Master's thesis of Engineering: Modelling of optimal sourcing strategy for future LNG demand

Modelling of Optimal Sourcing Strategy for Future LNG Demand

A thesis submitted in fulfilment of the requirements for the degree of Master of

Israt Mustary

B Sc in Computer Science & Engineering

Darul Ihsan University, Bangladesh

Engineering

School of Engineering

College of Science, Technology, Engineering and Maths

RMIT University

October 2020

Declaration of Originality

I certify that except where due acknowledgement has been made, the work is that of the

author alone; the work has not been submitted previously, in whole or in part, to qualify

for any other academic award; the content of the thesis is the result of work which has

been carried out since the official commencement date of the approved research

program; any editorial work, paid or unpaid, carried out by a third party is

acknowledged; and, ethics procedures and guidelines have been followed.

I acknowledge the support I have received for my research through the provision of an

Australian Government Research Training Program Scholarship.

Israt Mustary

Date: 06 October 2020

List of Publications

The following is a list of publications arisen from this work:

Mustary, I., Chowdhury, H., Loganathan, B., & Alam, F. (2017). Development

of a Computational Model for Optimal Sourcing of LNG. Energy Procedia, 110,

597-603.

Alam, F., Saleque, K., Alam, Q., Mustary, I., Chowdhury, H., & Jazar, R.

(2019). Dependence on energy in South Asia and the need for a regional

solution. Energy Procedia, 160, 26-33.

iii

Acknowledgments

At first, I want to express my whole-hearted thanks and gratitude to my senior

supervisor, Professor Firoz Alam for his guidance, encouragement, advice, constructive

criticism and assistance. Special thanks to Dr Harun Chowdhury and Dr Bavin

Loganathan for their assistance with this study.

Declaration of Originality ............................................................................................. ii

List of Publications....................................................................................................... iii

Acknowledgments ........................................................................................................ iv

Table of Contents .......................................................................................................... v

List of Tables ................................................................................................................ ix

List of Figures .............................................................................................................. xi

Nomenclature and Units ............................................................................................. xiii

Abbreviations and Acronyms ..................................................................................... xiv

Abstract ......................................................................................................................... 1

Global oil scenario ...................................................................................... 6

Global coal scenario .................................................................................... 7

1.1 Global energy scenario ................................................................................ 3

Future sources of primary energy ................................................................ 9

Global LNG market and future demand .................................................... 11

1.2 Advantages of LNG .................................................................................. 10

LNG transportation system ....................................................................... 14

1.3 Factors for optimal sourcing strategy for LNG importers ......................... 17

Production capacity of the source terminal ............................................... 18

Distance between source and destination terminals .................................. 18

Price at the source ..................................................................................... 18

Risk on the transportation route ................................................................ 18

Trade dispute between source country with other nations ........................ 19

Source country’s involvement in internal/external conflicts or warfare ... 20

Bilateral relationship between source and destination country ................. 20

Trade relationship/volume of export-import between source and

destination country .................................................................................... 20

Any economic or UN sanctions on the source country ............................. 20

Medium/long-term political risk index of the source country ................... 20

Political stability index of the source country ........................................... 21

Economic strength of the source country .................................................. 20

1.4 Objective and scope .................................................................................. 21

1.5 The rationale of the research ..................................................................... 21

2.1 Current body knowledge on optimal sourcing strategy of LNG ............... 23

Ship routing and scheduling ...................................................................... 24

Sailing cruise speed ................................................................................... 25

Type and size of the ship ........................................................................... 26

Type and frequency ................................................................................... 26

Sea condition and route ............................................................................. 26

Sourcing strategy ....................................................................................... 28

2.2 Summary of current studies on LNG sourcing and research gap .............. 29

2.3 Research questions .................................................................................... 30

3.1 Methodology ............................................................................................. 31

3.2 Price estimation model .............................................................................. 33

Development of a computational model using computer software ........... 37

Important parameters associated with LNG transportation cost ............... 33

Methodology ............................................................................................. 41

3.3 Development of the optimal sourcing model ............................................ 41

4.1 Sources of research data ............................................................................ 44

4.2 Data of LNG exporting countries .............................................................. 44

4.3 LNG carriers and charter rate .................................................................... 53

4.4 Major LNG shipping routes and tolls ........................................................ 53

4.5 LNG market and wholesale LNG price ..................................................... 54

5.1 Introduction ............................................................................................... 57

Current energy scenario of Bangladesh .................................................... 57

Distance measurement ............................................................................... 61

5.2 Implementing of the cost estimation model for Bangladesh ..................... 61

Model outputs and data processing ........................................................... 63

Cost estimation .......................................................................................... 62

5.3 Implementing the optimal sourcing model for Bangladesh ...................... 64

Shortlisting by data filtering ...................................................................... 64

Risk assessment and ranking on 12 risk factors ........................................ 65

6.1 The dependency of shipping cost on distance and ship capacity .............. 77

Effect of distance on shipping cost ........................................................... 77

Effect of ship capacity on shipping cost .................................................... 78

Effect of delivery time on ship speed ........................................................ 79

6.2 Effect of the canal on unit price and delivery time ................................... 79

Case study results and discussion .............................................................. 81

General Conclusions ................................................................................. 83

7.1 Conclusions ............................................................................................... 83

vii

Specific Conclusions ................................................................................. 84

7.2 Recommendation for future study ............................................................. 85

Sea route measurements data ...................................................................................... 91

viii

List of Tables

Table 1.1: Global export capacity of LNG - current and future data .............................. 12

Table 1.2: Global import capacity of LNG (current and future data) ............................. 13

Table 2.1: Summary of related published research work ............................................... 29

Table 3.1: Input variables ............................................................................................... 38

Table 3.2: Output variables ............................................................................................. 38

Table 3.3: Formulation for output parameters ................................................................ 39

Table 3.4: Risk factors matrix for points calculation ...................................................... 43

Table 4.1: Data for Algeria ............................................................................................. 46

Table 4.2: Data for Angola ............................................................................................. 46

Table 4.3: Data for Argentina ......................................................................................... 46

Table 4.4: Data for Australia ........................................................................................... 46

Table 4.5: Data for Brunei .............................................................................................. 47

Table 4.6: Data for Cameroon ........................................................................................ 47

Table 4.7: Data for Canada ............................................................................................. 47

Table 4.8: Data for Egypt ............................................................................................... 48

Table 4.9: Data for Equatorial Guinea ............................................................................ 48

Table 4.10: Data for Indonesia ....................................................................................... 48

Table 4.11: Data for Libya .............................................................................................. 48

Table 4.12: Data for Malaysia ........................................................................................ 49

Table 4.13: Data for Mauritania ..................................................................................... 49

Table 4.14: Data for Mozambique .................................................................................. 49

Table 4.15: Data for Nigeria ........................................................................................... 49

Table 4.16: Data for Norway .......................................................................................... 50

Table 4.17: Data for Oman ............................................................................................. 50

Table 4.18: Data for Papua New Guinea (PNG) ............................................................ 50

Table 4.19: Data for Peru ................................................................................................ 50

Table 4.20: Data for Qatar .............................................................................................. 50

Table 4.21: Data for Russian Federation ........................................................................ 51

Table 4.22: Data for Trinidad and Tobago ...................................................................... 51

Table 4.23: Data for United Arab Emirates (UAE) ........................................................ 52

Table 4.24: Data for United States of America (USA) ................................................... 52

Table 4.25: Data for Yemen ............................................................................................ 52

Table 4.26: Capacity and speed of LNG carriers ............................................................ 53

Table 4.27: Exporting and importing countries based on market ................................... 55

Table 5.1: LNG infrastructure data for the destination: Bangladesh .............................. 60

Table 5.2: Shortlisted countries ...................................................................................... 65

Table 5.3: Price comparison ........................................................................................... 65

Table 5.4: Distance and delivery time comparison ......................................................... 66

Table 5.5: Capacity comparison ..................................................................................... 67

Table 5.6: Risk on the route comparison ........................................................................ 68

Table 5.7: Risk comparison of the trade dispute with other nations ............................... 69

Table 5.8: Risk comparison based on conflicts internal/external ................................... 70

Table 5.9: Bilateral relationship comparison .................................................................. 70

Table 5.10: Volume-export-import comparison .............................................................. 72

Table 5.11: Economic strength comparison .................................................................... 73

Table 5.12: UN sanctions risk comparison ..................................................................... 73

Table 5.13: Medium/long-term political risk comparison .............................................. 74

Table 5.14: Political stability index comparison ............................................................ 75

Table 5.15: Risk assessment values for each country for each factor ............................ 76

Table 5.16: Points for each risk factor and the total points for each country ................. 76

Table 5.17: Optimal sources for Bangladesh with ranking ............................................ 76

Table 6.1: Optimal destinations for Bangladesh with the country ranking .................... 81

List of Figures

Figure 1.1: Primary direct energy consumption by source ............................................... 3

Figure 1.2: Percentage of total world energy consumptions from 1970 to 2019 ............. 4

Figure 1.3: Current fuel shares of primary energy in 2019 .............................................. 5

Figure 1.4: Energy consumption trend over 1991 to 2019 ............................................... 6

Figure 1.5: Energy consumption trend over 1991 to 2019 ............................................... 7

Figure 1.6: Energy consumption trend from 1991 to 2019 .............................................. 7

Figure 1.7: Coal export market share by country in 2019 ................................................ 8

Figure 1.8: Future global energy source ........................................................................... 9

Figure 1.9: Renewable Energy Share of Global Power Production in 2019 ................... 11

Figure 1.10: LNG Trade volumes from 1994 to 2014 ..................................................... 11

Figure 1.11: Trend of global export and import capacity ............................................... 14

Figure 1.12: A typical LNG process ............................................................................... 15

Figure 1.13: A typical LNG transportation network by ship .......................................... 15

Figure 1.14: LNG exporting facility at Gladstone in Queensland, Australia ................. 16

Figure 1.15: LNG import facility system using FSRU located at Moheshkhali,

Bangladesh ..................................................................................................................... 17

Figure 2.1: LNG value chain .......................................................................................... 23

Figure 3.1: Flow chart for the optimal sourcing model .................................................. 32

Figure 3.2: Basic costing model ..................................................................................... 33

Figure 3.3: Source and destination terminals on Google Maps ...................................... 35

Figure 3.4: Distance measurement with Google Maps web service............................... 36

Figure 3.5: Distance measurement with Netpas software .............................................. 36

Figure 3.6: Excel Model with sample data and calculations of outputs ......................... 40

Figure 4.1: Sources on Google Maps ............................................................................. 45

Figure 4.2: Spot chartering rate (voyage charter type) from 2015 to 2019 .................... 53

Figure 4.3: Major LNG shipping routes and canals ....................................................... 54

Figure 4.4: LNG global market map .............................................................................. 55

Figure 4.5: Wholesale LNG price data from 2010 to 2020 in different markets ............ 56

Figure 5.1: Primary sources of energy in Bangladesh from 1965 to 2019 ..................... 58

Figure 5.2: Gas production in Bangladesh from 2000 to 2019 ...................................... 58

Figure 5.3: Sector-wise gas consumption of Bangladesh in 2018 .................................. 59

Figure 5.4: GDP per capita for Bangladesh from 1990 to 2019 ..................................... 59

Figure 5.5: FSRU at Moheshkhali, Bangladesh ............................................................. 60

Figure 5.6: Location of the FSRU installed in Bangladesh ............................................ 61

Figure 5.7: Distance matrix map from sources to the destination: FSRU Bangladesh .. 62

Figure 5.8: Cost estimation model: sample calculations with data and assumptions ..... 63

Figure 5.9: Snapshot of the processed data for Bangladesh as a destination ................. 64

Figure 5.10: Shipping routes from 12 sources in google maps ...................................... 68

Figure 5.11: Total import growth by market from 2013 to 2018 .................................... 71

Figure 5.12: Total export growth by market from 2013 to 2018 .................................... 71

Figure 5.13: GDP per capita, constant 2010 dollars from 2012 to 2017 ........................ 72

Figure 5.14: Medium/long-term political risk (1=low, 7=high) ..................................... 74

Figure 5.15: Political stability index (-2.5 weak; 2.5 strong) ......................................... 75

Figure 6.1: Shipping cost per unit ($/GJ) as a function of distance. .............................. 77

Figure 6.2: Shipping cost per unit ($/GJ) as a function of ship capacity. ....................... 78

Figure 6.3: Delivery time as a function of ship speed .................................................... 79

Figure 6.4: Price calculations for Algeria ....................................................................... 80

Figure 6.5: Optimal sources of LNG for Bangladesh ..................................................... 81

xii

Nomenclature and Units

Kilometre km

Metric ton = 1,000 kilograms Mt

MMBtu million British Thermal Units

TWh Terawatt-hour

Barrel 1 barrel = 159 litres

xiii

Abbreviations and Acronyms

LNG Liquefied Natural Gas

Carbon dioxide CO2

CIS Commonwealth of Independent States

BP British Petroleum Company plc

EIA Energy Information Administration

OECD Organization for Economic Cooperation and Development

MTPA Million Ton Per Annum

FSRU Floating Storage and Regasification Unit

UN United Nations

GPS Global Positioning System

CIF Cost, Insurance and Freight

xiv

Abstract

Recent data indicates that there is an increasing global demand for natural gas in many

countries as these countries use imported gas to meet their shortfalls. On the other hand,

several countries have started exporting gas in the form of liquified natural gas called LNG.

The trend shows that both export and import are in rising with a second-degree order which

indicates a rapid rise in this energy sector. As many countries are switching from coal to gas

to meet the global carbon emission target by 2050, the trend will be continuing until the

target is reached. As a result, there will be increasing demand for LNG trade in future and the

need for the quest for good trading partners is crucial for the future energy demand of the

importing country. To minimize the overall cost of LNG import ensuring the energy security,

the potential country needs to select an optimum strategy for importing LNG from a suitable

location/country.

Several studies have been carried out by optimising only the cost. They have not considered

some other risk factors such as political and economic factors in their modelling. Currently,

there is no generalised model available to determine the optimal sourcing locations

incorporating important factors including risks in international LNG trading. In this study, 12

risk factors were taken into consideration to evaluate the risk. Therefore, the main objective

of this study is to develop a model for optimal sourcing of LNG by optimising the cost as

well as ensuring energy security for the destination / imported country.

A computational model has been developed to estimate the shipping cost between the source

and destination terminals. This model can be used to estimate the CIF (Cost, Insurance and

Freight) price of LNG from one destination to another including all the cost components of

shipping. Additionally, a strategic tool: optimal sourcing model has been developed to

determine the optimal sourcing countries according to their level of stability. A total of 12

risk factors have been taken into consideration for this optimization technique. This model

can provide a list of best possible sourcing by analysing the risk and benefit on each of the 12

factors and rank them accordingly.

The optimal sourcing model has been developed considering Bangladesh as a prospective

LNG importing country. Results from the model have provided a list of suitable sources. The

results from this study matched well with the Bangladesh government policy for selecting the

souring countries for their future LNG supply for long term basis.

The effectiveness of the model is confirmed with the matching of data for Bangladesh. It is

therefore expected that the model can be used for other countries who want to select optimal

sourcing countries for their future LNG demand by ensuring energy security. Although the

model has been developed for LNG importing country for selecting the best possible sourcing

country, it can also be used to find the prospective future buyer of LNG.

Also, the impact of ship types has been evaluated. Results confirmed that proper selection of

ship for their capacity can minimize the cost significantly. Larger capacity ship with high

speed can be more economical and can reduce the shipping cost up to 40% compared to

smaller ships with relatively lower speed.

Furthermore, the effect of using the canal has been evaluated with a case study. Results

confirm that the uses of Suez Canal to transport LNG from Algeria to Bangladesh can reduce

the delivery time around 35%. However, there are cost, and risk involved to travel through

the channel.

