6/5/2011

48260 Engineering Project Critical Chain Project Management (CCPM)

Management References and Sources
 E.M. Goldratt, Critical Chain, North River  Critical Chain Project Management
Press, 1997.
Critical Chain Project is a method of planning and managing
 Newbold, R, Project Management in the Fast
Management – Lane The St
Lane, St. Lucie Press,1998.
Press 1998 projects that puts more emphasis on
 Leach L P, Critical Chain Project Management, the resources required to execute
An Overview 2nd edition, Artech House, 2005 project tasks.
 These are available in the library  This is in contrast to the more
 CCPM presentation by Ms. Lisa Stensdale traditional Critical Path and PERT
methods, which emphasize task order
1 2
and rigid scheduling.

Usually….. Estimation Variation Where does all the safety go?

Create Then estimate becomes commitment 50% confidence  Parkinson's law: work Level of
WBS expands to fill the time Effort
available
• Allowance for
uncertainty, Murphy
 Student syndrome:
90% confidence
etc. “why start now if it isn’t
•Dedicated resource
due until next week?” Assigned
How long Due Date
will these
3  Delays are passed on – Date
take? 10 weeks
6
•Non - dedicated
resource not gains
+ Allowance for Murphy •Additional multi-
tasking allowance
 Multitasking
+ other jobs 10
Ask
For + interruptions
Task ~ 3 weeks And so a 3-week “just-in-time” task becomes a 10 week “just-in-case” task
Estimates
Estimate 4 5 6

Multi-tasking Goldratt’s Response? Constraints on Projects

Tasks Task A 3 days
assigned to a Task B 3 days
Lost
Productivity  These may be…
resource 3 days
The project critical path (CP)
Task C
 Critical Chain Project Management 
Should happen: Task A Task B Task C
 Resources in conflict – on CP perhaps
3 days 3 days 3 days  Application of his Theory of Constraints to
Could happen: A B C A B C
project management  Dates that cannot be moved
6 days
6 days
6 days
Does happen: A B C A B C
7 days
7 days
7 days

Source: Unknown 7 8 9

1
 6/5/2011

Critical Path Critical Chain Critical Chain (CC)

The sequence of dependent events that
Critical Chain

A1=6 A2=12 D1=7 A1=6 A2=12
prevent a project from completing in a
B1=3 B2=4 B3=11
B1=3 B2=4 B3=11 shorter time interval.
What if these tasks
use the same C1=8 C2=14 D1=7
 The CC is determined by:
C1=8 C2=14 resource ?  Task technical dependencies, and
 Resource dependencies
Feeder Paths
 The key constraint of your project.
Critical Path

10 11 12

Creating a Critical Chain

CP vs CC? Critical Chain Schedule
 The critical path is the longest path  Build the WBS
 If resources are always available in unlimited Identify dependencies
through the project network and has quantities, then a project's critical chain is

 Gather estimates with 50% confidence
the least amount of slack
slack– the longest identical to its critical path.
path  1 set – 50%
chain of tasks based upon task  2 sets – 50% and 90% - check for difference
dependencies.  1 set – 90% - and cut in half to get to 50%
 Build your project schedule using LATE
 The critical chain is the longest chain of FINISH (rather than “early start”)
tasks that considers both task  Resolve resource contentions (leveling)
dependencies and resource  Start with the resources on the critical chain first!
dependencies. 13 15
Source: Lisa Stensdale

Creating a Critical Chain

Schedule An Example Continued
 Identify the critical chain  Start with the project network  Use average (50% confidence) duration
 Exploit the critical chain to identify any re- for tasks – or use 50% of 95% duration
sequencing that can shorten the duration
 Add
dd theh Project Buffer
ff to the
h end d off the
h A 10
A:10 B6
B:6 C 12
C:12
critical chain A:10 B:6 C:12 E:20
D:10 F:12
 Add Feeding Buffers to all non-critical chains
Remove safety
that feed the critical chain – step 3 of TOC E:20
A:5 B:3 C:6 to shorten tasks
 Resolve new resource contentions E:10
D:10 F:12 D:5 F:6
16 17 18

2
 6/5/2011

Continued What About the Safety? The Project Buffer

 We do not build in any contingency at the Task level Critical Chain
 Find critical path – and do the resource  We move all the contingency to the Project level - we call Committed end
leveling this the Project Buffer A1=6 A2=12
date

