Handling uncertainties in the FTTH planning process (workshop FTTH EU Conference 2016)

FTTH Council Europe, February 2016

Handling uncertainties in the planning process

Marlies Van der Wee

Jeroen Goossens

Peter Bonne

James Wheatley

3 Sessions

13:00 – 14:45 Eir’s experiences

Key Learnings from FTTH deployment in Ireland

15:00 – 16:30 New Zealand’s UFB deployment

Answers to important FTTH deployment barriers

16:45 – 17:30 Handling Business case uncertainties

Exploring ways to improve your business case

2

Speakers

3

Marlies Van Der Wee

iMinds

James Wheatley

GE

Lomme Devriendt

Orbit

Jeroen Goossens

FiberPlanIT

Handling uncertainties in the business case

Marlies Van der Wee

4

Uncertainties in the entire process…

5

Model

Evaluate

Refine

Scope

Subdivideproblem

Collectinput

Processinput

Infrastructure

Processes

Investmentanalysis

Sensitivityanalysis

Value network analysis

Game theory

Realoptions

Revenues

A lot of uncertain parameters

6

Geographic / demographic / economic Area type

Population density

Level of education

Income

Legal Right of Way and licenses

Regulation

Infrastructure Existing networks / equipment

Reuse of locations (poles, buildings)

Market Competition

Adoption

Scope

How to handle uncertainty?

7

Model

Evaluate

Refine

Scope

Subdivideproblem

Collectinput

Processinput

Infrastructure

Processes

Investmentanalysis

Sensitivityanalysis

Value network analysis

Game theory

Realoptions

Revenues

Sensitivity analysissimulates impact of input variation

8

2.5

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

90% 92% 94% 96% 98% 100% 102% 104% 106% 108% 110%

NPV(millionEUR)

Parametervariation

Adoption

Trenching

Fiber

0.90 0.94 0.98 1.02 1.06 1.10

Pro

ba

bilit

y

0.69 0.82 0.94 1.07 1.19 1.31

Pro

ba

bilit

y

0.90 0.94 0.98 1.02 1.06 1.10

Pro

ba

bilit

y

Real optionsallow to value flexibility to react to uncertainty

Weak aspect of NPV evaluation Assumes strict planning, with no flexibility

Real projects Anticipate on changing market circumstances

Solution: “real options thinking” principle

9S = value share on exercise date

va

lue

ca

ll o

n e

xe

rcis

e d

ate

X

Scale up/down

Study/Start

Shut down

Upfront choice between big and small cabinetbased on flexibility in later extension (given uncertain uptake)

10

Options reduce negative outcomeUpfront most flexible (small cab) is chosen

10% risk: NPV < €6,000,000 8.3% risk: NPV < €6,000,000

Install small

cabinets

11

Game theory

12

Game theory is a discipline aimed at

modeling situations in which decision

makers have to make specific actions

that have mutual, possibly conflicting,

consequences.

Prisoners’ dilemmaNon-Pareto Optimal Nash Equilibrium

Ref: J. Nash, 1950, Equilibrium points in n-person games, Proc. of the National Academy of Sciences 36(1), 48-49.

Pareto

optimum

Nash

Equilibrium

1 year 1 year 3 years 0 years

0 years 3 years 2 years 2 years

(betray)

DEFECT(stay silent)

COOPERATE

(be

tray)

DE

FE

CT

(sta

y s

ile

nt)

CO

OP

ER

AT

E

13

Competition has big impactas shown by game theory

14

expected pay-off

vs.

actual pay-off in case of

competition

Competition has big impactas shown by game theory

15

Pareto optimum

Nash

Equilibrium

Competition has big impactas shown by game theory

16

React to competition and uncertaintyusing options, games and sensitivity

17

Refine

Sensitivityanalysis

Game theory

Realoptions

0.90 0.94 0.98 1.02 1.06 1.10

Pro

ba

bili

ty

0.69 0.82 0.94 1.07 1.19 1.31

Pro

ba

bili

ty

0.90 0.94 0.98 1.02 1.06 1.10

Pro

ba

bili

ty

Uncertainty on the value

of input parameters

SENSITIVITY ANALYSIS REAL OPTIONS GAME THEORY

Effect of timing on

decision making

Impact of competition on

the business case

1,0 1,2 0,1

0,3 0,1 2,0

A1 A2 A3

B1

B2

How data quality affects planning and design

Jeroen Goossens

18

Quality of input data for network design

Geo-referenced data (GIS) is essential

(streets, demand points... of area)

Accurate unit costs (labour, material)

Geo-marketing data

ROW (rights of way)

…

19

Input GIS data examples

How many apartments

(homes) in the building?

