Ocean Engineering 122 (2016) 84–96

Contents lists available at ScienceDirect

Ocean Engineering
journal homepage: www.elsevier.com/locate/oceaneng

Feasibility and commercial considerations of LNG-fueled ships

O. Schinas b,n, M. Butler a
a
INTECSEA B.V., Poortweg 6, 2612 PA Delft, The Netherlands
b
Hamburg School of Business Administration, Maritime Business School, Alter Wall 38, 20457 Hamburg, Germany

art ic l e i nf o a b s t r a c t

Article history: Regulatory initiatives obligate maritime operators to reduce emission via retrofitting propulsion and
Received 17 January 2016 exhaust treatment systems or by using alternative fuels. Although these initiatives facilitate the emer-
Received in revised form gence of LNG as a marine fuel, the proliferation of LNG technology is dependent on the relationship that
14 April 2016
defines the competitiveness of conventionally fueled and LNG-fueled ships.
Accepted 24 April 2016
This article proposes a methodology to evaluate the commercial incentives that are required to
promote LNG as a marine fuel, which can be used to assess policy initiatives that encourage the use of
Keywords: alternative technologies and estimate their market impact. To develop the context and support this
Air emissions methodology, the following areas are discussed: (1) international regulatory frameworks and regional
Maritime transportation
initiatives supporting the use of LNG as a marine fuel; (2) commercial and operational attributes that
LNG
determine the competitiveness of LNG-fueled ships; and (3) challenges preventing the adoption of LNG
MARPOL Annex VI
as a marine fuel, including the uncertainties and less-quantifiable concerns expressed by the maritime
operators.
& 2016 Elsevier Ltd. All rights reserved.

1. Introduction Since that time, substantial work from funded research on air
emissions has been presented throughout academic sources – such
The interest from research and business communities, regard- as the contributions of Miola and Ciuffo (2011); Miola et al. (2011)
ing the viability of liquefied natural gas (LNG) as a marine fuel, and Buhaug et al. (2009); Faber et al. (2009) in the field of air
continues to increase, particularly when the price of conventional emissions regulations, which sparked further research and pub-
marine fuels (bunkers) surges, freight rates decline, or the regional lications related to alternative fuels. The work of Corbett et al.
price of natural gas and, subsequently, that of LNG, decrease. Fig. 1 (2007) summarized the negative impacts of ship emissions that
depicts this clearly. During the boom years of 2004–2008, market generated numerous policy initiatives.
indices, such as Clark Sea and the Baltic Dry Index (BDI) surged, The technology itself has been used in marine fuel applications
conventional bunker prices (HFO 380cSt prices in Rotterdam are for steam turbines and dual-fueled diesel engines since the early
indicatively considered) were at lower levels, and LNG fuel was an 1970s – i.e. LNG is stored in on-board cryogenic tanks, but because
the tank volume is constant and the insulation is not flawless,
afterthought. In contrast, the freight market recession at the end of
ambient heat creates boiled-off LNG vapors (venting), which can
2008 – where the BDI experienced a drop of almost 90% – and the
be used by the propulsion system. However, sufficient interest to
increase in bunker prices made conventionally fueled ships un-
generate widespread investment has been elusive.
profitable and LNG fuel a consideration. However, despite the
For maritime operators, resistance to LNG-fueled ships is re-
substantial drop of bunker prices in 2015, the attractiveness of
lated to the following areas of concern:
technical solutions involving commercially viable alternative fuels
remains due to regulatory requirements. 1. cost of bunkers;
Aside from price, the increasing interest in LNG as a marine fuel 2. changes in the employment of the ships; and
can largely be attributed to regulatory initiatives that focus on the 3. impact on all other operations.
abatement of air pollution from ships and dictate the consumption
of low-emission fuels. As early as 1997, the environmental issues of Assuming that regulatory issues related to safety, security, and
using LNG as marine fuel were considered (see e.g. Cockett, 1997). occupational hazards are resolved (see e.g. International Maritime
Organisation, 2014), these areas are either promoted as excuses to
n
Corresponding author.
avoid investment or remain obscured to potential investors and
E-mail addresses: [email protected] (O. Schinas), operators. Besides, using LNG as a marine fuel does not address all
[email protected] (M. Butler). problems for the range of operating scenarios and environments,

http://dx.doi.org/10.1016/j.oceaneng.2016.04.031
0029-8018/& 2016 Elsevier Ltd. All rights reserved.
 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96 85

Fig. 1. Evolution of market indices and conventional bunker prices.