The developed model is suitable for any destination country to find out the optimum sourcing

strategy. On the other hand, LNG exporting countries can also use the same model to identify

optimal destination countries for future trade. The model can also be used for optimising the

sourcing strategy for other energy imports such as oil and coal by updating the input database

based on specific types of commodity.

Introduction

In this chapter, an overview of global energy scenario with a focus on LNG transportation

and optimal sourcing strategy for future demand is presented. The chapter concludes with the

major objectives and layout of the work.

1.1 Global energy scenario

Energy is the main driving force to accelerate the economic progress of a nation. The

progress of modern civilization depends on how much energy they can utilize for the

development of the nation. As the economic growth of a country rises, the power

consumption data also increased accordingly. A prosperous and advanced nation requires

access to sustainable sources of energy. For the past few decades, hydrocarbon fuel source

predominantly oil and coal have contributed to nearly 80% of the total global consumptions

reaching its peak in the 1970s (BP, 2019) and (U.S. Energy Information Administration,

60000

Oil

50000

Gas

40000

Coal

Solar

30000

s r u o h - t t a w a r e T

Hydropower

Nuclear

20000

Wind

Other renewables

10000

Biofuels

1970

1980

1990

2000

2010

2019

2019) as shown in Figure 1.1.

Figure 1.1: Primary direct energy consumption by source [adapted from (BP, 2019) and (U.S. Energy Information Administration, 2019)]

However, the demand has been steadily decreasing due to political (1973 oil embargo) and

environmental issues as shown in Figure 1.2. The need for energy security and cleaner fuel

source, as hydrocarbons produce carbon dioxide and other greenhouse gases which have a

detrimental effect on global warming, has led to developed countries to diversify their energy

source (Perera, 2018). The increase in the percentage of total world energy from sustainable

energy such as solar, wind, biofuels and nuclear energy is shown in Figure 1.2. The total

global energy consumption in terms of percentage for oil and gas has been decreasing

steadily until the early 2000s (BP, 2019). The total global oil consumption has decreased

from nearly 47% in the 1970s to nearly 39% in the 2000s. A similar trend can be observed for

coal demand. Whereas the total global nuclear consumption has increased by nearly 49 times

from nearly 0.13% in the 1970s to 6.73% in the 1990s. Similar trends can also be observed

for gas consumption. The total global gas consumption has been steadily increasing from

50.00

45.00

Oil

40.00

Gas

y g r e n e

35.00

Coal

30.00

d l r o w

Solar

25.00

Hydropower

l a t o t f o

20.00

n o i t p m u s n o c

Nuclear

15.00

Wind

10.00

Other renewables

e g a t n e c r e P

5.00

Biofuels

0.00

1970

1980

1990

2000

2010

2019

17% in the 1970s to 22% in the 2000s.

Figure 1.2: Percentage of total world energy consumptions from 1970 to 2019 [adapted from (BP, 2019) and (U.S. Energy Information Administration, 2019)]

However, during the 2000s, the percentage of total world energy consumption from coal

increased from 25% to nearly 30% in 2010s. This can be attributed to the demand for a cheap

source of energy from developing countries led by China and India. Developing countries

tend to prefer cheap energy sources e.g., coal to keep the energy prices low, however, cheap

energy sources produce carbon dioxide and other greenhouse gases which have a detrimental

effect on global warming. The challenge between providing economic development and

protecting the environment, have been driving the energy source demand and consumption in

the last decade. New technologies and cheaper alternative will help developing countries

move to the less carbon-intensive global energy economy. Since the 2010s, renewable energy

has seen an increase in total world energy consumption (REN21, 2018) and (IRENA, 2019).

Countries are moving away for high CO2 emitting sources such as oil and coal to more green

and sustainable energy such as renewables and gas. Another driver is the Paris Agreement

where countries pledged to combat climate change by keeping global temperature rise below

2 degree Celsius above pre-industrial levels (Rogelj et al., 2016).

Looking at the current global energy mix as shown in Figure 1.3, hydrocarbons are still

having the largest share of primary energy led by oil at 33.1%, gas at 24.2% and coal at 27%.

Renewable and sustainable energy sources particularly wind, solar and hydro energy makes

Nuclear 4.3%

Hydro 6.4%

Oil 33.1%

Renewables 5.0%

Coal 27.0%

Gas 24.2%

up nearly 11.4% of the share.

Figure 1.3: Current fuel shares of primary energy in 2019 [adapted from (BP, 2019)]

The energy consumption trend from different regions is shown in Figure 1.4. The increase in

global energy consumption has been spurred by the demand from Asia especially China and

India. Since the 2000s, the energy consumption in this region increased by 1.3% followed by

the Middle East region around 1.24%, Africa around 0.73%, Latin America around 0.42%,

CIS regions around 0.22% and North America around 0.02%. It is interesting to note that the

energy consumption in Europe decreased by around 0.04%.

16000

14000

12000

Middle East

10000

Africa

Asia

t n e l a v i u q e l i o f o

8000

Latin America

6000

e n n o t

North America

4000

n o i l l i

CIS

2000

Europe

1990

1995

2000

2010

2015

2019

2005 Year

Figure 1.4: Energy consumption trend over 1991 to 2019 [adapted from (BP, 2019)]

However, it is interesting to note that energy consumption has decreased in most of the region

except North America, in this decade compared to the 2000s. The largest decrease in energy

consumption is in the Asia Pacific region followed by the Middle East, Africa, Latin

America, CIS regions and Europe. The energy consumption decreased by 0.75%, 0.72%,

0.4%, 0.3%, 0.11% and 0.01%. Respectively. One factor that may be contributing to the

decrease in energy consumption in these regions is due to advancements in technologies

Global oil scenario

which is more energy efficient (Council, 2019).

The total world proved oil reserves are around 1733.9 thousand million barrels at the end of

2019 (BP, 2019). The top ten countries with highest oil reserves are shown in Figure 1.5.

Venezuela has the highest oil reserves at 303.8 thousand million barrels which are around

17.5% of the total world reserves. It is followed by Saudi Arabia, Canada, Iran, Russian

Federation, Kuwait, UAE, US and Libya with the total around 297.6, 169.7, 155.6, 145,

107.2, 101.5, 97.8, 68.9 and 48.4 thousand million barrels respectively.

2000

1800

1600

1400

1200

1000

s l e r r a b n o i l l i

800

600

m d n a s u o h T

400

200

Venezuela

Canada

Iran

Iraq

Kuwait

UAE

Libya

World

Saudi Arabia

Russian Federation

Global coal scenario

Figure 1.5: Energy consumption trend over 1991 to 2019 [adapted from (BP, 2019)]

The world’s current coal deposits are estimated to be around 1,069,636 million tons (BP,

2019) as shown in Figure 1.6. The major coal deposits are in the US, Russia, China, India and

Australia. Coal currently supplies 41% of global electricity and is expected to grow in

300000

250000

200000

150000

s e n n o t n o i l l i

100000

50000

Russia

Australia

China

India

Indonesia Germany Ukraine

Poland Kasakhstan

coming decades,

Figure 1.6: Energy consumption trend from 1991 to 2019 [adapted from (BP, 2019)]

The leading coal exporters accounted for nearly 95% of all coal exports in 2019

(Cunningham, Van Uffelen, & Chambers, 2019). The top eight coal exporting countries are

shown in Figure 1.7. The largest coal exporter was Indonesia, which exported nearly 455

million tonnes (Mt) in 2019 or nearly 38.3% of total global coal exports. Australia exported

nearly 393 Mt followed by Russia. United States, South Africa, Colombia, Canada, and

500

450

400

350

300

s n o t

250

n o i l l i

200

150

100

Indonesia Australia

Russia

Colombia

South Africa

Mongolia.

Figure 1.7: Coal export market share by country in 2019 [adapted from (BP, 2019)]

According to (IEA, 2019), global coal use is projected to remain flat. The total coal demand

in Organisation for Economic Co-operation and Development (OECD)1 regions and China is

predicted to decline offsetting the growth in India and other non-OECD Asian nations.

To reduce the dependency on coal and environment pollution, the use of natural gas for

power generation is increasing rapidly. Currently, generation by natural gas is around (23%),

International Energy Agency and Energy Information Administration (EIA) (2018). Most

peaking power generation in developed countries (Australia, USA, and the UK) is undertaken

by natural gas. However, due to limitation of access to other fuels, some countries fully or

over 80% of their power generation are achieved by using gas (Qatar, United Arab Emirates,

Bangladesh, and Nigeria), International Energy Agency (IEA) (2018). The fast-developing

nations and some developed nations in Asia use natural gas for power generation. These

countries do not have enough gas from their domestic source. Hence, they import gas from

1Current OECD member States: Australia, Austria, Belgium, Canada, Chile, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, UK, and US.

foreign countries as Liquefied Natural Gas (LNG) to make easier transportation.

Future sources of primary energy

By 2060 the global power generation will increase to 44,914 TWh, an increase of around

89% when compared to 2019, according to the BP statistics (BP, 2019). The non-fossil

energy source is predicted to provide around 54% of the total global power generation as

shown in Figure 1.8. The nuclear and hydropower sector is expected to lead the way with

15% each followed by wind, solar and biomass at 13%, 7% and 4% respectively. Gas and

coal are projected to be a major contributor from fossil fuel sources at 27% and 18%

Biomass 4%

Oil 1%

Solar 7%

Gas 27%

Wind 13%

Hydro 15%

Coal 18%

Nuclear 15%

respectively, Energy Information Administration (EIA) (2019),

Figure 1.8: Future global energy source [adapted from (BP, 2019)]

The oil demand is projected to experience a reduction in coming years due to the Paris

Agreement2 which aims to stabilize the global average temperature and limit greenhouse gas

emissions. Unless new oil reserves are discovered, fossil fuel is projected to deplete by 2070

at current world energy consumption rate. Coal is forecast to fuel 18% of the total global

power production by 2060. Since 2000, coal consumption for power has increased by 64%.

As mentioned previously, coal remains a major primary source of energy in Asia due to low

cost which is around (2.5 ¢ per kWh) and is expected to be the main power source in the

2 The Paris Agreement entered into force on 4 November 2016, thirty days after the date on which at least 55 Parties to the Convention accounting in total for at least an estimated 55 % of the total global greenhouse gas emissions have deposited their instruments of ratification, acceptance, approval or accession with the Depositary.

foreseeable future (BP, 2019).

The use of Nuclear Power is on the rise globally. In 2017, 68 nuclear reactors for power

generation were under construction with a total capacity of 66,000 MW (Alam, 2019) (Cobb,

2020). Forty of such units under construction were in China, India, Russia and South Korea.

Liquefied Natural Gas (LNG) is expected to play a role in power generation.

1.2 Advantages of LNG

The countries that do not have access to their natural gas are importing natural gas by

pipeline (if geographically possible) and natural gas as liquid form (e.g. Liquefied Natural

Gas - LNG). Natural gas accounts for around 24.23% of the world’s power generation around

39292 TWh, (BP, 2019). Gas in the form of LNG is the easiest means to import as these

regions are not easily served by pipelines. Japan remains the world’s largest importer of LNG

around 22% of the world’s LNG import followed by China (17%), South Korea (11%), India

(7%) and Taiwan (5%)

LNG accounted for nearly three-quarters of capacity growth between 2000 and 2019 in

developed economies. The shift towards gas power generations is due to lower capital cost

and shorter project lead times compared to coal and nuclear power plants. Furthermore,

flexibility in providing either peak or baseload power or supplement to the intermittent

renewable source makes it an attractive energy source.

The use of Liquefied Natural Gas (LNG) for power generation has been rapidly increasing as

an alternative to coal. The leading top ten exporters accounted for nearly 59% of total global

exports in 2019. The order of top five exporters by share is Qatar (29.9%), Australia (17.2%),

Malaysia (9.7%), Nigeria (7.2%), Indonesia (6.4%), Algeria (4.5%), Russia (4.2%), Trinidad

(4.1%), Oman (3.2%) and Papua New Guinea (2.9%), . The total exports in Million Ton Per

Annum (MTPA) are shown in Figure 1.9. Australia export is nearly half of Qatar’s export but

is expected to surpass Qatar by 2021.

PNG 3%

Oman 3%

Trinidad 5%

Russia 5%

Qatar 34%

Algeria 5%

Indonesia 7%

Nigeria 8%

Malaysia 11%

Australia 19%

Global LNG market and future demand

Figure 1.9: Renewable Energy Share of Global Power Production in 2019 [adapted from (IGU, 2020)]

The total global LNG trade has been steadily increasing for the last decade as more countries

import and export LNG. The global LNG trade in volumes in million tons per annum

(MTPA) from 1994 to 2014 is shown in Figure 1.10.

Figure 1.10: LNG Trade volumes from 1994 to 2014 [adapted from (IGU, 2015)]

The Asia Pacific trade in the largest trade flow route accounting to almost 77.3 MT. This is

no surprise as the highest exporters Australia and Malaysia and the highest importers Japan,

China, Singapore, Thailand and South Korea. The trade flow from the Middle East to Asia

accounted for about 36.3 MT led by Qatar which remains the third-highest trade flow. The

other trade routes being Africa, which exports mainly to Europe and Asia, followed by North

America, Russian, Latin American and European trade flows. Table 1.1 listed the global

export capacity of LNG including the current and future data until 2025.

Table 1.1: Global export capacity of LNG - current and future data [adapted from : (IGU,

Current capacity until 2019

Future Capacity by 2025

2020)]

Country

Sum of Capacity (MTPA)

93.95 87 77.1 48.96 32 30.85 30.2 25.5

87 77.1 46.6 30.5 26.76 26.55 25.5 22.2

United States Australia Qatar Russia Malaysia Indonesia Nigeria Algeria

and

Australia Qatar United States Malaysia Russia Indonesia Algeria Nigeria Trinidad Tobago

16.28

14.8

and

14.8 14 12.2 10.4 7.2

12.2 10.4 7.2 6.9 6.7

Mozambique Trinidad Tobago Canada Egypt Oman Brunei

Arab

Egypt Oman Brunei Papua New Guinea Yemen United Emirates Angola

6.9 6.7

5.8 5.2

Arab

Papua New Guinea Yemen United Emirates

Peru Norway Equatorial Guinea Libya Cameroon Argentina Grand Total

4.45 4.2 3.7 3.2 2.4 0.5 429.86

5.8 5.2 4.45 4.2 3.7 3.2 2.5 2.4 0.5 545.99

Angola Peru Norway Equatorial Guinea Libya Mauritania Cameroon Argentina Grand Total

On the other hand, currently, there are 37 countries are importing LNG in 2019 with a total

capacity of the LNG receiving facility around 819 MTPA within a more than 50% utilization

rate. 8 more countries: Jordan, Greece, Philippines, Ghana, Croatia, Vietnam, Cyprus, El

Salvador are building their import terminals and expected to rise in the global LNG demand

by 2025 with a 12% increase in a total global capacity. Table 1.2 listed the global import

capacity of LNG including the current and future data until 2023.