B1=3 B2=4 B3=11

Individual Tasks can now be late without making the
Project late.
C1=8 C2=14 D1=7

The Project due date is protected as long as Project Buffer

the accumulated delay along any one chain
is less than the project buffer Feeder Paths

19 21

Feeder Buffers Protect the CC Resource Buffers

 Feeders Buffers Resource Alerts

 Protect the critical chain from being Project

impacted by non-critical chains Buffer
 Place at the end of non-critical chains Project
P j t
before they connect to the critical chain i.e. Buffer
between the feeder paths and the critical FB
chain. Protect the Critical Chain by de- FB
coupling the feeding chain with a
FB = Feeding Buffer Feeding Buffer.

22 23 24

How Much Buffering? Buffer Placement Continued The Example with Buffers
 There are various ways to size the  Insert buffers – project, feeder and
buffers: resource – 50% of path duration A1=3 A2=7
 50% of the sum of task durations on the  Project and Feeder Buffers – additional
chain safety time added to the network
B1=2 B2=3 FB=3 B3=6

Conservative
Resource buffer – time to alert a resource


 SSQ that it will be required to work on tasks C1=4 C2=8 FB=6 D1=4 PB = 10
 Square root of the sum of the squares of the
differences for the tasks along the chain
 Has issues

25 26 27

3
 6/5/2011

Recap: Manage/control by: Critical Chain Project Tracking

 Watching the Buffers  For each task, collect:
 Actual Start Date
Safety
 Reporting  Days Remaining
Safety Safety End
 Buffer usage  Actual End Date
Calculate the impact on the project and feeding
Safety

 Action Thresholds 
buffers
Remove the safety from the individual tasks
 Priority of tasks  If a task is late or expected to be late, consume days
Put it where the project benefits from the appropriate buffer and DOCUMENT WHY
 If a task is early (completes ahead of estimate),
replenish the appropriate buffer
Earliest
 Monitor trends in buffer utilization for issues
Possible Expected
Finish
 Use a fever chart to assess project health
Finish 28 29 30
Source: Lisa Stensdale Source: Lisa Stensdale

100.0%

90.0%

Key Notions Summary

December
80.0%
Execute contingency plan
70.0%

 Don’t schedule in conflicts – e.g.  A Quick overview of Critical Chain Method

Buffer Utilization

60.0%

50.0% resource conflicts  Read Goldratt’s book for more information

Buffer the project – not individual tasks
October

%B

Actual case data now becoming available

40.0%
Identify contingency plan 
30.0%
 Protect the critical chain
 Can be applied to multiple project scenarios
20.0%
All is good  Buffer the resources on the critical
10.0%
July chain  Project is monitored by looking at % CC
0.0%
complete, and other items.
27/09/2005
4/10/2005
11/10/2005
18/10/2005
25/10/2005
1/11/2005
8/11/2005
15/11/2005
22/11/2005
29/11/2005
6/12/2005
13/12/2005
20/12/2005
27/12/2005
3/01/2006
10/01/2006
17/01/2006
24/01/2006
31/01/2006
7/02/2006
14/02/2006
21/02/2006
28/02/2006
7/03/2006
14/03/2006
21/03/2006
28/03/2006
4/04/2006
11/04/2006
18/04/2006
25/04/2006
2/05/2006
9/05/2006
16/05/2006
23/05/2006
30/05/2006
6/06/2006
13/06/2006
20/06/2006
27/06/2006
4/07/2006
11/07/2006
18/07/2006
25/07/2006
1/08/2006
8/08/2006
15/08/2006
22/08/2006
29/08/2006
5/09/2006
12/09/2006
19/09/2006
26/09/2006
3/10/2006
10/10/2006
17/10/2006
24/10/2006
31/10/2006
7/11/2006
14/11/2006
21/11/2006
28/11/2006
5/12/2006
12/12/2006
19/12/2006