nor is LNG the panacea for resolving air emissions issues, parti- slow steaming and the options available to operators, given the
cularly if the concerns regarding the effect of methane slips on new sulfur limits. Finally, works considering the possible exten-
increasing the emission profiles of LNG-fueled ships are not ad- sions of ECAs – such as of Schinas and Bani (2012) and Chang et al.
dressed (IPCC, 2007; Corbett et al., 2014; Brynolf et al., 2014). (2014) – highlight the societal benefits and different aspects of the
Therefore, a thorough understanding of the risks and limitations is problem, while the work of Panagakos et al. (2014) indicates the
necessary. possible modal shift that increases the relevant externalities.
The issue of marine bunkers has not been widely discussed by In contrast to the commercial issues, the impacts of using LNG
the academic community, despite its importance for the financial as a marine fuel on employment (chartering and scheduling) and
survival of operators (Stopford, 2009, see Table 6.1, p. 224). It is operations (bunkering facilities, engine performance and relia-
indicative that relatively few papers focused on hedging and price bility, etc.) have been widely discussed in the literature. The fol-
forecasting have been published, beyond the contributions of lowing works are mentioned to show the width and depth of the
Menachof and Dicer (2001), Alizadeh et al. (2004), Schinas and related analysis: Burel et al. (2013) analyzes maritime traffic, and
Stefanakos (2013), Stefanakos and Schinas (2014), and Stefanakos identifies ship types and trades with a higher probability of
and Schinas (2015). This paper builds upon these contributions shifting to fuels with reduced emission profiles. Chang et al. (2008)
and examines the price of LNG as a marine fuel in comparison to provides a reliability analysis of the propulsion systems and Liva-
conventional marine fuels – i.e. heavy fuel oil (HFO), marine diesel nos et al. (2014) provides a technical analysis of the propulsion
oil (MDO), and marine gas oil (MGO). plants and a life cycle analysis of the costs. From a technical point
The option of using LNG as a marine fuel has attracted the in- of view, many of the concerns are addressed.
terest of researchers, mainly due to the sulfur limits introduced by In short, due to the following facts and risks that will be ana-
the International Maritime Organisation (IMO) in Annex VI of the lyzed in detail in the later sections of this study, the majority of
International Convention for the Prevention of Pollution from operators are not willing to switch to LNG or more expensive
Ships (MARPOL). Annex VI sets the limits for greenhouse gases conventional marine fuels with reduced emissions profiles (e.g.
(GHG) and non-greenhouse gases (nGHG) emitted from the en- MDO and MGO), and are unwilling to shift operational practices
gines of ships, and has impacts on ship operations in the high seas, and technologies:
as well as in Emission Control Areas (ECAs). Relevant academic
publications, indicating the promising potential of LNG as a marine 1. There is no market trend or commitment that indicates ships
fuel, are widespread (Ma et al., 2012; Yang et al., 2012; Burel et al., with reduced emissions profiles will attract more cargoes or be
2013; Holmgren et al., 2014; Thomson et al., 2015). In an effort to able to charge higher freight rates than their competitors to
further explore the impact of the regulation, Schinas and Stefa- increase revenues. Consequently, any increase of the cost ele-
nakos (2012) present a model that estimates the impact of oper- ments will lead to a deterioration in profit margins (Psaraftis
ating in an ECA on the life-cycle cost of assets. Brynolf et al. (2014) and Kontovas, 2010).
consider LNG as an option, yet they highlight the negative effect of 2. There are operational risks associated with the global avail-
methane slip and suggest regulatory action. Fagerholt et al. (2015) ability of fuels, such as MGO and LNG. There are many concerns
present an optimization model that minimizes the cost of routing related to the availability of LNG-bunkering facilities, particu-
in and out of ECAs, and also indicates the possible negative side larly the geographic and routing restrictions associated with
effect of increasing CO2 emissions when switching fuels. In their using LNG-fueled vessels (i.e. vessels can only serve specific
work, the LNG option is not considered. Yet, the push-down, pop- markets and routes).
up phenomenon, presented in the book edited by Psaraftis (2016), 3. At present, there is no after-market for LNG-fueled ships.
is illustrated with numerical examples. These recent works are 4. There is regulatory uncertainty regarding not only the future
based on the relevant contributions of Psaraftis and Kontovas limits for currently restricted emissions, but also the probable
(2010, 2013, 2014) that address the issue of routing under given limits for known pollutants that are not yet restricted, which
environmental constraints. Pertinent work on slow steaming – LNG-fueled ships may or may not address.
such as of Corbett et al. (2009); Wang and Meng (2012), and
Doudnikoff and Lacoste (2014) – further explores the benefits of The validity of the above considerations and the impacts on
86 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96