A. Current capacity until 2019

B. Future Capacity by 2023

Country

Japan

Table 1.2: Global import capacity of LNG (current and future data) [adapted from : (IGU, 2020)]

Japan South Korea China United States Spain India UK France Turkey Mexico Taiwan Brazil Thailand Singapore Italy Pakistan Netherlands Indonesia Bangladesh Canada Malaysia Belgium United Arab Emirates Portugal Kuwait Egypt Chile Argentina Jordan Poland Jamaica Greece Israel Lithuania Colombia Dominican Republic Panama Grand Total

Sum of Capacity (MTPA) 210.5 125.4 77.42 45.4 43.8 39.5 38.1 25 21.4 16.8 14 12.8 11.5 11 11 9.5 9 8.6 7.55 7.5 7.3 6.6 6 5.8 5.8 5.7 5.5 3.8 3.8 3.6 3.6 3.3 3 3 3 1.9 1.5 818.97

South Korea China India United States Spain UK Kuwait Turkey France Mexico Thailand Brazil Taiwan Indonesia Singapore Italy Pakistan Netherlands Bangladesh Canada Malaysia Belgium United Arab Emirates Bahrain Portugal Egypt Chile Jordan Argentina Jamaica Poland Greece Lithuania Philippines Israel Colombia Russia Ghana

Sum of Capacity (MTPA) 211 125.4 91.42 58.5 45.9 43.8 38.1 27.8 26.8 25 19.8 19 18.4 14 11 11 11 9.5 9 7.55 7.5 7.3 6.6 6 6 5.8 5.7 5.5 3.8 3.8 3.6 3.6 3.3 3 3 3 3 2.7 2

Dominican Republic Croatia Panama Vietnam Cyprus El Salvador

Grand Total

1.9 1.9 1.5 1 0.6 0.5 916.57

Figure 1.11 shows the trend of global export and import capacity. Both export and import are

in rising with a second-degree curve which indicates a rapid rise in this energy sector. As

many countries are switching from coal to gas to meet the global carbon emission target by

2050, the trend will be continuing until the target is reached. As a result, there will be

increasing demand for LNG trade in future and the need for the quest for good trading

1000

900

800

)

700

Importers Importers future demand Exporters Exporters future production

600

A P T M

(

500

400

y t i c a p a C

300

200

100

0 1960

1970

1980

1990

2000

2010

2020

2030

Year

partners is crucial for the future energy demand of the importing country.

LNG transportation system

Figure 1.11: Trend of global export and import capacity

LNG is natural gas cooled down to condensate at around -161 °C. Natural gas consists

primarily of methane. Impurities and heavy hydrocarbons from the gas fuel are removed

before the cooling process. The density of LNG is roughly in the range of 410 to 500 kg/m3.

The volume of the liquid gas is approximately 1/600th of the gaseous volume at atmospheric

pressure. It is odourless, colourless, non-toxic and non-corrosive. Hazards include

flammability after vaporization into a gaseous state, freezing and asphyxia in liquid form, it is

inflammable (IGU, 2015). LNG is cost-efficient to transport over long distances where

pipelines do not exist. A Typical LNG Process is illustrated in Figure 1.12 (Mustary,

Chowdhury, Loganathan, & Alam, 2017).

Figure 1.12: A typical LNG process

Ocean standard specially built ship with different capacity and propulsion systems and types

are used to transport LNG via sea routes between loading and unloading terminals. A Typical

LNG transportation network is illustrated in Figure 1.13 including the infrastructures required

in the loading and unloading terminals.

Figure 1.13: A typical LNG transportation network by ship, adapted from (Mustary et al.,

2017).

A typical LNG export facility system requires the following infrastructures:

• Liquefaction Unit

• Storage Unit

• Shipping terminal

A typical LNG import facility system requires the following infrastructures:

• Shipping terminal

• Storage Unit

• Regasification Unit

A storage unit can be land-based (onshore) or floating type. Also, the storage unit and

regasification unit can be combined into a single unit called Floating Storage and

Regasification Unit (FSRU) as shown in Figure 1.14. FSRU is usually installed and docked

into the receiving terminal and connected to the main grid so that gas can be supplied on

demand. Also, can be refilled by the transporter LNG carrier as shown in Figure 1.15.

Figure 1.14: LNG exporting facility at Gladstone in Queensland, Australia, [adapted from (Loganathan, 2018)]

Figure 1.15: LNG import facility system using FSRU located at Moheshkhali, Bangladesh [adapted from (PetroBangla, 2018)]

1.3 Factors for optimal sourcing strategy for LNG importers

It is important to minimise the transportation cost as low as possible and maximise energy

security as high as possible. Some locations are good for a group of the country but not for

all. Several factors involved in trade from one location to another. These are:

• Production capacity of the source terminal

• Price at the source

• Distance between source and destination terminals

• Risk on the transportation route

• Trade dispute between source country with other nations

• Source country’s involvement in internal/external conflicts or warfare as geopolitical

conflicts in source countries can disrupt LNG supply

• Bilateral relationship between source and destination country

• Trade relationship/volume of export-import between source and destination country

• Economic strength of source country as economic strength provides internal security

which provides LNG supply security

• Any economic or UN sanctions on source country which can create supply disruptions

• Medium/long-term political risk index of the source country as a politically stable

country can provide energy security is for a sourcing country is safer

• Political stability index of the source country as the politically stable country provides

Production capacity of the source terminal

safer and reliable energy security

Every source terminal has its own specific LNG production capacity per year expressed in

MTPA. Higher production capacity indicates higher energy security. A country with a higher

production capacity can be more reliable than that with a lower capacity. As a result, the risk

of supply can be minimised. Therefore, it is important to make sure the source terminal can

Price at the source

supply the minimum base load at the destination terminal to ensure the energy supply.

Price plays an important role in the selection of an optimal source. The unit price of LNG

varies depending on the market location and type of contracts, for example, short-term or

long-term contracts. Usually, short term or spot purchase contracts are subjected to more

price volatility whereas long-term contract offers less fluctuation in price and ensures more

Distance between source and destination terminals

energy security.

Delivery time depends on route length. If the route length increase delivery time is also

increased accordingly. Distance is not only increasing the delivery time but also increased the

cost. So, it was preferable to select the sources from the same region or market rather than

Risk on the transportation route

going to a distant place. The chances of risk increasing with the increasing of the distance.

There are several risks involved in LNG trade:

• Typical issues with LNG transportation on the surface:

o Excessive corrosion of pipeline infrastructures

o Risk of pipeline integrity loss

o Blocking of furnace nozzles

o Operating problems at end-users

o Contamination of instrumentation and control valves

o Storage of impurities on the pipe network

• Safety and Accidents:

o Leaking and evaporation

o Risk of fire and thermal radiation hazards

o Risk of Subversive Attacks to LNG Facilities, Infrastructures and Transports

• Risk Assessment

o Risk assessment can be qualitative or quantitative.

o Qualitative risk assessment deals with the identification of possible accident

scenarios and attempts to estimate the undesirable consequences.

o The goal is to identify events that appear to be most likely and those with the

most severe consequences and focus on them for further analysis.

o Frequency analysis can be done to evaluate the risk using historical

information on accident frequencies.

In this study, any risk of attack or blockade on the route during the LNG transportation will

be considered for evaluation by analysis the historical data on the surrounding area. For

example, to transport LNG from some countries, the ship may need to use canals (Suez or

Panama) to pass through the channel. Because of the blocked due to maintained and

construction, there may be risk involved in this route. Similarly, the ship needs to pass

through a channel where there are conflicts in the surrounding area, for example, Strait of

Trade dispute between source country with other nations

Hormuz.

Possible destination restriction due to trade sanctions can create disruptions to energy

security. Increase in tariffs imposed on the source country will affect the export price. It is

economically advantaging to choose sourcing that is politically stable and that has no major

trade disputes.

Source country’s involvement in internal/external conflicts or warfare

The sourcing country is involved in any war or conflicts with the other nation it may impact

the trade and there will be always risking in the trade. Internal conflicts also a great risk to

infrastructure and trade. Possibilities of war and conflicts create very volatile satiation and is

Bilateral relationship between source and destination country

an unreliable source of energy supplier.

The bilateral relationship allows for more flexibility in supply volume and cost flexibility

Trade relationship/volume of export-import between source and

destination country

between source and importing country. This allows for lower importing cost.

Trade relationship allows for more flexibility in supply volume and costs flexibility. Allows

for a potential LNG supply and sustainability/security of supply over the medium to long

Economic strength of the source country

term.

Offers guarantees for exported volumes and a fixed price formula. It also allows for the

development of the regional gas and LNG supply (long-term and spot/short-term) in regional

and international markets and this could offer increased supply volume for future increase in

Any economic or UN sanctions on the source country

demand for LNG supply.

UN sanctions on sourcing country will severely affect the energy security of an importing

Medium/long-term political risk index of the source country

country. Countries must look for alternate supply country which will affect the cost.

One of the strategies to choose a potential sourcing country is to assess the potential of a risk

caused by political uncertainty usually beyond one year. A matrix is used as a quantitative

model to quantify countries into seven categories from 1 (low risk) to 7 (high risk). Countries

that are in volatile regions usually have a higher risk of political instability and is not a

Political stability index of the source country

reliable source of energy security.

Another sourcing strategy is to assess the political stability index of the source country. The

index of Political stability is the measure of the likelihood that a government will be

overthrown by unconstitutional or violent means. Political stability and economic growths are

deeply interconnected. A change in government can sometimes have a significant impact on

business. The new government may be less business-friendly and this can affect the taxation

and regulation which will have an impact on energy security.

1.4 Objective and scope

Currently, there is no such model available to find the optimal sources for a given destination

incorporating all the risk factors identified in section 1.3 to ensure energy security. Therefore,

the main objective of this study is to develop a model for optimal sourcing of LNG by

optimising the cost as well as but also ensure energy security for the destination.

In this study, the global shipment by sea route for LNG transportation will be considered.

Weather condition, ship efficiency, tanker risks are not considered for this study. The study

focuses on developing a model incorporating risk factors including the geographical

locations, foreign affair, economic and political conditions of the source and destinations.

A detailed literature review has been performed in Chapter 2.

1.5 The rationale of the research

Recent data indicate that the liquification capacity is increasing exponentially over the year

and the degasification capacity is increasing globally as more and more countries started

LNG importing into this market to meet their future energy demand. Also, the LNG price is

decreasing in recent years and the price is expected to be decreased shortly. On the other

hand, there is a global target to limit carbon emission by 2050. To meet this target more and

more countries are switching from coal to natural gas. As a result, there will be an increasing

demand for LNG soon and a lot more country will be coming to the market. So, there is a

need for a generalised mathematical model to find out the optimum sourcing strategy. As

currently there is no such a model existing to optimise the source and destination. To

minimise the cost and to provide energy security. To address this problem, a computational

model has been created to optimise the cost of the destination country a lot of countries who

are going to enter the market will can use this model to optimise. LNG exporting countries

can also use the same model to identify optimal destination countries for future trade. This

model can be used for other commodities such as coal, and oil.

Literature Review

This chapter reviews the current literature related to the main objective of this study. A

summary of the current knowledge is tabulated based on the research parameters and gap in

the research were identified. Research questions were formulated.

2.1 Current body knowledge on optimal sourcing strategy of

LNG

The increases in demand for global LNG have attracted increasing attention in the LNG value

chain (Cho, Lim, Kim, & Biobaku, 2018) & (Christiansen, Fagerholt, Nygreen, & Ronen,

2013). The LNG value chain compromises of 5 phases. The first 2 phases are the Gas

production and the liquefaction & LNG storage phase. These 2 phases are located in the

producing country. The stored LNG is then loaded onto an Ocean ship and is sent to the

sourcing country which is the third phase. The fourth and fifth phase is the unloading stage in

the sourcing country. The LNG value chain is shown in Figure 2.1.

Figure 2.1: LNG value chain [adapted from (Cho et al., 2018)]

LNG Inventory Routing aims to assist the producer and supplier in providing optimal

production inventory schedule and transportation plans to satisfy the demands from the

customer (Cho et al., 2018). It also helps the LNG supplier to maximise profit. In the first

three phases, the LNG supplier must work out how to maximum capacities for production

and storage, the total number of LNG carriers per type, and the shipping schedule.

The third phase, the shipping phase, the ships are either operated by the producer as a mixed

fleet of ships. These ships are either owned by the producer or by one, or a group of

customers. However, all ships are included in the fleet scheduled by the producer. The third

phase the shipping phase will be further reviewed in this thesis. Efficient design and

operation of the world fleet increase its productivity and revenue. Ship routing and

scheduling and optimal sourcing strategy are the focus of this review, is the major

Ship routing and scheduling

determinant of the fleet productivity.

LNG shipping presents several additional challenges beyond the coordinated planning of

inventories and routes. As the demand for Liquefied Natural Gas (LNG) transportation

increases, it is important to have an LNG procurement strategy which optimises the total cost

to deliver the required LNG volume supply commitments and to optimise the ship fuel

consumption (Palti-Guzman, 2018).

An LNG ship generally can store different types of LNG however it can only store a single

type of LNG on a single voyage. It is paramount that an LNG ship is filled up to capacity at

the loading port to avoid sloshing and minimise the transportation cost (Kuo et al., 2009). The

sourcing county capacity and the LNG ship loading capacity is an important factor.

Boil-off is another factor that affects the volume of LNG delivery capacity. The longer the

distance from the sourcing country to the customer, the higher the evaporation. It is estimated

between 0.11% and 0.13% of the total tank capacity will evaporate each day (Dobrota, Lalić,

& Komar, 2013). It is also a normal practice to keep a certain amount of LNG in the tank to

keep the LNG tank onboard a ship cools during the return journey. An empty tank will

require 24 hours cooling down period at the loading port which is uneconomical.

Another important factor that determined the duration of the voyage is the cruising speed, the

sailing conditions and the policies at the port (Aydin, Lee, & Mansouri, 2017). Some ports

only accept their owners ship for regasification terminals.

The transportation costs consist of several components. The fixed costs are the time charter

rates, while the variable costs are port and canal fees that are determined by the ship type and

the contract served. A third variable component is the bunker cost, which is dependent on the

ship size, the load on board and the duration of the voyage.

Since the fixed costs cannot be changed during the time horizon, the cost of sailing a

scheduled voyage is assumed to only be dependent on the capacity class of the ship, the

duration of the voyage and the regasification terminal visited.

Sailing cruise speed

Hennig et al. (2017) carried out to illustrates how operations research can be applied to large-

scale maritime crude oil transportation. Complex variables like quantity requirement at the

loading port, problems of pickups and delivery and ship routing was assessed. They defined

their model using ship speed, quantity dependent pickup/delivery times and costs, restrictions

on port access, ship weight and volume capacity and time window sizes of several days. They

introduce continuous variables to distribute the cargo among the different routes in their

model. They suggested using cutting planes into the solution algorithm and devising a

branch and cut and price algorithm to optimise the cost. However, Hennigs and his colleagues

did not include the optimal sourcing strategy in their model (Hennig, Nygreen, Furman,

Song, & Kocis, 2017).

Kontovas et al. (2011) examined speed reduction to reduce fuel consumption and curb

emissions. Their investigation was aimed at reducing port service time. They investigated an

operational scenario, focusing on speed reduction, regarding container vessels. They also

looked at the relation between engine load and specific fuel consumption using regression

analysis to identify a more appropriate relation between fuel consumption and vessel speed

for container vessels. They found that speed reduction under certain conditions is beneficial

in terms of reducing emissions. The proposed (the ‘‘booking by rendezvous’’ system) to

minimise the time in port which can offset the loss of time due to speed reduction. However,

Kontovas and his colleagues did not investigate the effects of sea conditions in their research

(Kontovas & Psaraftis, 2011).

Ronen (2017) developed a model relating to the annual operating cost of containerships on a

route to their sailing speed and developed a procedure to identify the optimal sailing speed.

In the model, the following assumptions were made. The first one is the route (sequence of

calling ports) is already specified. The second assumption is a weekly service frequency is

required, and the third is the vessels that operate on the route have similar physical and

economic characteristics and for practical purposes can be considered identical (usually sister

ships operate on a route). However, Ronen’s model did not investigate the effect of sea

conditions in research (Ronen, 2017).

Type and size of the ship

Rakke et al. (2011) formulated a maritime routing problem (MIP) model using a mixed-

integer programming (MIP) formulation. The goal was to create an annual delivery program

(ADP) to fulfil the long-term contracts at minimum cost while maximizing revenue from

selling LNG in the spot market. Variables for each ship, customer contract and time. problem

is presented, and it is based on a priori generation of all possible scheduled voyages within

the planning horizon. The variable studied include number of ships, the number of contracts

for each gas type, the length of the planning horizon, the number of births for each gas type,

and a range for the number of scheduled voyages needed are given. A rolling horizon

heuristic (RHH) is proposed due to the size and complexity of the problem. The RHH solves

the problem by iteratively solving sub-problems with shorter planning horizons. The RHH

finds solutions to the problem within a reasonable amount of time and creates very good

Type and frequency

ADPs according to the problem owner (Rakke et al., 2011).

