estimating activity time & cost

© 2012 Dr. Kenneth F. Smith, PMP

ESTIMATING ACTIVITY

TIME & COST

1 Critical Path Method


AFTER IDENTIFYING ACTIVITIES

& SEQUENCING THEM

ACCURATELY ESTIMATING ACTIVITY DURATIONS

& TIMING

IS THE NEXT IMPORTANT ASPECT OF

PROJECT SCHEDULING



“I hate to criticize, but you’ve only been employed here

for two days and you’re already three weeks behind in

your work.”


© 2014 Dr. Kenneth F. Smith, PMP 4 Project Management

Fundamentals

RISK ANALYSIS

Phase II

1. Activity Time Estimating

2. Decision Tree Analysis of

5 major risks

with focus on Probability & Impact on

Overall Project

Schedule & Budget


ESTIMATING ACTIVITY DURATIONS

• TIME is fundamental to all Projects

• Most Project Managers focus on managing

TIME

• Most Project Management Software focuses on

TIME

• Control of time is often used to control costs


© 2012 Dr. Kenneth F. Smith, PMP 6 Critical Path Method

Key Concepts

“Touch-time” = The amount of direct on-the-job

working time needed to perform an Activity

“Waiting (Idling) time” = Time when an Activity is on-

going (i.e. started, but not yet finished) but with nothing

happening

“Elapsed time” = Touch time plus Waiting time

“Opportunity time” = Waiting time when – if given the

resources – productive work could be done on the

activity, or another task, or activity on another project.)

© 2012 Dr. Kenneth F. Smith, PMP 7 Project Management Fundamentals

In many project working environments –

particularly matrix organizations where

equipment and personnel resources are

shared, outsourced and/or team members

are often concurrently assigned to

support multiple projects that are

otherwise beyond the control of

individual project managers –

Touch time for an activity may be as little as

10% of Elapsed time!


Thus, in most Program and Project situations massive amounts of

Opportunity time exist

Since the Project Schedule (time plan)

defines the Project Budget (money over time)

Effective time management is critical


Estimating Project Duration & Costs is not

Easy

Estimating as Accurately as possible is

important, but

• Time/Cost/Quality affect each other

• Estimating Methods affect accuracy

• How much detail do you need?

– Top Down

– Bottom up

• In the Real World, Estimates and Actuals are

usually different



Types of Estimate

• Concept oriented (Rough order of magnitude)

– High level definition

– Requirements not fully defined

– Accuracy of -25% to +75%

– Informal techniques used

• Function oriented (budgetary estimate)

– Requirements are well defined


• Implementation oriented (definitive estimate)




Estimation Methods and

Techniques

• Analogous (i.e. based on similar

projects, or processes; usually Top

Down)

• Parametric Modeling (i.e. “Rule

of Thumb” formulas or ratios)

• Detailed (Bottom up)



Analogous by components; based on

familiarity with similar projects

Computer

Project

100 %

30 % 10 % 60 %

5 % 10 % 10 % 5 %

System Testing Computer

Software

Computer

Hardware

Hardware

Design

Built

Hardware

Testing

Facility

Hardware

Testing

Software

Design

Software

Codes

10 %

Tests Design

Usually by allocating available

time durations, or percentages

to each level


50 weeks

15 weeks 30 weeks 5 weeks

5 weeks 10 weeks 15 weeks

2.5 weeks 2.5 weeks 5 weeks 5 weeks

20 % 30 %


Parametric Modeling

Parametric modeling uses key project characteristics (parameters) in a mathematical model to predict project costs.

Examples:

In software development, the number of lines of code to predict cost.

In construction, use the per- square meter of living space to estimate cost.


Furthermore, “contractors estimate the cost of corruption in the construction process at P125 per square meter.

It’s worse in mining, where one Tsinoy has been overheard grumbling that he is being asked to fork out P200 million for a permit.”

[Reported by Ana Marie Pamintuan in her Opinion column: Sketches “Business

Costs” The Philippine Star, 1 July 2013.]

© 2012 Dr. Kenneth F. Smith, PMP 17

Critical Path Method

“Pork Barrel”

Priority Development Assistance Fund (PDAF)

Under the PDAF System in the Philippines:

Each Congressional Representative gets 70 million pesos

(approximately US$1.75 million ) 30 million pesos per annum for “Soft” Projects – i.e.