2. Regulatory framework

The justification for introducing LNG-fueled ships, beyond LNG

carriers, stems from the advent of air emission regulations in-
tended to reduce the emissions of GHGs, such as carbon dioxide
(CO2), and nGHGs, such as sulfur oxides (SOx) and nitrogen oxides
(NOx). From the early stages, it became evident that it was ne-
cessary to implement a single technology for the abatement of
both GHGs and nGHGs, as the implementation of the technologies
reducing nGHGs, such as SOx, also resulted in fuel consumption
increases, thereby increasing CO2 emission (see, e.g. Psaraftis,
2016; Fagerholt et al., 2015). Given this necessity for a holistic
technical solution, the technology for LNG-fueled engines with
reduced emission profiles has gained momentum and attracted
the interest of manufacturers, policy-makers, and operators.
However, the use of LNG is not the only option for reducing
Fig. 2. Schematic representation of NOx reduction targets based on the rated en- emissions and, hence, there is no absolute need for LNG as a
gine speed. marine fuel. Scrubbers and fuel switching can also be deployed,
and until 2020–2025, in most areas, HFO can be used. The option
managing operations will be discussed in the coming sections. of desulfurized HFO – i.e. with a content of less than 0.5% of sulfur
In order to assess the commercial feasibility of LNG-fueled – is also examined.
ships and to propose a methodology to evaluate the commercial The positive impacts of using LNG as a marine fuel imply that a
incentives required to promote LNG as a marine fuel, the following tight control of methane slip is achievable during all operations. As
steps of analysis are considered: a GHG, methane is considered to be more harmful than CO2 be-
cause it traps more heat energy in the atmosphere (Myhre et al.,

Section 2 briefly describes the international regulatory frame- 2013). However, methane also dissipates from the atmosphere
works and regional initiatives that support the use of LNG as a more quickly than CO2. For LNG-fueled vessels, methane escapes
marine fuel, as well as important studies, academic contribu- the propulsion system as a result of incomplete combustion or
tions, and arguments related to the field of air emissions from venting from the pipes after the engine has stopped, especially in
ships. Otto cycle engines. Generally, incomplete combustion is avoided

Section 3 is devoted to the analysis of LNG pricing as a marine by injecting LNG directly into the cylinder to better control the
fuel and energy parcel. Such an analysis is not available in the combustion gas mixture or by using the so-called lean burn, where
literature, despite broad acknowledgment that bunker prices a significant surplus of air is added; any methane that is not
are a determinant for investment and technology adoption (see burned in a timely manner may react with the excess air on its
e.g. Schinas and Stefanakos, 2014). way out to the exhaust valve. After treatment of the exhaust gases

Section 4 and especially Section 4.1 discuss the regulatory, is also a possibility, in a similar way as scrubbing for SOx Smith
commercial, and operational attributes that determine the et al. (2014). Further investigation of the potential issues related to
competitiveness of LNG-fueled ships vis-à-vis conventionally using methane in vessel propulsion systems are critical towards
fueled ships and derive a condition relating the cost elements to the effective abatement of GHG from ships; yet, as long as IMO
assess the competitiveness of LNG as an alternative marine fuel. upholds emission standards that focus on individual vessels, LNG-
This relationship can be used to assess policy initiatives, in fueled propulsion plants are a way to meet these requirements.
terms of subventions of discount of fees, which encourage the The issues related to nGHG and Annex VI including ECA, as well
use of alternative technologies and estimate their market as to the sulfur Directive 1999/32/EC, as amended by the EU Di-
impact. Section 4.2 identifies the remaining challenges and rectives 2012/33/EC and 2005/33/EC and the California Air Re-
considerations for increasing market acceptance of LNG as a source Board (CARB) rules will not be analyzed any further. Many
marine fuel and evaluates the prospects for local -gold plating- texts and sources are available (see e.g. European Commission,
initiatives as a means toward accelerating the abatement of the 2011, 1999, 2005; 2012b, European Commission, 2013; Schinas,
externalities of air emissions. 2013; DNV-GL, 2014; Schinas, 2015; Psaraftis, 2016), and

Section 5 reiterates the key questions, regulatory frameworks Figs. 2 and 3 are provided only to remind the reader of the es-
(Section 2), and findings of Sections 3 and 4. tablished limits and implementation dates.
Despite the regulatory momentum, the commercial and fi-
nancial uncertainties associated with LNG pricing, the availability

Fig. 3. MARPOL Annex VI and European sulfur regulation limits.

