Improving forecasting to optimise programme and project decisions by "Chris Beach" at Project Controls Expo 2017, Arsenal Stadium, London

Copyright @ 2011. All rights reserved
Improving forecasting to optimise
programme and project decisions
Chris Beach
Project Controls Expo – 16th Nov 2017
Emirates Stadium, London

About the Speaker
Chris is Strategy Director of the
Programme and Project Solutions service
at Wood, supporting customers in
environment and infrastructure sectors in
the UK, Europe and Middle East.

About the Topic
The presentation will provide a perspective on:
 How well we can predict the future success of projects and
what can be learned from other industries
 Importance of considering uncertainty in forecasts
 Using trends to improve forecasting
 Use of scenario analysis to understand the range of
possible project results

Your Participation
Live.voxvote.com

Of course my project will be a success!
 On time
 On budget
 Quality achieved
 Happy stakeholders
 Zero harm
 Benefits delivered

Of course my project will be a success!

Decisions will be required
Strategic (Direction)
 Project Scope
 Pass a stage gate
 Contract strategy
 Which option to select
 Risk appetite
Tactical (Control)
 Specification detail
 Resource allocation
 Supplier payment
 Sequencing of work
 Allocation of contingency
Others?

Decision-making process
Assumes
 Full and accurate
information is available
 Measurable criteria exist
for which data can be
collected and analysed.
 Decision-maker has the
cognitive ability, time, and
resources to evaluate
each alternative against
the others

Role of Project Controls
Support decision-makers by:
 Creating structure
 Simplify complexity
 Converting data into
Information

Forecasting - Lessons from other
industries – Resource Planning
 18,000 flights cancelled
from November to
March
 Grounding 25 of its 400
aircraft
 Suspended 34 routes
over the winter
 Cost of €34.5m

industries – Weather Forecasting
 Weather data up to a
500-meter resolution
from 200,000 personal
weather stations across
the globe, providing 2.2
billion forecasts daily.
 Hurricane Harvey –
detailed modelling
incorporating historic
data and monte-carlo
simulation

industries – Election Forecasting
Ahead of election
probability of
Clinton win -
 New York
Times – 85%
 Huffington
Post – 98.3%

What do these examples tell us?
Requirements for a reliable
forecast:
 Robust historic information
 Incorporation of uncertainty
 Understanding the drivers of
trends
 Simulation modelling of
potential outcomes

Factors affecting ability to forecast
Proximity
PriorExperience/
Similaritytopast
Low
High
Close Far

0
100
200
300
400
500
600
P0 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12
Glidepath
Forecast out-turn cost
Using Trends to Validate EAC
Actual cost incurred

Measuring Forecast Accuracy
-60.00%
-40.00%
-20.00%
0.00%
20.00%
40.00%
60.00%
80.00%
P0 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12

Deterministic vs Stochastic Forecasts
Therefore
 Modelling of potential
range of outcomes can
aid decision-making
(Stochastic Forecasts)
 The outcomes of Projects are uncertain and have
inherent risk
 Single point (deterministic) forecasts will be incorrect

Monte Carlo Simulation
100%0%
Number of
results
from
simulation

Fan Curves
P95
P50
£
t
Baseline
P85

Earned Value
£
t
Budget
(Planned)
Earned
Actual
Forecast
PM Forecast

Basic Earned Value
£
t
Budget
(BAC)
Earned
(BCWP)
Actual
(ACWP)
Forecast
(EAC)
CV
SV
CPI = BCWP
ACWP
SPI = BCWP
BCWS
TCPI = BAC – BCWP
EAC - ACWP
(BCWS)

Bullseye charts

Programme / Portfolio Bulls-eye
£4.83m
(3)
£7.84m
(5)
£21.65m
(9)
£3.48m
(1)
£1.31m
(1)
Intervention
Additional
Attention

Evaluating Estimate to Complete
Productivity to complete
Productivity to complete

Earned Schedule
Ahead of
Plan
Behind
Plan
Trend
PM Forecast
P1
P2
P3
P4
P0 P1
P5
P6
P7
P8
P9
P10
P11
P12
P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14
EarnedSchedule(SVt)
Actual Time

Schedule Adherence - the P-Factor
 P-Factor demonstrates how well
the activities completed adhere
to the plan
 Calculate P-Factor
0.84
0.86
0.88
0.90
0.92
0.94
0.96
0.98
1.00
1.02
P0 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12
Pfactor
Min (PVES EVAT )
PVES
ES = Earned Schedule Date
AT = Actual Time Date
∑
∑

Success - Delivering the return on
investment

Forecasting Return on Investment
0.50
0.70
0.90
1.10
1.30
1.50
1.70
1.90
2.10
P0 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12
BCR
Time

Summary
 Project delivery is driven by management decisions
 Project Controls must support decision-making through
independent structured analysis
 Accurate forecasts are fundamental to effective decision-
making
 Good forecasts derived by using:
 Range of analytics to validate the forecast
 Incorporation of uncertainty to simulate future
outcomes
 Making data interpretable to facilitate decisions

Enjoy the rest of the day

Improving forecasting to optimise programme and project decisions by "Chris Beach" at Project Controls Expo 2017, Arsenal Stadium, London

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Improving forecasting to optimise programme and project decisions by "Chris Beach" at Project Controls Expo 2017, Arsenal Stadium, London

Similar to Improving forecasting to optimise programme and project decisions by "Chris Beach" at Project Controls Expo 2017, Arsenal Stadium, London (20)

More from Project Controls Expo

More from Project Controls Expo (20)

Recently uploaded

Recently uploaded (16)

Improving forecasting to optimise programme and project decisions by "Chris Beach" at Project Controls Expo 2017, Arsenal Stadium, London