Meng et al. (2011) propose an optimal operating strategy problem arising in the liner

shipping industry that aims to determine service frequency, containership fleet deployment

plan, and sailing speed for a long-haul liner service route. The problem is formulated as a

mixed-integer nonlinear programming model that cannot be solved efficiently by the existing

solution algorithms. this paper proposes an efficient and exact branch-and-bound based e-

optimal algorithm. Meng and colleague determined the type and number of containerships for

the route, frequency of service, and sailing speed for each leg that minimizes average daily

Sea condition and route

operating costs (Meng & Wang, 2011).

Lindstad et al. (2013) assess the shipping profits, costs and emissions by speed and as a

function of sea and freight market conditions. They propose a model consisting of five main

equations, of which the power element is the most important. The variables studied are

propeller efficiency, the power needed for still water conditions, the extra power required for

waves, the power needed for wind, and the necessary auxiliary power, as a function of a

vessel’s speed and cargo load. The results show that significant cost and emissions reductions

can be achieved and that the maximum economic speeds based on hydrodynamic

considerations even in a good freight market are lower than the design speeds. However, their

model did not consider political risk (Lindstad, Asbjørnslett, & Jullumstrø, 2013).

Norstad et al. (2011) investigated the optimal sailing speed by considering a heterogeneous

fleet of tramp vessels with a set of mandatory and optional cargoes and associated port time

windows. They assigned cargoes to vessels and determined the optimal speed for each sailing

leg while maximizing profit (income from optional cargoes minus fleet operating costs).

They demonstrated that by incorporating sailing speed as a decision variable when planning

vessel routes significantly improves fleet utilization and profit (Norstad, Fagerholt, &

Laporte, 2011).

Stalhane et al. (2011) studied the ADP planning problem for one of the world’s largest

producers of LNG. They developed a construction and improvement heuristic (CIH) by

studying the products, inventory and berth capacity at the loading port and a heterogeneous

fleet of ships (Rakke et al., 2011).

Kim et al. (2012) looked at a single vessel on a single route to minimize the bunker fuel-

related costs and the cost of the ship’s time. They used an algorithm to minimize the fixed

ordering, purchase, and inventory carrying costs associated with bunker fuel together with

ship time costs; and environmental costs associated with greenhouse gas emissions. It

determines the optimum ship speed, bunkering ports, and amounts of bunker fuel for a given

ship’s route. The results indicate that the shipping speed and CO2 emissions are highly

sensitive to the factors considered (Kim, Chang, Kim, & Kim, 2012).

Yao et al. (2012) looked at a single shipping line with a homogeneous set of vessels and

specified average sailing speed. Their research considered the potential loss of cargo revenue

due to bunker fuel weight. They determined that the bunkering ports, bunker fuel quantities

and sailing speed adjustments for each leg minimize the total bunker fuel-related costs (Yao,

Ng, & Lee, 2012).

Windeck & Stadtler (2011) studied whether environmental influences and the use of

additional propulsion systems can influence the cost structure or the on-time delivery of a

liner shipping network and its schedule. Their study proposes a model and a heuristic solution

algorithm for network design where the focus is on the reduction of emissions from fuel

consumption while taking into account waves, currents and wind (Windeck & Stadtler,

2011).

Sourcing strategy

Geng, Ji, Fan, & Shaikh (2017) studied to determine the optimal portfolio strategies

for liquefied natural gas (LNG) importation based on multi-objective optimisation models.

They used a quantitative maritime transportation risk indicator by considering maritime

transportation distances and the risk of pirate attacks. They also considered the economic

risks of importation, the exporting countries' political risks and the maritime transportation

risks associated with developing optimal LNG importation strategies (Geng, Ji, Fan, &

Shaikh, 2017).

Berle, Norstad, & Asbjørnslett (2013) tried to systematically address the vulnerability in a

maritime transportation system for cost efficiency calculations. They used a formal

vulnerability assessment approach to create quantitative measures of disruption risk and test

the effect of mitigating measures. They used supply chain simulation using heuristics-based

planning tools to quantify the impact of disruption scenarios and mitigating measures. They

found the exemplary assessment of a maritime liquefied natural gas (LNG) transportation

system illustrates the potential for providing quantitative data about the cost of disruptions

and the effects of mitigating measures, which are foundations for more precise cost efficiency

estimates. The combination of risk assessment methods and inventory routing simulation of

maritime supply chain problems is a novelty. Quantifying vulnerability, effects of disruptions

and effects of mitigating measures in maritime transportation systems contribute to a little-

researched area (Berle, Norstad, & Asbjørnslett, 2013).

2.2 Summary of current studies on LNG sourcing and research

gap

The number of published research articles in this area is significantly limited. The limited

publications may be categorised according to study parameters. work summarises these

distributions.

Table 2.1: Summary of related published research work

References

Price

Speed

Distance

Capacity/Size of Ship

Risk on the route

Economic condition

Political condition

Foreign affair

Yes

Hennig et al. (2011)

Yes

Rakke et al. (2011)

Ronen (2011) Yes

Yes

Meng et al. (2011)

Yes

Lindstad et al. (2013)

Yes

Reinhardt et al. (2012)

Yes

Norstad et al. (2011)

Yes

Stålhane et al. (2012)

Yes

Kim et al. (2012)

Yes

Yao et al. (2012)

Yes

Windeck et al. (2011)

Yes

Geng et al. (2017)

Yes

Berle et al. (2013)

Literature review summary as shown in Table 2.1 indicates that currently, there is no

generalised model available to determine the optimal sourcing locations incorporating

important factors including risks in international LNG trading.

2.3 Research questions

Following research questions have been formulated to address in this study:

• What are the important parameters that influence the cost or security of LNG

transportation?

• How to estimate the cost of LNG transportation?

• How the cost of LNG transportation can be minimized?

• How the risk of LNG transportation can be assessed?

• How cost-effectively LNG can be transported from the exporting country to an

importing country?

• What will be the cost prediction model for an importing country?

• How can the model be implemented to find out optimal sourcing country?

Methodology and Model

Development

In this chapter, initially, a model has been developed to estimate the LNG price including the

transportation cost via sea route for a given destination as an input. Necessary parameters

have been identified to formulate the computational model. Additionally, a qualitative and

quantitative analytical methodology has been devised to assess the risk to ensure energy

security and rank the sourcing locations for an intended destination.

3.1 Methodology

To develop a generalised optimal LNG sourcing model for a given destination, the following

2 models were devised:

• Price estimation model

• Optimal sourcing model

Following steps were carried out for price estimation model:

• Identification of important parameters associated with LNG transportation via sea

routes

• Definition of all the input and output variables for the model

• The formulation for the output variables

• Development of a computational model using computer software

• Testing the model using sample data and verify its correctness

Following steps were carried out for optimal sourcing model:

• Identification of key factors that affect the LNG trade between exporting and

importing countries

• Development of the optimal sourcing model by qualitative and quantitative output

data analysis to rank the sourcing locations for a given destination

A detailed flow chart is shown Figure 3.1 for the optimal sourcing model for this study.

Figure 3.1: Flow chart for the optimal sourcing model

3.2 Price estimation model

A basic costing model for LNG price from source to the destination is illustrated in Figure

3.2. It shows that the total cost at the destination can be calculated if three components of cost

(i.e. gas production cost, liquefaction cost and shipping cost) are known. In this model, gas

production cost and liquefaction cost for unit volume have been collected from the external

data sources. Also, the transportation cost can be calculated depending on the location of the

sources and the destinations and route preferences. Collection of the necessary data for this

model will be discussed in chapter 4.

Figure 3.2: Basic costing model

The model also requires input values for the properties of LNG carrier: ship capacity and

Important parameters associated with LNG transportation cost

speed including the chartering rate and associate cost components.

Transportation cost of LNG mainly depends on the following factors:

• Selection on type of LNG carrier: capacity, speed

• Type of charting: Chartering fee including brokerage charge

• Time of the booking for the journey

• Route length (depends on the location of the source and destination)

• The toll on the route: canal charges (e.g. Panama or Suez charge)

• Other charges: insurance of goods, demurrage

3.2.1.1 Speed and capacity of LNG carrier

• Loading and unloading time at the ports

There are various types of ships are used to transport LNG from one port to another. Capacity

and speed of the LNG carrier are two important factors that are directly related to

transportation cost. The unit cost of LNG transportation decreases with the increase of the

capacity. On the other hand, transportation cost decreases with the increase of ship speed as

the delivery time reduces as well as the total chartering fee. LNG carrier with newer

propulsion system is more fuel-efficient than that of the older ones. As a result, charting fee

of the recently built LNG carriers is relatively cheaper than the older technology ships. If the

LNG carrier is fuel-efficient, then it will consume less fuel. As a result, it can be hired with

3.2.1.2 Methods of chartering

less cost and the overall transportation cost will be lower.

Following are the three main chartering methods used for LNG carrier hiring:

• Demise charter

• Voyage charter

• Time charter

Demise type charter is usually for a long-term contract where charterer pays for all operating

expenses, including fuel, crew, port expenses and protection and indemnity insurance and

hull insurance. This type of chartering is useful for a long-term plan. On the other hand,

voyage and time charter are usually for a short-term basis. A voyage charter is the hiring of a

vessel and crew for a voyage between a load port and a discharge port. A time charter is the

hiring of a vessel for a specific period of time; the owner still manages the vessel, but the

charterer selects the ports and directs the vessel where to go. The charterer pays for all fuel

the vessel consumes, port charges, commissions, and a daily hire to the owner of the vessel.

The charterer pays the vessel owner on a per-ton or lump-sum basis. The owner pays the port

costs (excluding stevedoring), fuel costs and crew costs. The payment for the use of the

vessel is known as freight. A voyage charter specifies a period, known as laytime, for loading

and unloading the cargo. If laytime is exceeded, the charterer must pay demurrage. In this

study, voyage charted was considered to estimate the transportation cost.

Long term Chartering like bareboat charter is cheaper than the short-term Chartering like

time and voyage charter. Chartering rate fluctuates during the demand and supply. Also, the

charter rate depends on the time of hiring. If the peak demand time the charter rate is higher

than the off-peak season. See the data below for spot hiring rate. Additionally, if the hiring

process is done with the help of a broker, then the broker's fees will be added on top of the

hiring fees.

3.2.1.3 Length of shipping route (depends on the location of the source and

destination)

Transportation cost directly dependent on the travel distance. Usually, an LNG carrier

requires more fuel to travel more distance. As a result, the cost increases with the increase of

route length. In this study, a methodology has been developed to estimate the distance

between the source and the destination terminals using GPS coordinates and Commercially

available Netpas software developed by Seafuture Inc. (Koria) and Google Maps web

mapping service developed by Google Inc. (USA).

As Google Maps can show the satellite image of the earth surface, it is possible to identify the

locations and the GPS coordinates of LNG terminals. Using Google Maps web service,

Global Positioning System (GPS) Coordinates: Latitude and Longitude including satellite

images were collected. Figure 3.3(a) shows the satellite image of the Gorgon LNG plant in

Western Australia for an example. Similarly, the destination port is identified as shown in

Figure 3.3.

Figure 3.3: Source and destination terminals on Google Maps

One of the locations of the source and destination are identified, the distance between the two

terminals can be measured by using the distance measurement tool in google maps as shown

in Figure 3.4.

Figure 3.4: Distance measurement with Google Maps web service

Also, Netpas software was used to determine the exact distance between the source and

destination terminals using the same GPS data as shown in Figure 3.5. The path shown in the

figure represents the actual sea route between these terminals.

Figure 3.5: Distance measurement with Netpas software

The measurements by both methods provide similar results with less the 1% difference.

Hence, any method can be used.

3.2.1.4 Toll on the route: canal charges (e.g., Panama or Suez)

It is also possible to reduce the distance by using a shortcut way such as a manmade canal

which connects two seas via a short cut, for example, Suez and Panama Canal. Based on the

source and destination location, distance can be significantly decreased via the canal. As a

result, the delivery time can be significantly reduced. However, there are charges for the

3.2.1.5 Other charges: insurance of goods and demurrage

benefit.

To minimise the risk for the loss of goods and properties insurance is required. This cost is

also added to the total transportation. Furthermore, demurrage to be paid for any delay for the

3.2.1.6 Loading and unloading time at the port

delivery of the goods.

Development of a computational model using computer software

This is also a factor as it may increase the cost.

For the developed cost estimation model, the following parameters of the LNG carrier need

to calculate the transport cost:

• Ship speed

• Vessel size

• Chartering rate and associated costs such as brokerage fee and insurance cost

• Loading and unloading time at the ports

The delivery time can be calculated by dividing the sea route length with ship speed.

Delivery time = route length / ship speed

Total chartering cost = (chartering cost/day) × total time of delivery including loading and

unloading of LNG at the ports

To develop the computer model suing MATLAB output and output variables defined to code

was developed to calculate the required output. Additionally, a spreadsheet model using MS

3.2.2.1 Input and output variables for the model

Excel has been developed for the same outputs and verify the results by both models.

MATLAB code was developed for the calculation of the required parameters. For this

purpose, input and output variables were defined as per requirement. Table 3.1and Table 3.2

listed all the input and output variables with their description and data types and unit. Table

3.3 shows all the formulation for the output variables.

Table 3.1: Input variables

Unit

$/GJ

Name d_country d_port_name d_port_capacity s_country s_port_name s_port_capacity s_lng_price s_type s_capacity s_speed c_fee c_charge l_port u_port b_fee i_rate

Description Destination country Name of the destination port Total capacity of the destination port Source country Name of the source port Total capacity of the source port LNG price at source Ship type Ship capacity Ship speed Chartering fee Canal charge Loading at source port Unloading at destination port Brokerage Fee (2% of chartering fee) Insurance of Goods ($2600/day)

Type Text Text Number MTPA Text Text Number MTPA Number Text Number Number Number Number Number Number Number Number

cm km/h $/day $/mcm day day $ $/day

Table 3.2: Output variables

Unit km day $ $ $ $ $ $ $ $/GJ $/GJ

Name distance duration c_fee_total b_fee_total i_cost_total c_fee_total s_cost_total lng_cost_total cif_price_total s_cost_per_unit cif_per_unit trip cost_annual

Description Distance between source and destination Duration from Source to Destination Total chartering cost Total Brokerage cost Total Insurance cost Total canal cost Total shipping cost Total LNG cost per trip Total CIF price per ship Shipping cost per unit LNG CIF price per unit LNG Number of trips required Total cost for annual capacity

Type Number Number Number Number Number Number Number Number Number Number Number Number Number

Table 3.3: Formulation for output parameters

Output variable distance duration c_fee_total b_fee_total i_cost_total c_fee_total s_cost_total lng_cost_total cif_price_total s_cost_per_unit cif_per_unit trip cost_annual

method / formula using google map = distance / s_speed = (l_port + u_port + duration) x c_fee = 0.02 x c_dee_total = 2600 x (l_port + u_port + duration) = (s_capacity / 100000) x c_charge = c_fee_total + b_fee_total + i_cost_total + c_fee_total = (s_capacity x 0.456 x 51.9 x 1.054615) x s_lng_price = s_cost_total + lng_cost_total = s_cost_total / (s_capacity x 0.456 x 51.9) = s_lng_price + s_cost_per_unit = (d_port_capacity x 10^6) / (s_capacity x 0.456) = trip x cif_price_total

3.2.2.2 MATLAB model

MATLAB codes have been developed to calculate the parameters as shown in the Excel

% Inputs variables and data ----------------------------------------- d_country = 'Bangladesh';% Destination country d_port_name = 'FSRU'; % Name of the destination port d_port_capacity = 3.5; % Total capacity of the destination port s_country = 'Australia'; % Source country s_port_name = 'Gorgon LNG T1-2'; % Name of the source port s_lng_price = 7.91; % LNG price at source s_type = 'Large Conventional'; %S hip type s_capacity = 150000; % Ship capacity s_speed = 37; % Ship speed c_fee = 80000; % Chartering fee c_charge = 0.2; %Canal charge l_port = 1; % Loading at source port u_port = 2; % Unloading at destination port b_fee = 0.02; % Brokerage Fee (2% of chartering fee) i_rate = 2600; % Insurance of Goods ($2600/day) distance = 5523; % Distance between source and destination %Calculations---------------------------------------------------------- duration = distance / s_speed; % Duration from Source to Destination c_fee_total = (l_port + u_port + duration) * c_fee; b_fee_total = b_fee * c_fee_total; i_cost_total = 2600 * (l_port + u_port + duration); c_fee_total = (s_capacity / 100000) * c_charge; s_cost_total = c_fee_total + b_fee_total + i_cost_total + c_fee_total; lng_cost_total = (s_capacity * 0.456 * 51.9 * 1.054615) * s_lng_price; cif_price_total = s_cost_total + lng_cost_total; s_cost_per_unit = s_cost_total / (s_capacity * 0.456 * 51.9); cif_per_unit = s_lng_price + s_cost_per_unit; trip = (d_port_capacity * 10^6) / (s_capacity * 0.456); cost_annual = trip * cif_price_total; % Display outputs--------------------------------------------------------- disp(s_cost_per_unit); disp(cif_per_unit); disp(lng_cost_total); disp(cif_price_total); disp(trip); % End of code---------------------------------------------------------

model. are given below:

3.2.2.3 Microsoft Excel model

Microsoft Excel spreadsheet analysis software was used to develop the computational model.