Education, Health & Social Services – and another 40 million pesos for “Hard” – i.e. infrastructure Projects.

Each Senator gets 200 million pesos (approximately US$5 million )

100 million pesos per annum for “Soft” Projects, and another

100 million pesos for “Hard” Projects.

“The scuttlebutt is that the commission received by erring legislators ranges

from 20% to 30% for Hard projects . . . and 60% to 70% for Soft projects – as

evidenced by the ghost foundations that are the target of the current expose.”

Dean Andy Bautista My Four Centavos

The Philippine Star, Saturday August 10, 2013, Opinion page 12


Detailed (Bottom Up) Items &/or

Components

Overseas trip

preparation

$1,000

Accommodation

& Transportation

Personal

Belongings

Travel

Documents

Passport Visas Clothing Other Items

Plane ticket Hotel Car Rental

&/or Taxis

$1,000 $750

$2,750

$1,000 $750

$1,750 $500

$150 $350

$5,000

Add Up Estimated

item Costs at

each level



How to Prepare a Project Schedule

– Identify Activities & Milestones (WBS)

– Prepare Network diagram (CPM)

– Estimate the level-of-effort (staff-hours) to

complete the activities

– Translate effort into duration (working time)

– Convert working time to calendar time

– Identify “Critical Path” & Slack/Float

– Assign staff to work on activities



Estimating Activity Times,

& Ultimately Project Duration

IMPLEMENTATION PLANNING

& SCHEDULING



Time estimates are typically Over-optimistic for one or

more principal reasons:

CLIENT DRIVEN The Client establishes the project

completion deadline before technical analysis,

consultation or project management feedback

CONTRACTOR RESPONSE In order to be “Fully

Responsive” Contractors (Project Managers) accept

the Client’s deadline to hopefully win the contract

Project Activity Duration Estimating

& Scheduling



After technical analysis Contractors either

• Arbitrarily “Cut and Paste” activity time estimates

to fit the Client’s pre-determined schedule.

• Use a standard ‘Beta Distribution’ “PERT 3 time” or

a “Triangular Distribution” 3 time estimating

method to compute an “Earliest Expected Time”

but in either case, tend to favor the “Optimistic

Time” over the computed Earliest Expected Time, or

the Most Likely Time estimate in “Cutting and

Pasting” their final schedule proposal.

However, IMPLEMENTATION experience is that even

the computed Earliest Expected Time, and the “Most

Likely Times” are over-optimistic & unrealistic!




Who Provides the Best Estimates?

• Those Accountable

for the work (i.e. First

Line Supervisors)

• Those who know the

work (Senior

Technical Specialists)

• Those who Actually

do the work

(Journeymen)



Who IMPOSES the WORST TIME

Estimates on Project Managers and

First Line Supervisors?

• HIGH LEVEL EXECUTIVES!!

– MARKETING DEPARTMENT PERSONNEL

• COMPANY CONTRACT NEGOTIATING

DEPARTMENT

CUSTOMERS!

Moreover, usually, none of the above have

any detailed knowledge or experience as

to what is involved or how long it actually

takes to do the work!



The Dilbert cartoon illustrates the problem faced by

project estimators, as well as some of the terms

commonly used in planning:

• Optimistic

• Most Likely

• Pessimistic,

• Earliest Expected, &

• Realistic

Times



RISK QUANTIFICATION

Determining

Risk Events &

Risk Tolerance



Triangular Distribution Technique

A 3-time probability-based

time estimating technique

sometimes used to estimate

activity durations when there

is uncertainty about their

time durations

te

= O + M + P

3

i.e. a Simple Average

or “Mean”

of the range of

possibilities



Triangular Distribution

Activity Duration Estimate

Optimistic + Most Likely + Pessimistic

3

Example:

Optimistic time is 10 weeks

Pessimistic time is 35 weeks

Most likely time is 15 weeks

10+ 15 + 35 = 60

3 3

te = 20 weeks Earliest Expected Time

With a 50% probability of being accurate.