 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96 87

attract the attention of policy-makers and regulators, as in the case

of the European Parliament (2015a, 2015b). Therefore, decisions
affecting the selection of technologies for onboard energy re-
quirements should also consider the provisos for these pollutants,
as part of proactive operational planning initiatives.
The abatement of CO2 is linked with the consumption of bun-
kers and CO2 is the GHG pollutant currently attracting interest
from regulators. Chapter 4 of MARPOL Annex VI, and specifically
Regulations 19, 20, and 21, as well as Resolution MEPC.245(66) of
2014, provide the necessary information for the implementation of
the Energy Efficiency Design Index (EEDI). The idea behind the
EEDI is to provide a measurement (or an indication) of energy
efficiency, based on CO2 emissions (g) per unit of transport (ton-
mile). It is also clear that CF, a non-dimensional conversion factor
Fig. 4. Required EEDI against existing and proposed reference lines (Bazari, 2012). between fuel consumption and CO2 emissions (see par. 2.1 of the
Resolution), ranges from 3.114 to 3.206, depending on the grade of
conventional fuel, as per the international classification. In con-
of LNG fuel (bunkering facilities), and the high additional invest- trast, the CF for LNG is 2.75 – i.e. a reduction potential of circa 14%.
ment required are the most significant barriers preventing the The concept behind the EEDI is clearly depicted in Fig. 4. An EEDI
adoption of LNG as a marine bunker. Based on the current market measurement is calculated for every new or existing ship, as per
situation, the only vessels with a high likelihood of adopting LNG the provisos of Regulation 19, based on the formula provided in
as a fuel are vessels operating on fixed routes (containerships, the Resolution MEPC.245(66) and the levels provided in Regula-
RoRo) and relatively large vessels engaged in regional trades, tion 21. The estimated EEDI measurement should be below the
especially in ECAs (Burel et al., 2013; Acciaro, 2014). That said, the reference line; otherwise, measures should be taken to reduce the
global sulfur cap coming into force in 2020 or 2025, together with emission of CO2 per unit of transport. The reference line will be
the sulfur limit imposed by the EU for EU waters in 2020, is reviewed by the IMO at the given years where lower reference
strengthening the position of LNG as a marine fuel. It is expected lines will be considered, thereby pushing operators to increase the
that when these regulations come into force, larger oceangoing unit energy efficiency. Recently, the EU adopted Regulation 2015/
vessels (bulkers and tankers) will consider LNG as a compliance 757 on the monitoring, reporting, and verification of carbon di-
option (DNV-GL, 2014). Given the current regulatory limits of oxide emissions from maritime transport (MRV Regulation). This
Figs. 2 and 3, operators effectively have three compliance options, regulation demands the collection and eventual publication of
which are categorized as follows:
verified annual data on CO2 emissions from all large ships calling
at EU ports from January 1, 2018, irrespective of where the ships

Fuel switch: operators can install dual-fuel systems, which al- are registered (European Commission, 2015).
low high-sulfur heavy fuel oil (HS-HFO) to be consumed when
the ship is operating outside an ECA, and low-sulfur heavy fuel CO2 from the propulsion + CO2 from the auxiliary
oil (LS-HFO), MDO, or MGO to be consumed while inside an − CO2 innovative technology
ECA. This solution complies with Regulation 14 for SOx; how- EEDI =
DWT × speed (1)
ever, it also increases operational risks, especially for ships that
frequently enter or operate within an ECA. Furthermore, SCR, The simplified form of the EEDI formula, provided in Eq. (1),
EGR, or a relevant technology must be used for the abatement of clarifies how the unit energy efficiency can be affected.
NOx (see e.g. Brynolf et al., 2014; Burel et al., 2013).

Add-on technology: operators can install an exhaust gas
Increasing the nominator implies a higher utilization of the ship
cleaning system (EGCS) that desulfurizes the exhaust gases. This – i.e. more loaded ton-miles per annum executed or an adjust-
option implies that HS-HFO can be consumed in all cases. As in ment of the speed. The utilization and the determination of the
the previous option, a relevant NOx abatement technology must average operating speed are market-related decisions: If the
also be installed. market is strong, then operators increase speed to offer more

LNG fuel: operators can use LNG as a marine fuel. LNG provides ton-miles; when the market is weak, slow steaming is also
significant reductions in SOx and NOx emissions, which allows considered. The volumes of trade cannot be affected.
operators to comply with existing and proposed regulatory
Decreasing the nominator implies reducing the CO2 emitted
limits. from the main and auxiliary engines or increasing the output, in
terms of improving the energy efficiency of the technology.
The difference between fuel prices among HS-HFO and LS-HFO Based on the EEDI, the use of LNG implies a lessening of the CO2
and the amortization schedule of the add-on technology or initial emission, due to the substantially lower CF.
investment determines the attractiveness of the available options,
which are thoroughly discussed in the literature (Schinas and Nevertheless, even if all technical measures apply, the policy
Stefanakos, 2013, 2014). Further analysis of these areas has been goal is not achieved. Therefore, market-based measures (MBMs)
deemed unnecessary for the purposes of this article. Nevertheless, should also be considered for implementation. Psaraftis (2008)
the issue of GHGs and especially the emission of CO2 from LNG- provides a thorough presentation of these measures that include
fueled ships attracts the interest of researchers and of the market, proposals leading to CO2 trade, funds (charge per ton of fuel), caps,
as highlighted in the literature (indicatively by Burel et al. (2013), and levies.
Brynolf et al. (2014), and Thomson et al. (2015)), as promoted by In conclusion, the international regulatory framework fosters
policy-makers (e.g. by the European Commission (2012a)), and LNG as a marine fuel through initiatives designed to abate NOx and
addressed by experts (such as (DNV-GL, 2014) and Semolinos et al. SOx, and reduce CO2 emissions. As thoroughly outlined in Section
(2011)). In the near future, the issues of methane (CH4), ammonia 2, the following dates emphasize the need for operators to begin
(NH3), and particulate matter ( PM2.5) emissions are expected to the transition to ships with reduced emission profiles:
88 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96