Status Date

Source: Lisa Stensdale 31 32 33

Other Places to get

AGILE Traditional Approach to PM
Information
 Excellent white paper AND Focus on upfront planning
 http://www.focusedperformance.com/articles/ccpm.ht Plan, execute, control and therefore a
LEAN
ml
 Real-life p
project
j example
p high probability of success
 http://pqa.net/ProdServices/ccpm/W05002005.html#In High degree of predictability needed

troduction
Yahoo Discussion Group PROJECT Scope and technology known
MANAGEMENT
 http://finance.groups.yahoo.com/group/CriticalChain/
 March 2003 Cutter Journal – lots of great articles
 http://www.cutter.com/research/freestuff/itj0303.pdf REFERENCE – Chapter 17 Gray and Larson, 2010, Project Management – the Managerial
process 5th edition, McGraw-Hill Irwin
 Book: “Critical Chain Project Management”, 2nd
edition, by Lawrence Leach
Source: Lisa Stensdale 34

4
 6/5/2011

Traditional PM v Agile BUT AGILE Project Management

Scope not known fully – Relies on incremental, iterative
development cycles
Many customers with many different
to complete less-predictable projects.
needs that may be difficult to articulate
– Is ideal for exploratory projects in which
New technology requirements need to be discovered and
new technology tested.
– Focuses on active collaboration between
the project team and customer
representatives.

Traditional v Agile Agile PM Iterative and Incremental

Traditional Agile
– Is related to the rolling wave planning
Design up front Continuous design
and scheduling project methodology.
Fixed scope Flexible
• Uses iterations (“time boxes”) to develop a
Deliverables Features/requirements
workable pproduct that satisfies the customer
Freeze design as early as possible Freeze design as late as possible and other key stakeholders.
Low uncertainty High uncertainty
• Stakeholders and customers review progress
Avoid change Embrace change and re-evaluate priorities to ensure alignment
Low customer interaction High customer interaction with customer needs and company goals.
Conventional project teams Self-organized project teams • Adjustments are made and a different iterative
cycle begins that subsumes the work of the
previous iterations and adds new capabilities to
the evolving product.

Advantages of Agile Limitations and concerns LEAN – based on lecture prepared for 48260 by Philip Fyvie 2009

– Useful in developing critical breakthrough • It does not satisfy top management’s need Developed by Taiichi Ohne at Toyota in
technology or defining essential features for budget, scope, and schedule control. 1950’s
– Continuous integration, verification, and • Its principles of self-organization and close Addressed significant flaws in practices of
validation
lid ti off the
th evolving
l i product.
d t collaboration can be incompatible with
mass production in US automobile
corporate cultures.
– Frequent demonstration of progress to industry
increase the likelihood that the end product • Its methods appear to work best on small
projects that require only five-nine dedicated Toyota Production System and Toyota
will satisfy customer needs.
team members to complete the work. Product Development System
– Early detection of defects and problems.
• It requires active customer involvement and
developed and gave Toyota significant
cooperation. strategic advantage

5
 6/5/2011

Basically Lean Principles Basically

• Lean approaches help us to do more • Eliminating waste • Lean approaches help us to do more
with the same amount of resources to • Build Quality in with the same amount of resources to
produce outcomes of superior quality • Create Knowledge produce outcomes of superior quality
• Move from to concept to cash as quickly • D f commitment
Defer i
as possible • Deliver fast
• Respect people
• Optimise the whole

Some thoughts on implementing LEAN Remember ALSO

• The ideas are expressed briefly and concisely.
• - Discard conventional, fixed ideas for production
Critical Chain Consider the implications of introducing
• - Think of how to do it – not why it can not be done Agile these paradigms on:
• - Do not make excuses. Start by questioning current practices
• - Do not seek perfection. Do it right away even if only 50% of target is Lean
met
• - Correct mistakes once Are only some of the variations on Project • Organisational structure
•
•
- Do not spend money. Creativity before capital
- Wisdom is brought out when faced with difficulties
Management that exist. • Organisational culture
• - Ask ”WHY” five times and seek root causes • Incentives and rewards
• - Seek the wisdom of ten people rather than the knowledge of one
• - There is an infinite capacity to improve everything LOOK for more in your workplace
• You can add to this your own thoughts as you find out what works
while doing it.