Program Evaluation & Review Technique

(PERT)

te

= O + 4M + P

6

i.e. a Weighted Average

or “Weighted Mean”

of the range of

possibilities


A 3-time probability-based

time estimating technique

more often used to estimate

activity durations when

there is uncertainty about

their time durations

“Beta Distribution”


PERT Method – Activity Duration

Estimate

Optimistic + (4 x Most Likely) + Pessimistic

6

Example:




10+(4x15)+35 = 10+60+35 = 105

6 6 6

te = 17.5 weeks, or 18 weeks (rounded up)

Earliest Expected Time



The “Triangular Distribution” Formula:

A Fundamental Flaw.

The Triangular Distribution formula to estimate Activity duration – i.e. the “Earliest Expected”

time -- is a simple average, so the same weight is given to the two extremes (i.e. best and worst cases) as the “most likely” estimate.

However, since the Optimistic and Pessimistic estimates are extremes -- by definition -- they are less likely to occur. Hence the formula and its resultant Earliest Expected Time at 50% probability computation is unrealistic!



The “PERT” Formula (Beta Distribution)

A Fundamental Flaw, & Caution

The PERT formula to estimate Activity duration – i.e. the “Earliest Expected” time -- is a weighted average that attempts to rectify the

undue bias in the Triangular Distribution.

Nevertheless, the probability of completing the Activity by the “Earliest Expected” time is still only 50%

In other words, at the outset, the Activity duration is under-estimated half of the time, so there is an equal likelihood the Earliest Expected time computed will not be met, but will be overrun.



Russian Roulette “RR”

Therefore, Project Managers

who use the Standard PERT

3-Time formula to estimate

project Activity timing run a

high risk of Failure . . .

Even Worse than Playing

RUSSIAN ROULETTE !!!




What is the Probability of Surviving “RR” with One Round in a Six Cylinder Chamber Revolver?

• Point the Gun at Your Head

• Spin the Chamber

• Pull the Trigger

• Probability of being killed is

1 / 6 = 17 %

Thus Probability of Surviving = 100% - Probability of Dying = 83%




And the Probability

of Surviving with:

Three Rounds in

the Chamber

= 3 / 6 = 50 %

This is the same probability as

the PERT “3 time” weighted-

average estimating formula!!!



Would You play Russian

Roulette with One bullet?

NO? Then why expect Project Managers

to Estimate & Schedule

Project Activities with the “PERT 3 Time Formula”

Which is the equivalent of

Russian Roulette with Three bullets?



PERT Formula — A Work-Around Solution

Ken Smith’s “Realistic Time” Strategy

1. Use the PERT beta distribution 3 time formula during the planning stage as the first step in estimating individual activity durations

2. Then add two standard deviations to the activity duration

This will adjust the likelihood of completing the activity by its scheduled time to 95%



NORMAL CURVE, RANGE, STANDARD DEVIATION,

and RELATED PROBABILITIES

X -1 SD +1 SD

-2SD +2SD

-3SD +3SD

68.26%

95.44%

99.74%

50%

-3 -2 -1 +1 +2 +3

0.13% 2.15% 13.54% 34.13% 34.13% 13.5% 2.15% 0.13%

MEAN

0 % 100% 95%



The MEAN & the Standard Deviation

• A Mean (Average) is a measure of “Central

Tendency” – i.e. the midpoint of a range of

values

• A Standard Deviation is a measure of Dispersal

from the Mean within that range – essentially the

Opposite of a Mean

• Each Standard Deviation from the mean

represents a fixed percentage

• The Probability of any value occurring in that

range is the size of its standard deviation from

the mean



Dr. Ken’s prescription for

improving activity duration estimating

ter = opt + 4 ML + pess + 2 Std. Devs

6

Realistic Activity Time

Take 1 PERT

+ 2 ESD’s


NOTE: Very Practical, but

Not in PMBOK or PMP Exam


Basic PERT/CPM Formula

Earliest Expected Time = opt + 4 ML + pess

6

Activity Estimated =

Standard Deviation

pess - opt

6


Two Estimated =

Standard Deviations

pess - opt

3


Ken’s “Realistic Time” Method:

Optimistic + (4 x Most Likely) + Pessimistic + 2 SDs

6

Example:




10+(4x15)+35 = 10+60+35 = 105 = 17.5 weeks

6 6 6

1 SD = (35 – 10)/6 = 25/6 = 4.16 & 2 SDs = 8.33

So Realistic Time = 25.83, or 26 weeks (rounded up)

For a 95% probability of Success in attainment.