Re-use street poles and

existing conduits?

Which type of trenching for

specific roads?

…

20

Garbage in, Garbage out

Bad input data

Inaccurate network design

More costly deployment

21

Strategic planning

1. Extrapolation

Design small area manually

Taken from previous projects

2. Excel model

3. Full designs (automated calculations)

Different levels of input data required

Different levels of output accuracy expected

22

Different quality at different stages of project

Strategic planning stage

Small errors allowed

Ballpark cost estimations

Network design

Generate to build plan

High accuracy input/output required

23

Impact of bad data quality

Bad volume and cost estimations

Equipment (BOQ), labour

Design errors

E.g. Insufficient fibre capacity in cables

Inefficient (costly) design

Cost savings possible through smart planning and design

Longer planning and deployment time

Verification, corrections of to build plan

Total budget overrun

24

Using Mobile Mapping to collect data

Lomme Devriendt

25

Introduction to Mobile Mapping

Mobile Mapping brings the full representation of 3D reality onto the

desktop, using sensors mounted on a mobile vehicle (car, train,

boat, bike, even a person).

26

Introduction to Mobile Mapping

Result : Full 3D View in 360° imagery and laser-pointcloud

With exaction positioning and measurement capabilities

27

What can we do with it ?

Reduces field trips

Extraction of road infrastructure

Inventory of assets

Visuals checking and judgement

Placement checks

Evaluate trenching options (ground works)

3D Analysis (Line of Sight)

And much more

28

From within the office !

Use Case 1 : Base map

Base maps are traditionally generated by

Photogrammetry (aerial)

Surveying (terrestrial)

3D Mapping allows ad hoc addition, correction, completion

29

Use Case 2 : Managing Assets

Inventory of Assets

Display

Check

Add

Build

Planning checks

Position

Surroundings

Ground

Connectivity

30

Some more examples (short movies)

Roadside Pole

Catenaries / Height above ground

Line of Sight

31

Verification of the To-Build plan

Check locations

Wall-mount

Ground situation

Asset positioning

Update planning

32

Cost Effectiveness

Typical case: Single Collection = Multiple Use

Basic Mapping uses

Use for Road infrastructure update

Use for Asset Inventory creation / update / verification

FTTx planning uses

Check planning

Prep operations

Further use

Continuous availability of 3D view in day-to-day operations

33

Companies using Mobile Mapping today

34

What happens post construction?

James Wheatley

35

Data, data, data everywhere…..

Low level design requires good quality input

data to realise efficient design

Subsequent operations & maintenance

processes also require good quality data

about the network to be held in the inventory

Processes dependent upon physical

inventory data include

Service fulfilment

Service assurance

36

Why the focus on data quality?

Network never built as designed despite

best efforts

Important to quickly and efficiently capture as-

built network changes

Reduce service fulfilment issues by digitising

as-built update process

Faster as-built returns mean up to date data in

inventory

Electronic returns improve data quality by

reducing errors

37

Field-based updates

Enable field teams to capture as-built changes

there and then in the field

Build into process flow defined in work

management solution

Service fulfilment

Less fall outs – data available when request is made

Service assurance

Accurate location of faults on new or changed network

Network design

Subsequent designs based on current state of the

network

38

Session Wrap-up

Jeroen Vanhaverbeke

39

Session Summary

Looked at how to manage uncertainties in the planning process

Explored concepts such as game theory and sensitivity analysis

Highlighted how data quality will impact the design results

Investigated how Mobile Mapping can be used to provide higher quality

data

Finally described how can improve data quality on the as-built network

held in the GIS-based network inventory through field-based as-built

updates

40

FTTH Council Europe, February 2016

Thank you for your attention!

Any questions?

www.technoeconomics.ugent.be