/or the need to
start a non-Critical Chain activity before the
Critical Chain.
56
Project lead time
FB
FB
PCB
(CC)
FB
FB
This Feeding Buffer creates a new resource
conflict but it is not resolved as all times are
estimates so there may or may not be a conflict
during actual execution and if so the buffer will
address it
The actual project start or end date is relative
to time needs of the project Drop Dead or Open
Ended
57
When not to produce
The feeding buffers tell us the right time (not
too early or late) to start non-critical tasks
Project lead time
FB
FB
PCB
(CC)
FB
FB
58
Critical Chain
FB
FB
PCB
(CC)
FB
FB
20-25 time advantage
Promised project lead time
PERT/CPM
59
What about a multiple project environment?
Using a drum, key resource, to stagger projects
The start of the next project would be based on
the placement of the last drum task in the
current schedule and the first drum task in the
next project with a buffer between the last and
first respectively
60
If the red resource is the drum
Project One
DB
Project Two
DB
61
Managing the schedule

• TOC - Critical Chain - Buffer Management
• Focus remains on the global perspective of the
  original project goals
• Progress is reported based on buffer status and
  estimated time for remaining tasks
• At the task level, the focus is on getting the
  job done as soon as possible while maintaining
  original content
• The buffers allow time to plan and react
  appropriately to variation in the schedule

62
Buffers are used to provide focus and early
warning to protect the critical chain and due
date
WATCH
OK Zone 3
ACT Zone 1
PLAN Zone 2
BUFFER MANAGEMENT
Remaining
Project Buffer
100 67
67 33
33 0
63
Project Control - Buffer Management
The mechanism for gathering data provides us a
glimpse into the future - so we can take action
before we are in trouble while also allowing us
time to not be pressured to act when actions are
not necessary. The organization gathers the
information for the status of the buffers in the
following way Each resource that is working on
the project gives a daily status of the time
they estimate they still need to work until the
task is complete. That information is used to
calculate whether any buffer time would be gained
or lost if these time estimates proved
true. This daily interaction is key to
reinforce new behaviors and to provide
opportunities to mentor resources.
64
Project Control - Buffer Management
Project Buffer Status
Day Task To Go Buffer
1 217 10 90
35 168 5 69
51 143 8 55
59 143 11 44
85 122 15 51
101 122 20 30
111 32 4 69
65
CCPM Project ControlWhen and where to work and
when and where to intervene
Do we have a way to determine how much safety is
left if there is a conflict for the
resources? Yes - The Project Buffers! We apply
Buffer Management via comparing buffer statusas
our control mechanism.
66
Who gets the scarce resource?
Buffer status report
OK
WATCH PLAN
ACT
Project A PCB Status
Project A FB1 Status
Project A FB2 Status
Project A FB3 Status
Project B PCB Status
Project B FB1 Status
Project B FB2 Status
Project B FB3 Status
Tasks competing for same resource
67
Process of Ongoing Improvement (POOGI)
Evaluating Resources

• Buffer impact (should track , -, and average)
• Number of charged or credited times they impact
  the buffer
• Charged duration of the impacts
• (Should be viewed from both absolute and relative
  perspectives)
• Causes for charged impacts
• Same problem over and over
• Different issues from time to time
• Poor at identifying potential problems
• Constantly understaffing project
• High levels of rework
• Ability to consistently reduce the cost and time
  to do similar work from project to project

68
How does this fit with our fundamental needs of
the system?

• Improve flow (order to cash in hand cycle time)
  as the primary objective
• Be translatable into practical mechanisms that
  guide the operation when not to produce
• Enable the need to abolish local efficiencies
• Include a focusing process to balance flow

69
Buffer Management
Here is a project that has a Critical Chain of 85
days, a Project Buffer of 43 days and a total
lead time of 128 days.
70
Buffer Management
The first task completes 5 days early, but the
second task takes 25 days to complete. Buffer
Management shows the Project Buffer to be in Zone
3. Net schedule variance 10 days. What
action should the Project Manager take?
Planned -gt actual
71
Buffer Management
After 20 days of work, the resource assigned to
the third task is projecting completion in 20
days. Buffer Management shows the Project Buffer
to be in Zone 2. What action should the Project
Manager take?
Planned -gt actual
72
Buffer Management
After 30 days of work, the resource assigned to
the third task is still projecting completion in
20 days. Buffer Management shows the Project
Buffer to be in Zone 1. What action should the
Project Manager take?
Planned -gt actual
73
Buffer Management
Here is a project has a Critical Chain of 85
days, a Project Buffer of 43 days and a total
lead time of 128 days.
74
Buffer Management
The first task completes 5 days early, but the
second task takes 25 days to complete. Buffer
Management shows the Project Buffer to be in Zone
3. The first task of the feeding path is
accomplished in the duration time. What action
should the Project Manager take?
Planned -gt actual
75
Buffer Management
After 20 days of work, the resource assigned to
the third task is projecting completion in 20
days. Buffer Management shows the Project Buffer
to be in Zone 2. The second task of the feeding
buffer is projecting completion in another 10
days. What action should the Project Manager take?
76
Buffer Management
After 30 days of work, the resource assigned to
the third task is still projecting completion in
20 days. Buffer Management shows the Project
Buffer to be in Zone 1. What action should the
Project Manager take?
77
Organizational Cultural Changes

• Resulting Effects
• Surges or peak demands on resources are minimized
  or non existent
• Natural human behaviors are used to create an
  environment of continuous improvement
• Projects are consistently delivered on or before
  committed dates, often under budget, and with all
  original scope objectives in place
• More projects can be accomplished within the same
  time and with the same resources

78
Key tactics of CCPM that could be used without
full implementation

• Necessity approach to Network building
• Negotiation process
• Critical Chain/path identification from end to
  beginning
• Tracking progress by asking how long to finish
• Minimize multi-tasking can make a conscious
  effort but will not have buffers to direct and
  enforce
• Modification of buffer based priority Each
  activity you make a commitment to has a due date.
  The time between making the commitment and the
  due dale 100. You can work whatever task has
  used the highest of its assigned buffer

79
CCPM as a growth strategy

• Critical Chain offers a new and refreshing
  approach and solution for the undesirable effects
  project managers and organizations can no longer
  afford to suffer from
• Critical Chain is a complete solution that deals
  with both the algorithms of scheduling and the
  impacts of, and on, human behavior
• It is a solution that identifies the correct data
  processing needed to support schedule creation
  and management

80
Conclusions

• Critical Chain offers a new and refreshing
  approach and solution for the undesirable effects
  project managers and organizations can no longer
  afford to suffer from
• Critical Chain is a complete solution that deals
  with both the algorithms of scheduling and the
  impacts of, and on, human behavior
• It is a solution that identifies the correct data
  processing needed to support schedule creation
  and management