1. ECA limits at 0.1% as of 2015 (Fig. 3). and produces minimal PM. Similar results are derived by Thomson
2. Phase 1 of EEDI (Fig. 4) in 2015. et al. (2015). Hence, the benefits of using LNG as a marine bunker
3. Tier III NOx in the North America ECA in 2016. are known and widely accepted, but the induced higher costs
4. Possible first year of application of the MRV Directive in Europe hinder its commercial acceptance. In that respect, Agnolucci et al.
in 2018 (European Commission, 2015). (2014) concluded that the direct operational benefits and the re-
5. Phase 2 of EEDI (Fig. 4) in 2020. levant incentives are split among owners and charterers – i.e.
6. EU sulfur cap of 0.5%. among the providers and the users of maritime transport services.
7. Phase 3 of EEDI (Fig. 4) in 2025. This is a very interesting argument that should be further con-
8. Possible additional ECAs. sidered when drafting similar policies and in the following
9. Possible operational requirements for the reduction of CO2. sections.
10. Possible global MRV regulation.
11. Possible global sulfur cap.
12. Possible regional and local (port-specific) initiatives.
3. Price of LNG as a marine bunker
Regional approaches and incentives are not limited to the
The advent of the LNG technology also implies that operators
abatement of SOx and NOx. The abatement of GHG is also of
will adjust their operational and commercial practices accordingly.
paramount importance for city ports. Hence, many ports provide
In contrast to conventional bunkers, where the supply and energy
incentives for operations that are deemed to be beyond minimum
content are relatively well known and understood, the supply of
compliance or peer performance standards. Some of these in-
LNG is more complicated and the capacity required to store the
centive regimes may be related to the Environmental Ship Index
same energy content is notably increased. Therefore, there is
(ESI), which is a measurement derived by the World Port Climate substantial ambiguity surrounding the pricing of LNG as a marine
Initiative (WPCI) of the International Association of Port and bunker.
Harbors (IAPH). The ESI determines seagoing ships that outper-
form the current emission standards. Based on the ESI of a ship, a
First, the various natural gas pricing mechanisms (e.g. hub-
special tariff or scheme of benefits applies for the ships that reduce based, oil -indexed, bilateral monopoly, regulated, and sub-
their CO2 footprint. Examples include: sidized) that serve the needs and purposes of natural gas as an
energy commodity (i.e. not as a marine bunker), combined with

The port of Hamburg offers a rebate (Hamburg Port Authority, the unknown costs of distributing LNG to a ship, obscure the
2013). transparency of LNG pricing for marine bunkering purposes.

The port of Rotterdam awards a “Green Trophy” (Vinkoert,
Second, the quality of LNG as a natural gas mixture, which
2012). varies depending on the natural gas reservoir, production pro-

Swedish ports have adopted and promoted simple concepts and cess, and liquefaction facility, determines its energy content,
ideas that could be further applied in other regions (Mellin and price, and usability as a marine fuel.
Rydhed, 2011).
Third, the liquefaction technology – i.e. transportation of natural

Singapore has extended the incentives by introducing the gas (NG) to a liquefaction facility, refrigeration/liquefaction
“Green Ship”, “Green Port”, and “Green Technologies” programs. process, cryogenic storage facilities, and transportation from
The “Green Ship” initiative aims to attract ships under the reg- LNG storage to the ship – determine the final price. Un-
istry and the admittance criterion is that the ship should exceed fortunately, these costs are not publicly available, and the
the EEDI limits set by the IMO. Qualifying ships will enjoy a known hub-based and oil-indexed prices do not include these
significant reduction of registration fees and tax burdens. The costs.
“Green Port” initiative promotes ships with reduced GHG foot-
prints and the “Green Technologies” initiative encourages local The most common approach to consider the price that marine
companies to develop or invest in relevant technology. operators will pay for LNG fuel is the sum of either the: Henry Hub
(HH) natural gas, UK National Balancing Point (NBP) natural gas, or
Apparently, these incentives reward “greener” operators by par- a European LNG import price, and a adjustment for bunkering
tially compensating the investment cost (MPA Singapore, 2015; (Table 1).
Tay, 2012). As per the analysis of Algell et al. (2012), there are only three
Burel et al. (2013, p. 414) states that LNG as a marine fuel abates ways for bunkering LNG:
approximately 80–85% of NOx, due to the lean combustion process;
practically eliminates SOx, as LNG does not contain sulfur; reduces 1. Ship to ship transfer.
CO2 emissions by 20–30%, relative to conventional marine fuels; 2. Truck to ship transfer.

Table 1
LNG plus supply cost, MDO, and MGO against HFO (Rotterdam).