Sample data were used in the Excel model as shown in Figure 3.6 to evaluate the outputs.

Destination data: (Inputs)

Country Name of the terminal Total capacity (MTPA)

Bangladesh FSRU at Moheshkhali 3.50

Source data: (from Database)

Country Name of the terminal Distance between source and destination (km) LNG price ($/GJ) at source

Australia Gorgon LNG T1-2 5,523 7.91

Shipment Cost Components: (from Database)

Large Conventional 150,000 37 $ 80,000

Type of the ship Ship capacity (cm) Ship speed (km/h) Chartering fee ($/day) Brokerage fee (% of chartering fee) Insurance rate ($/day) Canal charge ($/mmBTU)

$ 2,600 $ -

Shipping Time:

Loading time at source port (day) Unloading time at destination port (day)

Duration from source to destination (day)

1.00 2.00 7.00

Total delivery time (day)

10.00 Total Shipping Cost:

Total chartering cost Total brokerage fee ( Assume: 2% of chartering fee) Total Insurance cost Total canal fee

$ 800,000 $ 16,000 $ 26,000 $ -

Total Shipping Cost

$ 842,000

Total Cost:

Total LNG Cost per Trip Total CIF Price per Ship Number of trips required

$ 29,612,332 $ 30,454,332 51

Total cost for annual capacity

$ 1,558,335,706

Unit Cost:

Shipping cost per unit LNG ($/GJ) CIF price per unit LNG ($/GJ) at destination

$ 0.25 $ 8.16

Figure 3.6: Excel Model with sample data and calculations of outputs

This model can be used to calculate CIF price at the destination by using the appropriate any

source and destination data. The benefit of this model is its Simplicity and Easy to edit,

modify and input new data. Both the exporting and importing countries can use this model to

estimate the cost. To verify the correctness of the model calculations, MATLAB codes were

used. Details are provided in the following section. Both Excel and MATLAB model provide

the same results. Therefore, the calculations and formula are correct, and any method can be

implemented.

3.3 Development of the optimal sourcing model

If there are multiple possible sources for a given destination, quantitative and qualitative data

analysis can be performed to rank the optimal sources according to their suitability for a

given destination. A point-based ranking system has been developed by analysis of the

following factors:

• Production capacity of the source terminal

• Price at the source

• Distance between source and destination terminals

• Risk on the transportation route

• Trade dispute between source country with other nations

• Source country’s involvement in internal/external conflicts or warfare

• Bilateral relationship between source and destination country

• Trade relationship/volume of export-import between source and destination country

• Economic strength of the source country

• Any economic or UN sanctions on the source country

• Medium/long-term political risk index of the source country

• Political stability index of the source country

Methodology

The description of the above factors has been discussed in Chapter 1.

The steps are given below:

• Price and delivery time estimation for each source

• Create a database of all sources for unit price, delivery time, distance, capacity

• Shortlisting by data filtering based on capacity and delivery time

• Ranking among shortlisted countries according to level: (High, Medium and Low) for

all 12 risk factors

• Setting points for each ranking for each country according to the point criteria

• Sum of all points for all 12 risk factors for all shortlisted countries

• Ranking of the shortlisting countries based on points

Initially, using the costing model (described in the earlier section), prices are calculated for

all possible sources and an output database is created with the following data fields:

• Country

• Unit price

• Delivery time

• Distance

• Capacity

Then, sources are shortlisted based on the following criteria:

• Capacity to a minimum threshold

• Delivery time to a maximum threshold

It has been mentioned earlier that higher the capacity, energy security is also higher. Also,

the risk is minimised if the delivery time is not higher.

Country ranking performed based on risk factors by qualitative and quantitative analysis. For

example, the unit price of LNG at the destination of Source 1 > Source 2 > Source 3. Then,

the rank will be as flows:

• Source 1 = High

• Source 2 = Medium

• Source 3 = Low

Then, the point is assigned according to the point criteria for factor 1 (Production capacity):

Low = 0, Medium = 1, High = 2. Therefore, the points for source 1 for factor 1 (Production

capacity) is 2, Source 2 = 1 and Source 3 = 0.

Similarly, all the risk factors are analysed, and level of risk is assessed based on collected

data for each source by comparative data analysis and level is assigned high medium, and low

according to the rank level. Table 3.4: Risk factors matrix for points calculation shows the

tabulation of data for the risk assessment for each source for each factor and the point criteria.

Table 3.4: Risk factors matrix for points calculation

Factors

Point criteria

1. Production capacity

Low=0, Medium=1, High=2

2. Price

Low=2, Medium=1, High=0

3. Distance

Low=2, Medium=1, High=0

4. Risk on the route

Low=2, Medium=1, High=0

5. Trade dispute with other nations

6. Conflicts internal/external

Low=2, Medium=1, High=0

7. Bilateral relationship

Low=0, Medium=1, High=2

8. Trade relationship/volume- export-import

9. Economic strength

Low=0, Medium=1, High=2

Source 1 Point (F1,S1) Point (F2,S1) Point (F3,S1) Point (F4,S1) Point (F5,S1) Point (F6,S1) Point (F7,S1) Point (F8,S1) Point (F9,S1)

Source 2 Point (F1,S2) Point (F2,S2) Point (F3,S2) Point (F4,S2) Point (F5,S2) Point (F6,S2) Point (F7,S2) Point (F8,S2) Point (F9,S2)

10. UN sanctions

Low=2, Medium=1, High=0

Point (F10,S1)

Point (F10,S2)

Low=2, Medium=1, High=0

11. Medium/long-term political risk

Point (F11,S1)

Point (F11,S2)

12. Political stability index

Low=0, Medium=1, High=2

Point (F12,S1)

Point (F12,S2)

Total Points:

Sum of Point (S1)

Sum of Point (S2)

Source n Point (F1,Sn) Point (F2,Sn) Point (F3,Sn) Point (F4,Sn) Point (F5,Sn) Point (F6,Sn) Point (F7,Sn) Point (F8,Sn) Point (F9,Sn) Point (F10,S2 n) Point (F11,Sn ) Point (F12,Sn ) Sum of Point (Sn)

Finally, the sources are ranked based on the total points for all 12 factors from the height to

lowest points.

Data Collection and Processing

In this chapter, collection and processing of necessary data from external sources for the

determination of required output parameters have been presented.

4.1 Sources of research data

To understand the methodology and effectiveness of the developed optimization model for

this work, a detailed case study was implemented for a country. Historical and recently

available data from different sources have been used for the research model evaluation.

Sources are listed below:

• International Gas Union (IGU)

• International Energy Agency (IEA)

• ICC Commercial Crime Services

• Energy Global LNG Service

• Waterborne LNG

• HIS Inc.

• British Petroleum (BP)

• ExxonMobil

• Royal Dutch Shell

• BG Group

• Chevron

• PERTAMINA (Indonesia)

• Petronas (Malaysia)

• Petro Bangla (Bangladesh)

• Google maps

• National Aeronautics and Space Administration (NASA)

• World bank

4.2 Data of LNG exporting countries

Currently, there are 22 countries in the world export LNG and 3 more countries will start to

export by 2023. To evaluate the optimal sources for a given destination, all the sources have

been taken into consideration for this study. Data from different sources were collected and

matched for all the possible sources of LNG producing and exporting country around the

globe and a database has been stored with all the following details:

• Country

• Name of the project

• Address/Location of the terminal

• Operation starts year

• Number of terminals

• Terminal capacity

Also, based on the address or location of the terminal of geographic location (GPS

coordinates) on the Google Maps of the terminal has been identified and included the source

database as shown in Figure 4.1.

Figure 4.1: Sources on Google Maps

For this case study, all the current sources (in operation) including the sources under

construction have been included for this study. The operation starts year, annual capacity and

name and GPS coordinates of the terminal for all 25 countries have been presented in Table

4.1 to Table 4.25.

Table 4.1: Data for Algeria

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1978

Arzew-GL1Z (T1-6) [35.818288, -0.261095]

1981

Arzew-GL2Z (T1-6) [35.818288, -0.261095]

2013

Skikda- GL1K rebuild [36.883598, 6.942913]

2014

Arzew-GL3Z (Gassi Touil) [35.818288, -0.261095]

Grand Total

7.90 7.90 8.40 8.40 4.50 4.50 4.70 4.70 25.50

Table 4.2: Data for Angola

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2013

5.20

Angola LNG T1 [-6.114803, 12.335089]

5.20

Grand Total

5.20

Table 4.3: Data for Argentina

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2019

Tango FLNG [-38.841527, -62.217814]

Grand Total

0.50 0.50 0.50

Table 4.4: Data for Australia

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1989

5.00

North West Shelf T1-2 [-20.586644, 116.763745]

5.00

1992

2.50

North West Shelf T3 [-20.586644, 116.763745]

2.50

2004

4.60

North West Shelf T4 [-20.586644, 116.763745]

4.60

2006

3.70

Darwin LNG T1 [-12.523552, 130.849467]

3.70

2008

4.60

North West Shelf T5 [-20.586644, 116.763745]

4.60

2012

4.30

Pluto LNG T1 [-20.615481, 116.744530]

4.30

2015

12.40

GLNG T1 [ -23.778337, 151.195885]

3.90

QC LNG T1-2 [ -23.778337, 151.195885]

8.50

2016

23.30

AP LNG T1-2 [ -23.778337, 151.195885]

9.00

GLNG T2 [ -23.778337, 151.195885]

3.90

Gorgon LNG T1-2 [ -20.795315,115.478228]

10.40

2017

9.65

Gorgon LNG T3 [ -20.795315,115.478228]

5.20

Wheatstone LNG T1 [-20.392012, 115.508831]

4.45

2018

4.45

Wheatstone LNG T2 [-20.392012, 115.508831]

4.45

2019

12.50

Ichthys LNG T1-2 [-12.534357, 130.881867]

8.90

Prelude FLNG [-13.862997, 122.395900]

Grand Total

3.60 87.00

Table 4.5: Data for Brunei

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1972

Brunei LNG (T1-2) [4.710263, 114.441647] Brunei LNG (T3-4) [4.710263, 114.441647] Brunei LNG T5 [4.710263, 114.441647]

7.20 2.88 2.88 1.44

Grand Total

7.20

Table 4.6: Data for Cameroon

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2018

Cameroon FLNG [4.02026, 8.97393]

Grand Total

2.40 2.40 2.40

Table 4.7: Data for Canada

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2024

LNG Canada T1-2 [51.857472, -129.782177]

14.00 14.00

Grand Total

14.00

Table 4.8: Data for Egypt

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2005

Damietta LNG T1 [31.364016, 30.299496] Egyptian LNG (Idku) T1-2 [31.364016, 30.299496]

Grand Total

12.20 5.00 7.20 12.20

Table 4.9: Data for Equatorial Guinea

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2007

EG LNG T1 [3.782558, 8.697219]

3.70 3.70

Grand Total

3.70

Table 4.10: Data for Indonesia

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1977

Bontang LNG TC-TD [0.083339, 117.486686]

1989

Bontang LNG TE [0.083339, 117.486686]

1993

Bontang LNG TF [0.083339, 117.486686]

1997

Bontang LNG TG [0.083339, 117.486686]

1999

Bontang LNG TH [0.083339, 117.486686]

2009

Tangguh LNG T1 [-2.425299, 133.132766]

2010

Tangguh LNG T2 [-2.425299, 133.132766]

2015

Donggi-Senoro LNG T1 [-1.255771, 122.584880]

2020

Sengkang LNG T1 [-3.907664, 120.343646]

2021

Tangguh LNG T3 [-2.425299, 133.132766]

Grand Total

5.60 5.60 2.80 2.80 2.80 2.80 2.80 2.80 2.95 2.95 3.80 3.80 3.80 3.80 2.00 2.00 0.50 0.50 3.80 3.80 30.85

Table 4.11: Data for Libya

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1970

Marsa El Brega [30.414909, 19.591813]

3.20 3.20 3.20

Grand Total

Table 4.12: Data for Malaysia

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1983

MLNG Satu T1-3 [3.265060, 113.057762]

1995

MLNG Dua T4-5 [3.265060, 113.057762] MLNG Dua T6 [3.265060, 113.057762]

2003

MLNG Tiga T7-8 [3.265060, 113.057762]

2017

MLNG T9 [3.265060, 113.057762] Petronas FLNG Satu [3.265060, 113.057762]

2020

Petronas FLNG Dua [3.265060, 113.057762]

Grand Total

8.40 8.40 9.60 6.40 3.20 7.70 7.70 4.80 3.60 1.20 1.50 1.50 32.00

Table 4.13: Data for Mauritania

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2022

2.50

Tortue/Ahmeyim FLNG T1 [19.24703, -16.87444]

2.50

Grand Total

2.50

Table 4.14: Data for Mozambique

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

3.40 2022

Coral South FLNG [-26.029246, 32.761307]

3.40

12.88 2024

Mozambique LNG (Area 1) T1-2 [-26.029246, 32.761307]

12.88

Grand Total

16.28

Table 4.15: Data for Nigeria

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1999

NLNG T1-2 [4.421813, 7.141173]

2002

NLNG T3 [4.421813, 7.141173]

2005

NLNG T4 [4.421813, 7.141173]

2006

NLNG T5 [4.421813, 7.141173]

2007

NLNG T6 [4.421813, 7.141173]

6.60 6.60 3.30 3.30 4.10 4.10 4.10 4.10 4.10 4.10 8.00

2024

NLNG T7 [4.421813, 7.141173]

Grand Total

8.00 30.20

Table 4.16: Data for Norway

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2007

Snohvit LNG T1 [70.683907, 23.598731]

Grand Total

4.20 4.20 4.20

Table 4.17: Data for Oman

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2000

Oman LNG T1-2 [22.665325, 59.412303]

2006

Qalhat LNG T3 [22.665325, 59.412303]

Grand Total

7.10 7.10 3.30 3.30 10.40

Table 4.18: Data for Papua New Guinea (PNG)

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2014

PNG LNG T1 -2 [-9.339396, 146.986901]

Grand Total

6.90 6.90 6.90

Table 4.19: Data for Peru

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2010

Peru LNG T1 [-13.259154, -76.313187]

Grand Total

4.45 4.45 4.45

Table 4.20: Data for Qatar

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1996

QatarGas I (T1) [25.936393, 51.614587]

1997

QatarGas I (T2) [25.936393, 51.614587]