ESTIMATING ACTIVITY DURATIONS FOR SCHEDULING UNDER CONDITIONS OF UNCERTAINTY

© 2014 Dr. Kenneth F. Smith, PMP [email protected]

NOTE: The Probability of SURVIVING Russian

Roulette is 83% !

Enter Your

DESIRED

Probability

% BELOW

PERT FORMULA

Ursula Kuehn's* Worksheet to Estimate

"P" in PERT Pessimistic Time Formula

PERT

FORMULA

Dr. Ken's

REALISTIC

FORMULA

( 0 + 4 ML + P )

/ 6 "What Could Go Wrong?"

Ac

tivity # Enter O, ML & P

Time Data in

cells Below

*Integrated Cost & Schedule

Control in Project

Management. 2nd Edition.

Management Concepts

Identify Top Three Risks

P =

MURPHY's

Law:

What if

ALL

THREE

Risks

Occur!

50%

PROBA-

BILITY

95.44%

PROBA-

BILITY

99.90

OPTIMIS

TIC

TIME

MOST

LIKELY

TIME Risk 1 Risk 2 Risk 3

EXPECTED

TIME

DURATION

REALISTIC

TIME Your Time

1 4 7 9 11 13 17 8.17 12.50 15.36

2 1 10 25 11.00 19.00 24.28

3 2 4 6 4.00 5.33 6.00

4 1 2 7 2.67 4.67 5.99

5 1 2 3 2.00 2.67 3.00

mailto:[email protected]


REALISTIC TIME COMPUTATION © 2012 Dr. Kenneth F. Smith, PMP

[email protected]

PERT FORMULA for Estimating

Activity Duration

Dr. Ken's

REALISTIC

FORMULA

Ursula Kuehn's* Worksheet for

Estimating Pessimistic Time

( 0 + 4 ML + P ) / 6

"What Could Go Wrong?"

Ac

tivi

ty

#

Enter O, ML & P Time

Data in cells Below

50%

PROBABIL

ITY

95%

PROBABILI

TY

*Integrated Cost & Schedule Control

in Project Management. 2nd Edition.

Management Concepts

OPTIMI

STIC

TIME

MOST

LIKELY

TIME

PESSIMI

STIC

TIME

EXPECTED

TIME

REALISTIC

TIME

1 2 3

MURPHY

:

Everythi

ng

1 3 7 20 8.50 14.17 15 17 14 20

2 1 10 25 11.00 19.00

3 2 4 6 4.00 5.33

4 1 2 7 2.67 4.67

mailto:[email protected]


REMEDY

CONTRACTOR RESPONSE: Contractors (Project Managers)

• Conduct the Technical Analysis using the supplement to the PERT 3-time estimating approach that improves the probability of developing a more realistic time estimate

• Provide feedback to the Client ASAP -- Brief the Client to demonstrate why their deadline is unrealistic, and appeal for duration extension, or reduction in Scope of Work before submitting a bid

• If Client insists on the Original Project Completion Deadline &/or Scope, either

• Get a change order ASAP if you are the successful bidder

• Don’t Bid and consider yourself lucky you won’t have to deal with the time, cost & quality problems that will inevitably arise!


& Scheduling



The “Critical Chain” Method (CCM)

CCM is another approach to deal with uncertainty by adding non-work time duration “Buffer” activities at various points in the Critical Path Network; and particularly just before a series of activities joins the Critical Path.

Buffer Activities on the Critical Path are

called “Project Buffers”

Buffer Activities that are not on the

Critical Path are called “Feeder Buffers”



REMEDY

CONTRACTOR RESPONSE: Contractors (Project Managers)

• Conduct the Technical Analysis using the supplement to the PERT 3-time estimating approach that improves the probability of developing a more realistic time estimate

• Provide feedback to the Client ASAP -- Brief the Client to demonstrate why their deadline is unrealistic, and appeal for duration extension, or reduction in Scope of Work before submitting a bid

• If Client insists on the Original Project Completion Deadline &/or Scope, either

• Get a change order ASAP if you are the successful bidder

• Don’t Bid and consider yourself lucky you won’t have to deal with the time, cost & quality problems that will inevitably arise!