MDO/HFO MGO/HFO LNG0/HFO LNG20/HFO LNG50/HFO LNG100/HFO LNG120/HFO LNG200/HFO

2005 1.96 2.17 1.47 1.56 1.69 1.90 1.99 2.33

2006 1.79 1.96 0.91 0.98 1.08 1.25 1.32 1.59
2007 1.66 1.82 0.80 0.86 0.95 1.09 1.15 1.38
2008 1.80 1.95 0.75 0.79 0.85 0.96 1.00 1.17
2009 1.39 1.49 0.44 0.50 0.58 0.72 0.78 1.01
2010 1.48 1.52 0.39 0.43 0.50 0.61 0.65 0.83
2011 1.52 1.53 0.26 0.29 0.34 0.42 0.45 0.58
2012 1.49 1.49 0.17 0.20 0.25 0.33 0.36 0.48
2013 n/a 1.52 0.25 0.28 0.33 0.42 0.45 0.59
2014 n/a 1.53 0.33 0.37 0.42 0.52 0.56 0.71
 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96 89

3. Intermediate tank to ship transfer (pipeline to ship). Table 2

Data for the X82DF-32880 kW with nine cylinders of Wärtsilä (2015, p. 83).
The related supply costs vary significantly, as local port conditions,
Fuel Specific consumption Daily consumption (est.) Daily costs
traffic, and availability of infrastructure affect the supply chain. gr tons USD
Detailed examples in the analysis suggest supply costs that vary kWh day day

from 50 USD/t of LNG (p. 102) to 630 USD/t (p. 104); intermediate
HFO 178.9 141.2 34.5 k
costs of 115 (p. 101) and 200 (p. 105) USD/t are also reported MDO 178.9 141.2 52.3 k
(Algell et al., 2012). LNG 138.7 109.5 20 k
For the purposes of this analysis, the prices for natural gas and pilot MDO 1.3 1.0
LNG will be converted into the standardized tonne of oil equiva-
lent (TOE) energy metric using the IGU methodology (IGU, 2012,
see reference 2, page 18) or IGU (2012) to allow a statistical Table 3
comparison with conventional marine bunkers (e.g. HFO, MDO, Numerical example of 1500 TEU-slots containership.
and MGO) based on energy content – i.e. USD/TOE. This will also
Conventional LNG-fueled
allow the relative comparisons in the following sections, as all Total 15,400 15,400
maritime market data and benchmarking are based on oil terms.
The energy contents per tonne of LNG, HFO, MDO, and MGO are CAPEX 4500 29% 5400 35%
based on composite values intended to approximate typical fuel OPEX 4500 29% 4500 29%
VOYEX 6400 42% 5500 36%
qualities available to marine operators, which are known to vary.
Considering the analysis of Algell et al. (2012), an estimated Port 1600 25% 1420 26%
Fuel 4800 75% 4080 74%
supply cost of 20, 50, 100, and 200 USD/TOE is added to the NG
and LNG prices to represent the free on board (FOB) price, which
interests ship-owners and operators. It is expected that the in- include, apart from other forms, a mixture of Gaussian pdfs, the
creasing adoption and maturity of LNG technology throughout time series is not normally distributed.
various sectors will reduce the total supply costs; hence, lower As in the case of HFO, the distribution of HH prices will be
costs are considered. The sum of HH FOB price and of the supply difficult to approximate. The results of the statistical analysis also
cost of LNG will be compared with the prices for MDO, MGO, and σ
suggest that the coefficient of variation (CV), CV = μ for HH natural
HFO fuels. LNG, MDO, and MGO are considered as desulfurized,
gas prices is 0.49, substantially lower than the CVs of HFO, MDO,
where no further treatments are required to comply with reg-
and MGO. Table 1 compares the average annual FOB energy prices
ulatory limits, but HFO has an extra operational cost for exhaust
(USD/TOE) for MDO, MGO, and LNG to Rotterdam HFO prices using
gas treatment with scrubbers. The analysis is based on the marine
unity values throughout a range of “supply costs” – i.e. where a
bunker prices reported in Rotterdam only because the price var-
value greater than one indicates that the alternative fuel has a
iations for conventional bunkers among the major bunkering hubs
higher price per unit of energy than HFO, given the cost of energy
– namely Singapore, Fujairah, and Houston – are negligible (see
and additional supply cost (0–200 USD/TOE).
Fig. 5).
The benefit of the statistically expected lower price of LNG as
A similar descriptive analysis is provided in the Annex, as per
bunker, is further supplemented by the substantial evidence from
the statistical examination of Schinas and Stefanakos (2013). Ta-
engine manufacturers, that the specific fuel oil consumption
bles 4–8 and 10, as well as Fig. 6 provide the results of the uni-
(SFOC) of engines consuming LNG or a mix of LNG and conven-
variate statistical analysis of the HFO and MGO prices in Rotter-
tional fuel (dual fuel engines) is lower than the SFOC of engines
dam, the FO prices in Houston, Singapore, and Fujairah, as well as
of the HH LNG prices, that update and complete the findings of the consuming HFO. Hence, the benefit of using LNG as a marine fuel
previous work (see Tables 11–13 and Figs. 7–9 in the Annex). Al- may be higher through reduced consumption (Groene, 2015;
though the the probability density function (PDF) and empirical Wärtsilä, 2015).
cumulative distribution function (ECDF) of the HH price may This advantage can be clearly illustrated when considering
dual-fuel engines. In this case, the engines can consume either
MDO or LNG plus MDO as pilot fuel. When burning MDO, manu-
facturers estimate a specific fuel oil consumption SFOCMDO that can
⎡ ⎤
be easily translated in terms of daily consumption in ⎣ tons
day ⎦
, the
value that is considered in commercial transactions (subject to the
necessary adjustments and considerations). When burning LNG,
manufacturers estimate the equivalent specific gas consumption
SGC and the necessary specific pilot fuel oil consumption SFOCpilot.
Should the case SGC + SFOCp < SFOCMDO be valid, and considering
that LNG is circa j% less expensive than MDO – i.e.
PriceLNG = j*PriceMDO – then the actual savings are estimated as:
(SGC*PriceLNG + SFOCpilot *PriceMDO ) − SFOCMDO *PriceMDO
=
SFOCMDO *PriceMDO
(SGC*j*PriceMDO + SFOCpilot *PriceMDO ) − SFOCMDO *PriceMDO
=
SFOCMDO *PriceMDO
(SGC*j + SFOCpilot ) − SFOCMDO
SFOCMDO (2)