1998

QatarGas I (T3) [25.936393, 51.614587]

1999

RasGas I (T1) [25.936393, 51.614587]

2000

RasGas I (T2) [25.936393, 51.614587]

2004

3.20 3.20 3.20 3.20 3.20 3.20 3.30 3.30 3.30 3.30 4.70 4.70

RasGas II (T3) [25.936393, 51.614587]

2005

RasGas II (T4) [25.936393, 51.614587]

2007

RasGas II (T3) [25.936393, 51.614587]

2009

QatarGas II (T4-5) [25.936393, 51.614587] RasGas III (T6-7) [25.936393, 51.614587]

2010

RasGas III (T6) [25.936393, 51.614587]

2011

QatarGas IV T7 [25.936393, 51.614587]

Grand Total

4.70 4.70 4.70 4.70 31.20 15.60 15.60 7.80 7.80 7.80 7.80 77.10

Table 4.21: Data for Russian Federation

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2009

Sakhalin 2 T1-2 [46.618985, 142.899874]

2017

Yamal LNG T1 [71.2733, 72.0725]

2018

Yamal LNG T2 [71.2733, 72.0725]

2019

Vysotsk LNG T1 [71.2733, 72.0725] Yamal LNG T3 [71.2733, 72.0725]

2020

Portovaya LNG T1 [71.2733, 72.0725] Yamal LNG T4 [71.2733, 72.0725]

2024

Arctic LNG 2 T1 [71.2733, 72.0725]

2025

Arctic LNG 2 T2-3 [71.2733, 72.0725]

Grand Total

9.60 9.60 5.50 5.50 5.50 5.50 6.16 0.66 5.50 2.40 1.50 0.90 6.60 6.60 13.20 13.20 48.96

Table 4.22: Data for Trinidad and Tobago

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1999

Atlantic LNG T1 [10.190200, -61.699874]

2000

Atlantic LNG T2 [10.190200, -61.699874]

2003

ALNG T3 [10.190200, -61.699874]

2005

ALNG T4 [10.190200, -61.699874]

Grand Total

3.00 3.00 3.30 3.30 3.30 3.30 5.20 5.20 14.80

Table 4.23: Data for United Arab Emirates (UAE)

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

1977

ADGAS LNG T1-2 [25.157938, 52.886555]

1994

ADGAS LNG T3 [25.157938, 52.886555]

Grand Total

2.60 2.60 3.20 3.20 5.80

Table 4.24: Data for United States of America (USA)

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2016

Sabine Pass T1-2 [29.742158, -93.869731]

2017

Sabine Pass T3-4 [29.742158, -93.869731]

2018

Cove Point LNG T1 [38.402772, -76.384514] Sabine Pass T5 [29.742158, -93.869731]

2019

Cameron LNG T1 [30.036871, -93.333062] Corpus Christi T1 [27.629194, -97.145189] Corpus Christi T2 [27.629194, -97.145189] Elba Island T1-T3 [31.860173, -80.583875] Freeport LNG T1 [28.931789, -95.316123]

2020

Cameron LNG T2-3 [30.036871, -93.333062] Elba Island T4-10 [31.860173, -80.583875] Freeport LNG T2-3 [28.931789, -95.316123]

2021

Corpus Christi T3 [27.629194, -97.145189]

2023

Calcasieu Pass LNG T1-18 [29.758258, -93.337934] Sabine Pass T6 [29.742158, -93.869731]

2024

Golden Pass LNG T1-3 [29.647474, -93.850415]

Grand Total

9.00 9.00 9.00 9.00 9.75 5.25 4.50 18.85 4.00 4.50 4.50 0.75 5.10 12.75 0.80 1.75 10.2 4.50 4.50 14.50 10.00 4.50 15.60 15.60 93.95

Table 4.25: Data for Yemen

Start Year - Terminal name [GPS Coordinates]

Sum of Capacity (MTPA)

2009

Yemen LNG T1-2 [13.979995, 48.169201]

Grand Total

6.70 6.70 6.70

4.3 LNG carriers and charter rate

LNG is transported by specially designed ships called LNG carriers with different vessel size.

Table 4.26 shows the capacity and speed for different types of LNG carriers.

Table 4.26: Capacity and speed of LNG carriers

Average Capacity (m3) Average design ship speed (knot)

Ship Type Small Small (Med-max) Small Conventional Large Conventional Q-flex Q-max

19,000 75,000 138,000 150,000 210,000 265,000

15.0 17.5 19.5 20.0 20.0 20.0

Figure 4.2 shows the spot chartering rate (voyage charter type) from 2015 to 2019. Cratering

fee varies with the supply and demand model.

Figure 4.2: Spot chartering rate (voyage charter type) from 2015 to 2019 [adapted from (IGU, 2020)]

For this study, a voyage charting rate is used to calculate the transport cost. As mentioned

earlier in section 3, in voyage chartering, the charter rate includes the bunker fuel, crew and

insurance of the ship. Brokerage fee and insurance fees can be dependent on the brokerage

company and the insurance company. In this study, a brokerage fee has been calculated using

2% of the chartering fee. Insurance fee was used $2600/day.

4.4 Major LNG shipping routes and tolls

Figure 4.3 shows the major LNG shipping routes around the globe. LNG carriers can follow

any route depending on the loading and unloading port. There is no cost or toll on the

international waters, however, there is the cost involved in passing through 2 canals: Panama

and Suez. These canals are used to minimize the route length, as a result, the delivery time

will be quicker. Any LNG carrier passing through this canal must pay the toll with a specific

rate that is updated each year by the canal operators. Most of the international routes are

shown in Figure 4.3.

Figure 4.3: Major LNG shipping routes and canals [adapted from (IGU, 2020)]

There are some limitation and risk of this canals. For example, canal closure due to

maintenance and construction. Also, there is a waiting period involved that varies with traffic

condition. For this study, a canal rate 0.2/MMBtu has been used.

4.5 LNG market and wholesale LNG price

The LNG market is divided into specified zones depending on the geographic location as

shown in Figure 4.4. For this study, the global LNG market is divided into 8 zones. Table

4.27 shows the list of exporting and importing countries based on global LNG marketplace.

Figure 4.4: LNG global market map [adapted from (IGU, 2020)]

Table 4.27: Exporting and importing countries based on market

Market

Latin America

Europe

Africa

Middle East

Asia

Asia Pacific

Former Soviet Union

North America

Peru

Norway

Algeria

Russia

Oman

Australia

Canada United States America (USA)

Trinidad and Tobago

Brunei

Angola

Qatar United Arab Emirates (UAE)

Indonesia

Cameroon

Yemen

s r e t r o p x E

Malaysia Papua New Guinea (PNG)

Egypt Equatorial Guinea

Libya

Mauritania

Mozambique

Nigeria

Bangladesh

Japan

Bahrain

El Salvador

Argentina

Belgium

Ghana

China

Philippines

Israel

Mexico

Brazil

Croatia

India

Singapore

Jordan

Panama

Chile

Cyprus

Pakistan

South Korea

Kuwait

France

Taiwan

Thailand

Colombia Dominican Republic

Greece

Vietnam

Jamaica

Italy

Lithuania

s r e t r o p m

I

Netherlands

Poland

Portugal

Spain

Turkey United Kingdom

LNG prices are not the same globally. The price of an importing country depends on the

geographical location of the plant and which marketplace it belongs to. Price also depends on

gas production and liquefaction cost as well as supply and demand on the global market.

Figure xx shows the gas price data from 2010 to 2020 in different markets. In this study, the

current wholesale or spot price in 2020 has been used from this data as shown in Figure 4.5.

Figure 4.5: Wholesale LNG price data from 2010 to 2020 in different markets [adapted from (IGU, 2020)]

Model Implementation

In this chapter, a case study was performed for Bangladesh as a prospective importer of LNG.

The developed optimal sourcing model was utilised to determine the optimal sourcing

locations for Bangladesh.

5.1 Introduction

Currently, 22 countries are involved in LNG trading and more countries are expected to start

their LNG exports. On the other hand, a significant number of countries are looking forward

to importing LNG from optimal sources to meet their increasing demand. These LNG

importers must consider minimising the cost and maximising energy security. There is

currently no model available in the open literature. In this study, a model has been developed

to select an optimal sourcing strategy based on their requirement. Any country can implement

this model to select the optimal location for sourcing LNG for the current and future global

marketplace. Bangladesh has been selected for this study as a destination country. Before the

model is implemented on any destination, it is important to understand the current energy

situation and future demand, economic condition, and other factors to provide an appropriate

Current energy scenario of Bangladesh

strategy for future energy security for the country.

Natural gas is the main primary source of energy in Bangladesh. As shown in Figure 5.1,

since 1965 the gas uses increased exponentially over the years as the domestic production

also increase significantly as shown in Figure 5.2.

The electricity sector is the main consumer of the domestic production of gas as shown in

Figure 5.3. Natural gas production in Bangladesh has steadily increased by an annual average

of 6% from 2004 to 2014. However, the growth of gas production stopped since 2016 as

shown in Fig xx despite the growing demand in the power sector. To overcome this situation,

Bangladesh started to import gas from external sources to meet it continuing demand for gas.

Figure 5.1: Primary sources of energy in Bangladesh from 1965 to 2019 [adapted from (IEA, 2020)

Figure 5.2: Gas production in Bangladesh from 2000 to 2019 [adapted from (PetroBangla, 2018)]

Figure 5.3: Sector-wise gas consumption of Bangladesh in 2018 [adapted from (PetroBangla, 2018)]

Bangladesh is a developing country with rapid industrialization and infrastructure

development activities. Currently, based on purchasing power parity (PPP) Bangladesh is the

28th largest economy in the world. Bangladesh, the country of 160 million people, wishes to

become a middle-income country by 2021 and a fully developed nation by 2041. Figure 5.4

7000

) $

Bangladesh

6000

Low income

Lower middle income

5000

l a n o i t a n r e t n i 7 1 0 2

4000

3000

t n a t s n o c ( P P P

2000

1000

, a t i p a c r e p P D G

0 1985

1990

1995

2000

2010

2015

2020

2025

2005 Year

shows the rapid growth of GDP per capita since 1990.

Figure 5.4: GDP per capita for Bangladesh from 1990 to 2019 [adapted from (WB, 2020)]

To accelerate industrialization and economic growth, Bangladesh needs to increase its power

generation capacity. Due to limited energy available in Bangladesh, it needs to import raw

energy (e.g., coal and gas) from abroad to ensure emery security for a sustainable period of

time. In 2018, Bangladesh has started LNG importing LNG from the global market.

Necessary data related to LNG projects and future planning have been collected from the

recent publication of PetroBangla, a government-owned national energy company of

Bangladesh. The collected data are presented in Table 5.1. Figure 5.5 shows the photo of

FSRU at Moheshkhali, Bangladesh (PetroBangla, 2018).

Table 5.1: LNG infrastructure data for the destination: Bangladesh

Terminal

Project Name

Location

Infrastructure

Capacity

Status

Type

(MTPA)

No

Moheshkhali -

Moheshkhali,

FSRU

3.75

in operation

Excellence

Bangladesh

(Offshore)

from 2018

Excelerate

Moheshkhali -

Moheshkhali,

FSRU

3.75

in operation

Excelerate

Bangladesh

(Offshore)

from 2019

Excelerate

Moheshkhali -

Moheshkhali,

Land-based

3.75

Under future

Matarbari

Bangladesh

(Onshore)

consideration

Moheshkhali -

Moheshkhali,

Land-based

3.75

Under future

Matarbari

Bangladesh

(Onshore)

consideration

Figure 5.5: FSRU at Moheshkhali, Bangladesh [adapted from (PetroBangla, 2018)]

5.2 Implementing of the cost estimation model for Bangladesh

Initially, the cost estimation model was used to estimate the cost and other paraments. The

Distance measurement

outputs will be used to determine and rank the optimal sourcing strategy for Bangladesh.

At first, the GPS coordinates were found out using Google Maps web application as shown in

Figure 5.6 for the snapshot of the exact location of the FSRU installed in Bangladesh. This

the destination origin point from where the distance will be measured from different sources

around the globe.

Figure 5.6: Location of the FSRU installed in Bangladesh

A total of 105 terminals from 25 countries considered for the possible sourcing location for

Bangladesh. Using the methodology developed in Methodology and GPS coordinate data for

each possible LNG source from the database created in Chapter 4, the individual distance was

measured and saved all the measured distances in a database for Bangladesh. Distance

measurement and maritime route for each source can be found in Appendix A.

The measured distance was processed using IBM SPSS Statistics and Microsoft Excel

software to analyse and plot the distance wise data from each source as shown in Figure 5.7.

Sources with distance up to 6000 km are shaded with green colour whereas yellow colour

shaded data represent the medium distance (between 6000 to 16,000 km). the long-distance

sources are shaded with the red colour where the distance is more than 16,000 km from

sources to the destination: FSRU Bangladesh. The distance matrix map also shows the LNG

capacity in each source with distance. Data shows that most of the Asian counties including

some middle east countries are in close distance from Bangladesh.

Cost estimation

Figure 5.7: Distance matrix map from sources to the destination: FSRU Bangladesh

Next, the database for Bangladesh that has been created earlier section was being used to

calculate other output parameters: unit price, delivery time from each location using the cost

estimation model developed in Chapter 3 and the necessary data collected earlier in Chapter

4. Figure 5.8 shows the Microsoft Excel price estimation model as an example for sample

calculations for a source with necessary data and assumptions.

Destination data: (Inputs)

Country Name of the terminal Total capacity (MTPA)

Source data: (from Database)

Country Name of the terminal Distance between source and destination (km) LNG price ($/GJ) at source

Shipment Cost Components: (from Database)

Type of the ship Ship capacity (cm) Ship speed (km/h) Chartering fee ($/day) Brokerage fee (% of chartering fee) Insurance rate ($/day) Canal charge ($/MMBtu)

Shipping Time:

Loading time at source port (day) Unloading time at destination port (day) Duration from source to destination (day) Total delivery time (day)

Total Shipping Cost:

Total chartering cost Total brokerage fee (Assume: 2% of chartering fee) Total Insurance cost Total canal fee Total Shipping Cost

Total Cost:

Total LNG Cost per Trip Total CIF Price per Ship Number of trips required Total cost for annual capacity

Unit Cost:

Shipping cost per unit LNG ($/GJ) CIF price per unit LNG ($/GJ) at destination

Bangladesh FSRU at Moheshkhali 3.50 Qatar QatarGas III 6,090 2.11 Large Conventional 150,000 37 $ 80,000 2% $ 2,600 $ - 1.00 2.00 7.00 10.00 $ 800,000 $ 16,000 $ 26,000 $ - $ 842,000 $ 7,896,622 $ 8,738,622 51 $ 447,151,705 $ 0.25 $ 2.36

Model outputs and data processing

Figure 5.8: Cost estimation model: sample calculations with data and assumptions

All 105 sources from 25 countries were evaluated and data were stored in a spreadsheet file

for further analysis with IBM SPSS software. Figure 5.9 shows the snapshot of the processed

data: Min CIF LNG price per unit, minimum delivery time, the min shipping cost per unit

sum of capacity for each country.

Figure 5.9: Snapshot of the processed data for Bangladesh as a destination

5.3 Implementing the optimal sourcing model for Bangladesh

Among all these possible destination data and calculated parameters, optimal sources for

Bangladesh were determined and ranked by implementing the optimal sourcing model

Shortlisting by data filtering

developed in Chapter 3. risk assessment and point-based ranking

Processed outputs from the previous cost estimation model, the results further processed with

filter criteria to shortlist all possible source countries for further optimization. The filtering

criteria are given below

• Shipping duration is maximum of 2 weeks

• Capacity more than 6.5 MTPA

These conditions were applied to minimise the risk. After applying these filtering conditions,

12 countries were found as shown in Table 5.2. These shortlisted sources will be further

assessed for 12 risks factors.