& Scheduling



A NOTE ON PROBABILITY

• A Risk Event could still occur Despite a Low Probability

and

• A Risk Event might not occur Despite a

High Probability

There are No Guarantees !!!



Team Exercise - Estimating

• Upgrade your CRITICAL PATH NETWORK based on the WBS plus any new information & feedback

• Estimate [“Best Guess!”] Time Durations for each Activity and Task in the WBS independently,

• USE a STANDARD DURATION TIME UNIT THROUGHOUT

• Do not attempt to Schedule the overall Project WBS]




• Add Time Estimates to your Activities

• Schedule your Critical Path Network

• Prepare it for Presentation




• Upgrade your WBS based on new information & feedback

• Estimate [“Best Guess!”] Resource Requirements and Costs for each Component, Activity and Task in your WBS

• Estimate [“Best Guess!”] Time Durations for each Activity and Task in the WBS independently, using the PERT formula. [NOTE: Do not attempt to Schedule the overall Project WBS]

• Prepare a Resource Register with the Estimated

Costs and Estimated Duration for each item in the

WBS



Calendar Time

Elapsed time

– Accounts for all time, not

just time spent on the

project

Working time

– Time spent working on

the project; Synonymous

with productive time



Usefulness of CPM & PERT

In planning:

it forces the project staff to carefully identify the tasks to be undertaken

Precisely determine the relationships of the tasks to each other

Allows planners to do “what-if” scenarios to determine the impact of task slippages

Helps to create more realistic estimates of project schedules



Key Time Estimating Issues that can be Explored

with the

CRITICAL PATH CONCEPT

• Calculating Realistic durations and timings -- Earliest & Latest -- for Activities & Milestones

• Determining reasonable deadlines

• Identifying “Hurry up & wait” and “Wait & hurry up” Activities

• Parallel Scheduling of Selected Activities to speed up project implementation

• Fast Track Facilitation

• Identifying Merge/Burst Points

• Identifying “bottle neck” Activities

• Performing “What if” simulations for forward planning



Critical Path Network with three buffers added

applying the Critical Chain method

A

B

D

G

C

E

F

H

I

J Z

2

4

2 3

3

5

4 1 1

5

6

4 3

3

4

19

16

12

11 8

11

12

5 2

5

4

8

2

0 7

2

2

7

8

8

16

6

5

4 20

6


9

3

Feeder Buffer

Project Buffer

3 15 1

© 2013 Dr. Kenneth F. Smith, PMP Project Management Fundamentals Slide 63

Applying the Decision Tree / Risk

Breakdown Structure Technique

in RISK ANALYSIS Phase II

Estimating PROJECT

Schedule Buffer &

Budget Contingencies, &/or

Management Reserves

# 1-8 Decision Tree Analysis

Template In Tool Kit

Project Management Fundamentals © 2013 Dr. Kenneth F. Smith, PMP

Decision Tree / Risk Breakdown Structure to Analyze the Probability of up to Five Sequential Risk Tiers

Etc., etc.,

resulting in

32 separate

different

possible

outcomes

for Five

Sequential

Risks

1

Risk

2

Risks

3

Risks Each of

which

affect the

schedule

and budget

in different

ways!

2 possible

outcomes 4 possible

outcomes 8 possible

outcomes

Slide 64


Exponential Contingency / Management Reserve Needed to address the Cascading Effect of Additional Risks

Likely to be Encountered

# Risks Encountered

Sc

ale

o

f C

on

tin

ge

ncy L

eve

l

to

a

dd

re

ss th

e in

cre

asin

gly

co

mp

ou

nd

e

ffe

ct

0 5 3 4 2 1

2

4

8

16

32

The Cumulative

Impact of

“Murphy’s Law”

“If anything can go

wrong, it will, and at the

worst possible time”

A Computer Program is needed to

handle all the possibilities and

outcomes

Slide 65


Revisit your Project Network and

update activity Durations based on

risk analysis data

Then add buffers

Team Activity



Develop a Time-Scaled Gantt/Bar

Chart from the CPM Network Data

67

Ac

tiv

itie

s &

Mile

sto

nes

Time

Executive Level Review of Implementation Plan

Milestone

Summary


TEAM EXERCISE

68

Prepare a GANTT/BAR CHART

for your Project’s

MAJOR ACTIVITIES & MILESTONES