As an example, based on the data and values available from

(Wärtsilä, 2015) for the power/speed rating point R1 – i.e. for the
Fig. 5. Unity value comparison of MDO, MGO, and LNG vs. HFO energy content
prices in Rotterdam, based on calculated USD/TOE prices. nominal maximum continuous rating (MCR) – see Table 2, where
90 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96

Table 4
FOB as per Henry Hub Statistics plus Supply Chain Cost of 0 USD/TOE.

Year Count Average Min Max Range STD Skewness Kurtosis

1997 249 98.80 70.23 186.89 116.66 20.36 1.19 4.64

1998 251 82.87 41.66 105.15 63.49 9.42  0.81 4.62
1999 250 90.24 64.68 123.01 58.33 14.07 0.15 2.28
2000 249 171.08 85.71 416.24 330.53 68.38 1.59 5.89
2001 250 157.10 67.06 409.10 342.04 71.54 1.29 4.77
2002 250 133.94 80.55 210.70 130.15 28.24 0.31 2.75
2003 250 217.10 157.93 733.29 575.36 50.78 5.42 48.43
2004 249 233.83 171.42 322.20 150.78 24.64 0.53 3.58
2005 241 344.66 219.43 610.68 391.24 104.94 0.94 2.57
2006 249 267.10 145.23 392.83 247.60 43.68  0.16 3.17
2007 252 276.46 210.30 362.68 152.37 28.11  0.18 2.83
2008 253 351.67 213.08 528.14 315.06 83.36 0.41 2.20
2009 252 156.44 72.61 242.05 169.43 32.74 0.55 3.17
2010 252 173.39 126.18 298.00 171.81 27.02 1.01 4.59
2011 252 158.57 112.69 195.23 82.53 18.80  0.54 2.51
2012 252 109.30 72.22 149.59 77.38 19.26 0.05 2.06
2013 252 148.06 122.21 179.35 57.14 12.69 0.36 2.40
2014 241 175.53 135.71 323.39 187.69 29.83 2.41 11.00

All years 4494 185.64 41.66 733.29 691.62 90.98 1.33 5.38

Table 5
HFO 380cSt. Rotterdam USD/ton.

Year Count Average Min Max Range STD Skewness Kurtosis

2005 52 235.69 141.50 291.50 150.00 38.60  0.74 2.71

2006 52 292.38 249.50 331.00 81.50 23.67  0.21 1.88
2007 52 345.99 219.50 495.00 275.50 74.20 0.25 2.22
2008 52 474.33 171.50 707.00 535.50 145.89  0.61 2.54
2009 52 352.73 173.50 466.00 292.50 84.75  0.40 1.71
2010 53 451.06 405.00 494.00 89.00 22.99 0.08 2.12
2011 52 619.51 511.50 669.50 158.00 39.33  1.55 4.71
2012 52 638.77 550.50 720.00 169.50 50.20 0.06 1.69
2013 52 594.49 567.00 646.00 79.00 17.23 0.87 3.74
2014 52 533.36 305.00 601.00 296.00 77.84  1.70 4.71

All years 1314 244.46 50.50 720.00 669.50 196.91 0.95 2.41

Table 6
HFO 380cSt. Fujairah USD/ton.