Table 5.2: Shortlisted countries

Country

Price at

Shipping

Price

Min

Sum of

Source

cost per

delivery

Capacity

per

($/MMBtu)

unit

time

(MTPA)

unit

($/GJ)

(day)

16.3

3.0

0.26

Mozambique

3.42

12.2

3.0

0.28

Egypt

3.44

25.5

3.0

0.37

Algeria

3.53

32.0

7.5

0.15

Malaysia

8.06

7.2

7.5

0.15

Brunei

8.06

30.9

7.5

0.19

Indonesia

8.08

87.0

7.5

0.22

Australia

8.10

6.9

7.5

0.26

Papua New Guinea

8.17

49.0

1.8

0.51

Russia

2.18

10.4

2.0

0.17

Oman

2.28

6.7

2.0

0.19

Yemen

2.30

77.1

2.0

0.21

Qatar

2.32

Risk assessment and ranking on 12 risk factors

Now, 12 shortlisted countries will be assessed for the following 12 risk factors using a point-

5.3.2.1 Price at the source

based optimising technique as discussed in chapter 3.

According to current wholesale LNG price at the source, this assessment has been done on

the selected countries. Data were classified into low, medium and high for each country by a

comparative data analysis among the selected 12 countries. Table 5.3 shows the values for

12 countries based on the CIF price per unit.

Table 5.3: Price comparison

Rank

Country

Price ($/MMBtu)

High

Russia

1.8

High

Qatar

2.0

High

Oman

2.0

High

Yemen

2.0

Medium

Algeria

3.0

Medium Mozambique

3.0

Medium

Egypt

3.0

Low

Australia

7.5

Low

Malaysia

7.5

Low

Indonesia

7.5

Low

Brunei

7.5

Low

Papua New Guinea

7.5

5.3.2.2 Distance and delivery time

This assessment has been carried out for the selected 12 countries based on the measured

distance from the destination and the calculated delivery time using the developed model in

this study. Data were classified into low, medium and high for each country by a comparative

data analysis among the selected 12 countries. Table 5.4. shows the values for 12 countries

based on distance and delivery time.

Table 5.4: Distance and delivery time comparison

Rank

Country

Min Distance (km) Min delivery time (day)

Algeria

High

12282

Russia

High

9463

Egypt

High

9315

Mozambique

High

8500

Papua New Guinea

High

7794

Medium Qatar

6090

Medium Yemen

5955

Medium Australia

5432

Low

Oman

5106

Low

Indonesia

4900

Low

Malaysia

3737

Low

Brunei

3926

5.3.2.3 Capacity

This assessment has been carried out for the selected 12 countries based on the total capacity

of the source country. Data were classified into low, medium and high for each country by a

comparative data analysis among the selected 12 countries. Table 5.5 shows the values for

12 countries based on annual capacity.

Table 5.5: Capacity comparison

Rank

Country

Total Capacity (MTPA)

High

Australia

87.0

High

Qatar

77.1

High

Russia

49.0

High

Malaysia

32.0

High

Indonesia

30.9

Medium Algeria

25.5

Medium Mozambique

16.3

Medium Egypt

12.2

Medium Oman

10.4

Low

Brunei

7.2

Low

Papua New Guinea

6.9

Low

Yemen

6.7

5.3.2.4 Risk on the route

Figure 5.10 shows the travel routes of the selected 12 countries in the google maps. it shows

that Algeria and Egypt need to pass through the Suez Canal whereas Qatar needs to travel

through the Strait of Hormuz.

According to current and historical data, this assessment has been done on the selected

countries, a risk assessment has been evaluated for this factor for each country by a

comparative data analysis among the selected 12 countries. Table 5.6. shows the evaluated

risk classification for 12 countries based on the data.

Figure 5.10: Shipping routes from 12 sources in google maps

Table 5.6: Risk on the route comparison

Risk on the route

Country

Qatar

High

Egypt

High

Algeria

High

Yemen

Medium

Australia

Low

Brunei

Low

Indonesia

Low

Malaysia

Low

Papua New Guinea

Low

Oman

Low

Mozambique

Low

Russia

Low

5.3.2.5 Trade dispute with other nations

According to current and historical data, this assessment has been done on the selected

countries, a risk assessment has been evaluated for this factor for each country by a

comparative data analysis among the selected 12 countries. Table 5.7 shows the evaluated

risk classification for 12 countries based on the data.

Table 5.7: Risk comparison of the trade dispute with other nations

Country

The trade dispute with other nations

Yemen

High

Qatar

Medium

Egypt

Medium

Russia

Medium

Algeria

Low

Australia

Low

Brunei

Low

Indonesia

Low

Malaysia

Low

Papua New Guinea

Low

Oman

Low

Mozambique

Low

5.3.2.6 Conflicts internal/external

According to current and historical data, this assessment has been done on the selected

countries, a risk assessment has been evaluated for this factor for each country by a

comparative data analysis among the selected 12 countries. Table 5.8 shows the evaluated

risk classification for 12 countries based on the data.

Table 5.8: Risk comparison based on conflicts internal/external

Conflicts internal/external

Country

Yemen

High

Egypt

High

Qatar

Medium

Russia

Medium

Algeria

Low

Australia

Low

Brunei

Low

Indonesia

Low

Malaysia

Low

Papua New Guinea

Low

Oman

Low

Mozambique

Low

5.3.2.7 Bilateral relationship

According to current and historical data, this assessment has been done on the selected

countries, a risk assessment has been evaluated for this factor for each country by a

comparative data analysis among selected 12 countries. Table 5.9 shows the evaluated risk

classification for 12 countries based on the data.

Table 5.9: Bilateral relationship comparison

Country

Bilateral relationship

High

Yemen

High

Algeria

High

Australia

High

Brunei

High

Indonesia

High

Malaysia

High

Oman

High

Qatar

High

Russia

Egypt

Medium

Mozambique

Medium

Papua New Guinea

Low

5.3.2.8 Trade relationship/volume-export-import

Figure 5.11and Figure 5.12 show the total import and export growth by market from 2013 to

2018 respectively. Based on the data, a risk assessment has been evaluated for this factor for

each country by a comparative data analysis among the selected 12 countries. Table 5.10

shows the evaluated risk classification for 12 countries.

Figure 5.11: Total import growth by market from 2013 to 2018 (WB, 2020)

Figure 5.12: Total export growth by market from 2013 to 2018 (WB, 2020)

Table 5.10: Volume-export-import comparison

volume-export-import

Country

Australia

High

Malaysia

High

Oman

High

Russia

High

Algeria

Medium

Indonesia

Medium

Yemen

Low

Brunei

Low

Qatar

Low

Papua New Guinea

Low

Egypt

Low

Mozambique

Low

5.3.2.9 Economic strength

Figure 5.13 shows the GDP per capita, constant 2010 dollars from 2012 to 2017 for 12

selected countries. Based on the data, a risk assessment has been evaluated for this factor for

each country by a comparative data analysis among the selected 12 countries. Table 5.11

shows the evaluated risk classification for 12 countries based on the data.

Figure 5.13: GDP per capita, constant 2010 dollars from 2012 to 2017

Table 5.11: Economic strength comparison

Economic strength

Country

Australia

High

Brunei

High

Qatar

High

Malaysia

Medium

Oman

Medium

Russia

Medium

Yemen

Low

Papua New Guinea

Low

Egypt

Low

Mozambique

Low

Algeria

Low

Indonesia

Low

5.3.2.10 UN sanctions

According to current and historical data, this assessment has been done on the selected

countries, a risk assessment has been evaluated for this factor for each country by a

comparative data analysis among the selected 12 countries. Table 5.12 shows the evaluated

risk classification for 12 countries based on the data.

UN sanctions

Country

High

Yemen

High

Egypt

Medium

Russia

Australia

Low

Brunei

Low

Qatar

Low

Papua New Guinea

Low

Mozambique

Low

Algeria

Low

Indonesia

Low

Malaysia

Low

Oman

Low

Table 5.12: UN sanctions risk comparison

5.3.2.11 Medium/long-term political risk

Figure 5.14 shows the medium/long-term political risk from 2014 to 2017 for 12 selected

countries. Based on the data, a risk assessment has been evaluated for this factor for each

country by a comparative data analysis among the selected 12 countries. Table 5.13 shows

the evaluated risk classification for 12 countries based on the data.

Figure 5.14: Medium/long-term political risk (1=low, 7=high)

Table 5.13: Medium/long-term political risk comparison

Medium/long-term political risk

Country

Yemen

High

Egypt

High

Papua New Guinea

High

Mozambique

High

Qatar

Medium

Indonesia

Medium

Malaysia

Medium

Russia

Medium

Australia

Low

Brunei

Low

Algeria

Low

Oman

Low

5.3.2.12

Political stability index

Figure 5.15 shows the Political stability index from 2009 to 2017 for 12 selected countries.

Based on the data, a risk assessment has been evaluated for this factor for each country by a

comparative data analysis among the selected 12 countries. Table 5.14 shows the evaluated

risk classification for 12 countries based on the data.

Figure 5.15: Political stability index (-2.5 weak; 2.5 strong)

Table 5.14: Political stability index comparison

Political stability index

Country

Australia

High

Brunei

High

Oman

High

Qatar

High

Papua New Guinea

Medium

Mozambique

Medium

Indonesia

Medium

Malaysia

Medium

Russia

Medium

Yemen

Low

Egypt

Low

Algeria

Low

5.3.2.13

Summary table

Table 5.15 shows all the risk assessment values and Table 5.16 shows all the respective

points and the total points for each country.

Table 5.15: Risk assessment values for each country for each factor

5.3.2.14

Final ranking

Table 5.16: Points for each risk factor and the total points for each country

Final ranking of the optimal sources for Bangladesh is presented in Table 5.17. Oman strands first place in the ranking tables with 22 points, followed by Australia with 21 points and then Malaysia.

Table 5.17: Optimal sources for Bangladesh with ranking

Total Points

Rank

Country

Oman

Australia

Malaysia

Brunei

Indonesia

Qatar

Russia

Egypt

Yemen

Papua New Guinea

Mozambique

Algeria

Results and Discussion

This chapter presents the results from this study on important parameters with their effects

with detailed analysis and discussion.

6.1 The dependency of shipping cost on distance and ship

capacity

To understand the dependency of shipping cost on distance, ship capacity, a case study was

carried out with selecting 5 sources with a range of shipping distance. The cost of shipping

and delivery time was calculated using the developed price estimation model for this study.

Effect of distance on shipping cost

Results are presented and discussed in the following sections:

Figure 6.1 shows the plots for shipping cost per unit for 6 different types of the ship for 5

$3.00

$2.50

y = 0.0002x + 0.1966 R² = 1

$2.00

) J G / $ ( t i n u r e p

Small Small (Med-max) Small Conventional Large Conventional Q-flex Q-max

$1.50

t s o c

$1.00

g n i p p i h S

$0.50

y = 5E-05x + 0.0992 y = 3E-05x + 0.0828 y = 3E-05x + 0.0841 y = 2E-05x + 0.0781 y = 2E-05x + 0.0756

2000

4000

6000

8000

10000

12000

14000

16000

Distance (km)

sourcing destinations located over a range of distance.

Figure 6.1: Shipping cost per unit ($/GJ) as a function of distance.

It is clear for Figure 6.1 that the shipping cost increases with the increase of distance for all

ship types. Additionally, the shipping cost decreases with the increase in ship size. Smaller

capacity can increase the shipping cost significantly. Furthermore, as the ship size increases

the rate of shipping cost increment decreases with increasing distance. Therefore, it can be

concluded that the larger ship is more economical for LNG transportation and the cost of

transportation is minimal if the selected sourcing terminal is closer from the destination.

Shipping cost for different ship sizes can be calculated using the following equations derived

from the data presented in Figure 6.1:

• For small ship: y = 0.0002 × d + 0.1966 -------------------------------------------------------- (1)

• For Med-Max ship: y = 5 × 10-05 × d + 0.0992 ------------------------------------------------ (2)

• For small conventional ship: y = 3 × 10-05 × d + 0.0828 -------------------------------------- (3)

• For large conventional ship: y = 3 × 10-05 × d + 0.0841 --------------------------------------- (4)

• For Q-flex ship: y = 2 × 10-05 × d + 0.0781 ----------------------------------------------------- (5)

• For Q-max ship: y = 2 × 10-05 × d + 0.0756 ----------------------------------------------------- (6)

Where, y is shipping cost per unit ($/GJ) and d is distance (km) form the source to the

Effect of ship capacity on shipping cost

destination.

Figure 6.2 shows the plots for shipping cost per unit as a function of ship capacity. It can be

seen that the shipping cost significantly drops with increasing ship capacity. There is a

$3.00

$2.50

$2.00

$1.50

) J G / $ ( t i n u r e p

$1.00

y = 6578.5x-0.799 R² = 0.9832

t s o c

$0.50

g n i p p i h S

50,000 100,000 150,000 200,000 250,000 300,000

Ship capacity (m3)

negative power correlation exits between unit shipping cost and ship capacity.

Figure 6.2: Shipping cost per unit ($/GJ) as a function of ship capacity.

Shipping cost can be calculated using the following equation derived from the data presented

in Figure 6.2:

• Shipping cost per unit ($/GJ): y = 578.5 × c-0.799 ----------------------------------------------- (7)

Where, c is the ship capacity in m3.

Effect of delivery time on ship speed

Figure 6.3 shows the plots for delivery time as a function of ship speed. It can be seen that the

delivery time has a negative linear correlation with ship speed. It means delivery time

decreases with the increase of ship speed. Collected ship properties data indicates that larger

ships like Q-max with an average capacity of 265,000 m3 have much 34% higher speed than

the smaller ships with an average capacity of 19,000 m3. As a result, the cost of shipping per

y = 0.0068x2 - 0.6672x + 21.535 R² = 1

) y a d ( e m

i t

y r e v i l e D

Ship speed (km/h)

unit will be significantly reduced.

Figure 6.3: Delivery time as a function of ship speed

Delivery time can be calculated using the following equation derived from the data presented

in Figure 6.3:

• Delivery time (day): t = 0.0068 × v2 – 0.6672 × v + 21.535 ---------------------------------- (8)

Where, v is the ship speed in km/h.

6.2 Effect of the canal on unit price and delivery time

As mentioned earlier, travel distance can be reduced significantly by using the sea route via

the canal. When a location of the LNG source needs has two way to travel to the destination,

either by using the international sea route without paying any route toll or by using the canal

by paying the toll. to understand the effect of the canal on the transportation cost and delivery

time, a case study was performed by calculating the output parameters for using two distance.

The measurements were taken for both the Suez Canal and without canal from Algeria to

Bangladesh. Price estimation model was used to calculate the output parameters: unit price

and delivery time suing the same shipping data. Figure 6.4 shows the calculation and

assumptions data for the excel model for both ways: with the canal and without the canal.

a. Calculations with canal b. Calculation with canal

Figure 6.4: Price calculations for Algeria

Results show that the canal reduced around 8000 km of distance (around 40%). However,

delivery time increased significantly from 14 days to 23 (around 35%). The price is slightly

cheaper of the reduction of travel time; however, the price is subjected to change due to the

toll rate for the canal.

Case study results and discussion

The developed strategic model was implemented to find out and rank optimal sources for Bangladesh. The results are tabulated in

Table 6.1 and presented in the global map in Figure 6.5.

Table 6.1: Optimal destinations for Bangladesh with the country ranking

Market

Source Country

Total Points Rank

Middle East

Oman

Asia Pacific

Australia

Asia Pacific

Malaysia

Asia Pacific

Brunei

Asia Pacific

Indonesia

Middle East

Qatar

FSU

Russia

Africa

Egypt

Middle East

Yemen

Asia Pacific

Papua New Guinea

Africa

Mozambique

Africa

Algeria

Figure 6.5: Optimal sources of LNG for Bangladesh

Results from the optimization model suggest that most of the countries from Asia pacific and

middle east suppliers of LNG are optimal locations for Bangladesh. Especially, in the Middle

East, the price is much lower than the Asia Pacific market. Also, the distance is not so far for

Asia pacific countries. However, the countries from the Asia pacific reason are politically

more stable than the middle east countries as there are some security issues are the concern.

