9-1

ELC 347 project management

Day 22

Agenda• Integrative Project

– Part 5 Feedback Posted– Part 6 (page 378) Due Dec 9– Completed project plan and presentation due December 13 @ 1 PM– Any of the first five sections can be resubmitted for rescoring prior to finals week.

The recorded score will the average of the original score and the score on the resubmitted section. Please notify me via email which sections you will be resubmitting.

• Assignment 7 corrected – 13 A’s, 1 B and one that I cannot read (Project 2010 files)

• Assignment 8 (Last One) Posted – Due Dec 6 @ 12:35 PM

• Exam 2 Graded– 2 A’s, 4 B’s, 9 C’s– Threw out two faulty questions

• Course Evaluations • Discussion on Critical Chain Project Scheduling

6-2

Schedule for rest of semester• Today

– Chap 11 • Dec 2

– Chap 12– More on MS Project

• Dec 6– Group Work – Assignment 8 due

• Dec 9– Chap 13 & 14– IP Part 6 due

• Dec 13 @ 1 PM– Quiz 3

• Chaps 11 - 15

– IP Project’s due

6-3 © 2007 Prentice-Hall, Inc

11-4

Critical Chain Project Scheduling

Chapter 11

© 2007 Pearson Education

11-5

11-6

Theory of Constraints & Critical Chain Project Scheduling

A constraint limits system output.The Goal – Goldratt

TOC Methodology1. Identify the constraint2. Exploit the constraint3. Subordinate the system4. Elevate the constraint5. Repeat the process

11-7

VariationCommon CauseInherent in the system

Special CauseDue to a special circumstance

Managers should• Understand the difference between the two• Not adjust the process if variation is common cause• Not include special cause variation in risk

simulation• Not aggregate discrete project risks

Dropping a marble

11-8

11-9

Dr J Edwards Deming

11-10

CCPM and the Causes of Project Delay

How safety is added to project activities1. Individual activities overestimated2. Project manager safety margin3. Anticipating expected cuts from management

time

25%

50%

80%90%

Gaussian Distribution

11-11

Wasting Extra Safety Margin

1. The Student Syndromea. Immediate deadlinesb. Padded estimatesc. High demand

2. Failure to pass along positive variationa. Other tasksb. Overestimation penaltyc. Perfectionism

3. Multitasking4. Path Merging

11-12

11-13

Critical Chain Solutions

Central Limit Theorem

V∑ = n * V , SD2 = n * V, SD = n1/2 * V1/2

Standard deviation of the sum is less than the sum of the standard deviations!

Aggregating risk leads to reduced risk

n

9-14

Activity durations estimated at 50% level

Buffer reapplied at project level– Goldratt rule of thumb (50%)– Newbold formula (2σ)

Feeder buffers for non-critical paths

11-15

11-16

11-17

CCPM

Original 50% ProbabilityActivity Duration Duration

A 10 5

B 6 2

C 14 7

D 2 1

E 8 3

F 12 6

Total 52 24

11-18

New estimate = 24 + (52-24)/2 = 38

11-19

CCPM Changes

Due dates & milestones eliminated Realistic estimates – 50% level not 90%

“No blame” culture

Subcontractor deliveries & work scheduled ES Non critical activities scheduled LS Factor the effects of resource contention Critical chain usually not the critical path Solve resource conflicts with minimal disruption

11-20

11-21

11-22

Critical Chain Solutions

Bob

Feeder Buffer

Feeder Buffer

Feeder Buffer

ProjectBufferBob

Bob

Buffers protect constraints and

prevent delays

11-23Joe.mpp

11-24

Critical Chain Project Portfolios

Drum – system-wide constraint that sets the beat for the firm’s throughput

• Drum – person, department, policy, resource• Capacity constraint buffer – safety margin

between projects

• Drum buffer – extra safety before the constraint

11-25

Applying CCPM to Project Portfolios

1. Identify the drum2. Exploit the drum

a. Prepare a schedule for each projectb. Determine priority for the drumc. Create the drum schedule

3. Subordinate the project schedules (next slide)4. Elevate the capacity of the drum5. Go back to step 2

11-26

Subordinating Project Schedules

• Schedule projects based on drum

• Designate critical chain

• Insert capacity constraint buffers

• Resolve any conflicts

• Insert drum buffers so the constraint is not starved

11-27

11-28

CCPM Critiques

No milestones used

Not significantly different from PERT

Unproven at the portfolio level

Anecdotal support only

Incomplete solution

Overestimation of activity duration padding Cultural changes unattainable