Revision Lecture
•
•
Guest lecture – APESMA
Comments on the Group Project
GROUP GROUP
• Reflection
PROJECT
• SPARK
PROJECT
Reflection and
• SFS – online
• Revision
•
•
Examination
Individual Questions
Comments SPARK

6
 6/5/2011

Week 1
Project Lifecycle
• Management

REVISION • Project
• Project Management
– Background
– History
– Models
– Lifecycle

Week 2 - Organisation
Organisational structure
• Stakeholders
• Customer
• ‘Need’
• SWOT Strategic
• O
Organisational
i ti l strategy
t t Management
– Vision/ Mission
– Goals and Objectives Process
– Projects
• Organisational Structure – Spectrum
– Advantages and disadvantages
• Choosing a location
• Interface Management FIGURE 2.1

Project Portfolio Management Week 3 - Scope Week 4 Cost estimating

What was this about??? • Project priorities • Why?
• Milestones • Top down / Bottom up
– Milestone Plan
• Screening Estimate
• WBS
• Feasibility Estimate
– Work Package
– Uses of WBS • Budget Estimate
Estimating • Tender Estimate
Responsibility matrix
Scheduling

7
 6/5/2011

Week 4 continued Week 5 - Quality Week 5 Scheduling

• Contingency fund • Quality Definition Logic:
– Activity on Arrow:
• Profit – Quality Plan • Dummy
• Direct Costs • Rules
– Quality Assurance
– Activity on Node
• Indirect Costs – Quality
Q alit Control
Cont ol • Duration
• Time Phased Budget • Quality Costs • Calculations:
• Risk – Forward Pass
– Identify risks – Backward Pass
– Critical Path
– Risk assessment
– Float – Total, Free
– Risk response
• Gantt Chart
• Change request

Week 6 - Scheduling Continued Week 7 - Resources Resources continued

• Precedence diagrams (lag–link networks) • Project Constraints • Resource Load
– Technical
• PERT – Physical • Resource Level
– 3 durations a,m,b – Time
– Calculate te, V – R
Resources
• Time constrained networks
– Calculate Z • Types of Resources
– Human • Resource constrained networks
– Look up Z tables to determine probability of – Plant and equipment
achieving a particular date – Materials
– Working Capital
– Technology
– Information

Week 8 - Crashing Crashing continued Crashing continued

Sometimes it is important to reduce the duration of a • Outsourcing project work • Cost slope
project.
• Scheduling overtime – $/day
Reasons for this include: • Dedicated core team
• Imposed Duration – perhaps by a political
announcement
• Do it twice – fast and correctlyy
• Fast tracking • Sensitive
S iti network
t k
• Market Factors – a 6 month delay in market release may
result in profit loss or loss of market share of about 30% • Critical chain • Insensitive network
• Incentive clauses may encourage early completion • Phased handover
• Liquidated damages clauses my motivate early
completion • Reduce scope
• Unforseen delays may require compression of remainder • Compromise quality
of the project • Adding resources

8
 6/5/2011

Week 9- Control Cycle Control Cycle

General Control Cycle
• What is this?
• What areas in a project do we control? Control is using the information collected
• Earned Value during the monitoring process to:
E al ate the projects
Evaluate p ojects performance
pe fo mance
Determine if corrective action is
required
Take appropriate action

The control cycle is a closed loop

Control overview Control overview Earned value

Establish baselines Establish baselines
PV – Planned Value
Monitor and report Monitor and report
Control - Time Control - Time EV – Earned Value
- Budget – Earned Value - Budget – Earned Value AC - Actual Value
- Risk - Risk
- Quality - Quality
- Organisation - Organisation
- Scope - Scope
Action Action
Problems in control Problems in control
Change control process Change control process

Week 10 - Legal issues Lean and Agile Critical Chain

Tort Law – Duty of Care Philosophy of these techniques?? Basis of the technique??
Contract types - risk
Contract Law Agreement

• Elements of a contract Consideration

• Contract Documents Intention

General conditions of Contract

• Extension of time
• Variations
• Extras
• Practical Completion
• Final Completion

9
 6/5/2011

Project Closure Audits Project Management Maturity

Model
• Project Delivery Why – To check to see that the project is
– Normal progressing satisfactorily
– Premature
– Perpetual
Pe pet al Wh – At any time but
When b usually
ll at key
k
– Failed milestones.
– Changed Priority - Should be planned (not surprise)
- Different issues will be considered
at different life cycle phases
FIGURE 14.1

CMMI v OPM3
Philosophy of these techniques??
What is the difference?

10