Qatar is a suitable selection, however, recently there are some political tensions between

Qatar and its neighbours which is a little concern for the future risk. On the other hand,

Australia is much stable in term of political stability. Some locations or terminal of Australia

are comparatively far away from Bangladesh and also the price is much higher. However,

some terminals located in the western part of Australia is suitable as the distance is much

lower than the terminal of the eastern part. Apart from the larger production of Australia

makes it a suitable sourcing location for Bangladesh. Some African countries are also suitable

sourcing location for Bangladesh and the price is lower. However, it may provide less

security.

According to the annual report of PetroBangla, Bangladesh government started importing

LNG in 2018 (PetroBangla, 2018). Bangladesh has selected Oman and Qatar as sourcing

countries and signed a long-term contract with them. Bangladesh also signed a contract with

Indonesia recently. The results from the developed model in this study also show those three

countries are the optimal countries for Bangladesh. The results from this study matched well

with the Bangladesh government policy for selecting the souring countries for their future

LNG supply for long term basis. The effectiveness of the model is confirmed with the

matching of data for Bangladesh. It is therefore expected that the model can be used for any

other country who wants to select optimal sourcing countries for their future LNG demand by

ensuring energy security.

In future, Bangladesh needs more LNG for its energy security and planned to build

infrastructure for importing more LNG for external sources. This model can be more

beneficial for Bangladesh to select more reliable LNG trading partners for its future energy

security.

Conclusions and recommendations

In this chapter, major conclusions for this research are outlined and recommendations are

provided for further study. tools of strategic formulation

7.1 Conclusions

The main objective of this research was to develop a model for the optimal sourcing strategy

of LNG by not only optimising the cost but also ensure energy security for a given

General Conclusions

destination. The following general and specific conclusions stem from this research.

A computational model has been developed to estimate the shipping cost between the source

and destination terminals. This model can be used to estimate the CIF (Cost, Insurance and

Freight) price of LNG from one destination to another including all the cost components of

shipping. Several input data used in this model have been assumed, however, these values

can be easily updated with more appropriate values to determine more realistic and reliable

output and the model can be easily configured for any destination by updating the input data.

Additionally, a strategic tool: optimal sourcing model has been developed to determine the

optimal sourcing countries according to their level of stability. A total of 12 risk factors have

been taken into consideration for this optimization technique. This model can provide a list of

best possible sourcing by analysing the risk and benefit on each of the 12 factors and rank

them accordingly.

The optimal sourcing model was implemented in Bangladesh as a prospective LNG

importing country. Results from the model have provided a list of suitable sources. The

results from this study matched well with the Bangladesh government policy for selecting the

souring countries for their future LNG supply for long term basis.

The effectiveness of the model is confirmed with the matching of data for Bangladesh. It is

therefore expected that the model can be used for any other country who wants to select

optimal sourcing countries for their future LNG demand by ensuring energy security.

Although the model has been developed for LNG importing country for selecting the best

possible sourcing country, this model can also be used to find the prospective future buyer of

Specific Conclusions

LNG.

Data analysis has been performed on the future LNG demand using the current and future

operations of LNG infrastructures for exporting and importing country with their annual

capacity. Results indicates a rapid rise in this energy sector as a significant number of

countries has started switching from coal to gas to meet the global carbon emission target by

2050.

In this study, mathematical expressions for the determination of shipping cost as a function of

distance from the source to destination for different types of ships have been developed. Also,

a mathematical equation has been developed to calculate the shipping cost as a function of

the capacity of the ship. Additionally, an empirical equation was developed to calculate the

delivery time as a function of ship speed.

Types of ship used have a significant impact on the shipping cost. Results confirmed that

proper selection of ship for their capacity can minimize the cost significantly. Larger capacity

ship with high speed can be more economical and can reduce the shipping cost up to 40%

compared to the smaller ship with relatively lower speed.

Using of the canal can be useful to transport LNG for some countries based on their

geographical locations. Canals such as Panama and sues can be cut the delivery time

significantly. Results confirm that the uses of Suez Canal to transport LNG from Algeria to

Bangladesh can reduce the delivery time around 35%. However, there are cost, and risk

involved to travel through the channel.

7.2 Recommendation for future study

After conducting the work presented here and reviewing the literature available in the public

domain, the following suggestions are made for further study:

a. More case studies on different countries can be performed to understand more of the

effectiveness of the developed model for this study.

b. Implementing the model for an exporting country to find an optimal buyer can be

done.

c. More risk factors can be added in the optimal coursing model to improve it's for

providing better security

d. Only 3-level of risk (high, medium and low) have been defined for point-based risk

analysis for this developed model in this study. However, more levels of risk can be

defined for the further improvement of the model.

e. The model can also be used for optimising the sourcing strategy for other energy

imports such as oil and coal by updating the input database based on the specific types

of commodity.

References

Aydin, N., Lee, H., & Mansouri, S. A. (2017). Speed optimization and bunkering in liner

shipping in the presence of uncertain service times and time windows at ports.

European Journal of Operational Research, 259(1), 143-154.

Berle, Ø., Norstad, I., & Asbjørnslett, B. E. (2013). Optimization, risk assessment and

resilience in LNG transportation systems. Supply Chain Management: An

International Journal.

BP. (2019). Statistical Review of World Energy. Retrieved from

https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/energy-

economics/statistical-review/bp-stats-review-2019-full-report.pdf

Cho, J., Lim, G. J., Kim, S. J., & Biobaku, T. (2018). Liquefied natural gas inventory routing

problem under uncertain weather conditions. International Journal of Production

Economics, 204, 18-29.

Christiansen, M., Fagerholt, K., Nygreen, B., & Ronen, D. (2013). Ship routing and

scheduling in the new millennium. European Journal of Operational Research,

228(3), 467-483. doi:10.1016/j.ejor.2012.12.002

Cobb, J. (2020). Highlights of the World Nuclear Performance Report 2019. Age, 20, 0.

Council, W. E. (2019). World Energy Scenarios 2019. Word Energy Council: London, UK, 1.

Cunningham, M., Van Uffelen, L., & Chambers, M. (2019). The Changing Global Market for

Australian Coal. Reserve Bank of Australia Bulletin.

Dobrota, Đ., Lalić, B., & Komar, I. (2013). Problem of boil-off in LNG supply chain.

Transactions on maritime science, 2(02), 91-100.

Geng, J.-B., Ji, Q., Fan, Y., & Shaikh, F. (2017). Optimal LNG importation portfolio

considering multiple risk factors. Journal of Cleaner Production, 151, 452-464.

doi:10.1016/j.jclepro.2017.03.053

Hennig, F., Nygreen, B., Furman, K. C., Song, J., & Kocis, G. R. (2017). Crude Oil Tanker

Routing and Scheduling. INFOR: Information Systems and Operational Research,

49(2), 153-170. doi:10.3138/infor.49.2.153

IEA. (2019). World Energy Outlook 2019. Retrieved from https://www.iea.org/reports/world-

energy-outlook-2019

IEA. (2020). Global Energy Review 2020. Retrieved from https://www.iea.org/reports/global-

energy-review-2020

IGU. (2015). IGU World LNG Report 2015 Edition. Retrieved from

IGU. (2020). World LNG report. Retrieved from https://www.igu.org/app/uploads-

wp/2020/04/2020-World-LNG-Report.pdf

IRENA. (2019). Renewable power generation costs in 2018. Report. International Renewable

Energy Agency, Abu Dhabi.

Kim, H.-J., Chang, Y.-T., Kim, K.-T., & Kim, H.-J. (2012). An epsilon-optimal algorithm

considering greenhouse gas emissions for the management of a ship’s bunker fuel.

Transportation Research Part D: Transport and Environment, 17(2), 97-103.

Kontovas, C., & Psaraftis, H. N. (2011). Reduction of emissions along the maritime

intermodal container chain: operational models and policies. Maritime Policy &

Management, 38(4), 451-469. doi:10.1080/03088839.2011.588262

Kuo, J., Campbell, R., Ding, Z., Hoie, S., Rinehart, A., Sandström, R., . . . Danaczko, M.

(2009). LNG tank sloshing assessment methodology-the new generation. Paper

presented at the The Nineteenth International Offshore and Polar Engineering

Conference.

Lindstad, H., Asbjørnslett, B. E., & Jullumstrø, E. (2013). Assessment of profit, cost and

emissions by varying speed as a function of sea conditions and freight market.

Transportation Research Part D: Transport and Environment, 19, 5-12.

doi:10.1016/j.trd.2012.11.001

Loganathan, B. (2018). Aerodynamic study of single stage multi-blade drag-based vertical

axis wind turbines.

Meng, Q., & Wang, S. (2011). Optimal operating strategy for a long-haul liner service route.

European Journal of Operational Research, 215(1), 105-114.

doi:10.1016/j.ejor.2011.05.057

Mustary, I., Chowdhury, H., Loganathan, B., & Alam, F. (2017). Development of a

Computational Model for Optimal Sourcing of LNG. Energy Procedia, 110, 597-603.

doi:10.1016/j.egypro.2017.03.191

Norstad, I., Fagerholt, K., & Laporte, G. (2011). Tramp ship routing and scheduling with

speed optimization. Transportation Research Part C: Emerging Technologies, 19(5),

853-865.

Palti-Guzman, L. (2018). The Future of Asia's Natural Gas Market: The Need for a Regional

LNG Hub. asia policy, 25(3), 101-126.

Perera, F. (2018). Pollution from fossil-fuel combustion is the leading environmental threat to

global pediatric health and equity: solutions exist. International journal of

environmental research and public health, 15(1), 16.

PetroBangla. (2018). Annual Report 2018. Retrieved from

http://rpgcl.portal.gov.bd/sites/default/files/files/rpgcl.portal.gov.bd/annual_reports/51

09918d_2074_4e02_867a_9c6ade85494e/AGM.pdf

Rakke, J. G., Stålhane, M., Moe, C. R., Christiansen, M., Andersson, H., Fagerholt, K., &

Norstad, I. (2011). A rolling horizon heuristic for creating a liquefied natural gas

annual delivery program. Transportation Research Part C: Emerging Technologies,

19(5), 896-911. doi:10.1016/j.trc.2010.09.006

REN21, R. (2018). Global Status Report, REN21 Secretariat, Paris. Retrieved from

Rogelj, J., Den Elzen, M., Höhne, N., Fransen, T., Fekete, H., Winkler, H., . . . Meinshausen,

M. (2016). Paris Agreement climate proposals need a boost to keep warming well

below 2 C. Nature, 534(7609), 631-639.

Ronen, D. (2017). The effect of oil price on containership speed and fleet size. Journal of the

Operational Research Society, 62(1), 211-216. doi:10.1057/jors.2009.169

U.S. Energy Information Administration, A. E. O. (2019). Levelized cost and levelized

avoided cost of new generation resources in the annual energy outlook 2020.

Washington, DC.

WB. (2020). World Development Report 2020. Retrieved from

https://www.worldbank.org/en/publication/wdr2020

Windeck, V., & Stadtler, H. (2011). A liner shipping network design–routing and scheduling

impacted by environmental influences. Paper presented at the International

Conference on Network Optimization.

Yao, Z., Ng, S. H., & Lee, L. H. (2012). A study on bunker fuel management for the shipping

liner services. Computers & Operations Research, 39(5), 1160-1172.

doi:10.1016/j.cor.2011.07.012

Bibliography

Anand, P., & Barkan, C. P. L. (2006). A geographic information system analysis of the

exposure of soil and groundwater to spills of hazardous materials transported by

rail. Transport. Res. Rec.: J. Transport. Res. Board, 12-19.

Anand, P. O. O. J. A. (2006). Cost-effectiveness of reducing environmental risk from railroad

tank car transportation of hazardous materials. University of Illinois at Urbana-

Champaign, Urbana, IL.

Barkan, C. P., Glickman, T. S., & Harvey, A. E. (1991). Benefit-cost evaluation of using

different specification tank cars to reduce the risk of transporting environmentally

sensitive chemicals. Transportation Research Record, (1313).

Biresselioglu, M. E., Demir, M. H., & Kandemir, C. (2012). Modeling Turkey’s future LNG

supply security strategy. Energy Policy, 46, 144-152.

BP (2014), BP Statistical Review of World Energy, June 2014, BP plc, London.

Christiansen, M., Fagerholt, K., Nygreen, B., & Ronen, D. (2013). Ship routing and

scheduling in the new millennium. European Journal of Operational Research, 228(3),

467-483.

Cranmore, R. G., & Stanton, E. (2000). Natural gas. Modern petroleum technology, upstream

volume. Chichester: Institute of Petroleum, John Wiley and Sons Ltd, 337-82.

Department of Industry (2014), ‘Gas Market Report’’, November.

Dinwoodie, J., Tuck, S., & Rigot-Müller, P. (2013). Maritime oil freight flows to 2050:

Delphi perceptions of maritime specialists. Energy Policy, 63, 553-561.

Erkut, E., Tjandra, S. A., & Verter, V. (2007). Hazardous materials

transportation. Handbooks in operations research and management science, 14, 539-

621.

EIA (Energy Information Administration) (2013), International Energy Outlook 2013, US

Department of Energy, Washington, DC.

EIA (2014), Annual Energy Outlook 2014: With Projections to 2040, US Department of

Energy, Washington, DC.

IEA (International Energy Agency) (2013), ‘Developing a Natural Gas Trading Hub in

Asia’’.

IEA (2014), Medium-term Gas Market Report 2014, OECD/IEA, Paris.

Halvorsen-Weare, E. E., & Fagerholt, K. (2013). Routing and scheduling in a liquefied

natural gas shipping problem with inventory and berth constraints. Annals of

Operations Research, 203(1), 167-186.

International Energy Agency (IEA), Natural Gas Information, IEA Publications, Paris, 2011.

IGU (2015) World LNG Report. IGU Publications, Paris, 2015.

Jacobs D (2011), ‘The Global Market for Liquefied Natural Gas’, Bulletin, September, pp

17–27.

Kumar, S., Kwon, H. T., Choi, K. H., Cho, J. H., Lim, W., & Moon, I. (2011). Current status

and future projections of LNG demand and supplies: A global prospective. Energy

Policy, 39(7), 4097-4104.

Lin, D. Y., & Tsai, Y. Y. (2014). The ship routing and freight assignment problem for daily

frequency operation of maritime liner shipping. Transportation Research Part E:

Logistics and Transportation Review, 67, 52-70.

List, G. F., Mirchandani, P. B., Turnquist, M. A., & Zografos, K. G. (1991). Modeling and

analysis for hazardous materials transportation: Risk analysis, routing/scheduling and

facility location. Transportation Science, 25(2), 100-114.

Maxwell, D., & Zhu, Z. (2011). Natural gas prices, LNG transport costs, and the dynamics of

LNG imports. Energy Economics, 33(2), 217-226.

Ruester, S., & Neumann, A. (2006, September). Economics of the LNG Value Chain and

Corporate Strategies - An Empirical Analysis of the Determinants of Vertical

Integration. In Paper presented at the 10th Annual ISNIE Conference.

Saat, M. R., Werth, C. J., Schaeffer, D., Yoon, H., & Barkan, C. P. (2014). Environmental transportation. Journal of hazardous risk analysis of hazardous material rail materials, 264, 560-569.

Woodward, J. L., & Pitbaldo, R. (2010). LNG Risk Based Safety: modeling and consequence

analysis. John Wiley & Sons.

Yoon, H., Werth, C. J., Barkan, C. P., Schaeffer, D. J., & Anand, P. (2009). An

environmental screening model to assess the consequences to soil and groundwater

from railroad-tank-car spills of light non-aqueous phase liquids. Journal of Hazardous

Materials, 165(1), 332-344.