Year Count Average Min Max Range STD Skewness Kurtosis

2005 52 258.06 173.00 319.00 146.00 39.68  0.52 2.28

2006 52 310.19 262.00 352.50 90.50 28.34  0.24 1.60
2007 52 374.73 258.00 513.00 255.00 68.26 0.41 2.36
2008 52 512.12 214.00 736.00 522.00 152.52  0.56 2.45
2009 52 371.85 221.50 495.00 273.50 87.51  0.39 1.53
2010 53 469.01 425.00 515.50 90.50 22.90 0.18 2.07
2011 52 654.43 530.50 705.50 175.00 38.51  1.78 6.07
2012 52 667.52 581.00 751.50 170.50 54.99 0.09 1.54
2013 52 616.87 590.00 663.00 73.00 16.77 0.90 3.70
2014 52 563.21 330.00 623.00 293.00 77.33  1.74 4.87

All years 1314 258.75 51.00 751.50 700.50 207.17 0.91 2.37

⎡ ⎤ 4. Regulatory and commercial considerations

savings of circa 20% in terms of daily consumption in ⎣ tons
day ⎦
, and
USD
savings of 30% in might be considered (as per the averages 4.1. Commercial and operational feasibility of LNG-fueled ships
day
estimated in Tables 4 and 9). Considering also the dual-fueled
auxiliary engines, the total savings in terms of daily consumption In the previous Section 2, the current and future regulatory
are increased. However, one should consider the option of running requirements, as well as some notable regional initiatives were
a similar engine on HFO with a scrubber, given the proper treat- outlined. Unless there is a regulatory initiative that dictates the use
of LNG-fueled ships, these ships will continue to compete with
ment and application of compliance technology. Even draft cal-
compliant conventionally fueled ships. However, the possibility of
culations can highlight the benefits of using LNG and support
a drastic regulatory initiative favoring LNG or alternative fuels
further optimism.
should not be dismissed,
In both markets, ship-owners provide a warranty of sea-
worthiness; therefore, their ships should comply with the
 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96 91

Table 7
HFO 380cSt. Houston USD/ton.

Year Count Average Min Max Range STD Skewness Kurtosis

2005 52 249.98 166.00 331.00 165.00 40.65  0.55 2.51

2006 52 302.23 254.00 349.00 95.00 26.99  0.27 1.75
2007 52 352.98 236.00 498.50 262.50 71.69 0.33 2.28
2008 52 499.10 223.50 747.00 523.50 148.88  0.46 2.36
2009 52 359.88 211.00 472.50 261.50 78.94  0.30 1.62
2010 53 450.22 402.50 493.00 90.50 22.06 0.04 2.33
2011 52 627.03 497.00 675.50 178.50 42.91  1.65 5.04
2012 52 645.85 552.50 730.50 178.00 49.48 0.10 2.01
2013 52 602.14 566.50 655.00 88.50 20.42 0.59 2.89
2014 52 546.98 310.00 611.50 301.50 80.52  1.66 4.59

All years 1314 247.86 0.00 747.00 747.00 201.95 0.92 2.38

Table 8
HFO 380cSt. Singapore USD/ton.

Year Count Average Min Max Range STD Skewness Kurtosis

2005 52 263.45 163.50 327.50 164.00 43.59  0.42 2.37

2006 52 312.54 255.50 355.50 100.00 27.79  0.38 1.94
2007 52 373.65 266.50 513.00 246.50 68.56 0.53 2.32
2008 52 508.13 222.50 745.50 523.00 153.27  0.51 2.34
2009 52 371.09 223.50 476.50 253.00 83.79  0.39 1.56
2010 53 465.15 420.50 510.50 90.00 23.81 0.20 2.03
2011 52 648.40 526.50 704.50 178.00 38.15  1.71 5.70
2012 52 662.76 573.50 741.50 168.00 55.25 0.10 1.51
2013 52 615.72 591.50 667.50 76.00 18.41 0.84 3.00
2014 52 560.83 337.50 622.00 284.50 74.44  1.70 4.89

All years 1314 260.21 52.00 745.50 693.50 204.27 0.91 2.37

Table 9
MDO Rotterdam USD/ton.

Year Count Average Min Max Range STD Skewness Kurtosis

2005 52 460.66 327.50 607.50 280.00 62.82 0.06 2.84

2006 52 523.73 460.00 597.50 137.50 43.21 0.17 1.42
2007 52 572.21 422.50 825.00 402.50 98.45 0.77 2.82
2008 52 853.52 397.50 1202.50 805.00 225.17  0.39 2.22
2009 52 489.83 327.50 615.00 287.50 82.30  0.29 1.80
2010 53 668.84 565.00 795.00 230.00 56.96 0.28 2.20
2011 52 941.67 795.00 1052.00 257.00 56.77  0.64 3.43
2012 52 952.73 823.00 1030.00 207.00 53.92  0.50 2.39
2013 19 923.11 826.00 1025.00 199.00 63.01  0.14 1.86

All years 1168 370.76 87.00 1202.50 1115.50 286.52 1.08 2.88

Table 10
MGO Rotterdam USD/ton.

Year Count Average Min Max Range STD Skewness Kurtosis

2005 52 509.13 357.50 645.00 287.50 67.47  0.05 2.62

2006 52 573.28 512.50 650.00 137.50 46.15 0.25 1.45
2007 52 628.64 467.50 855.00 387.50 104.98 0.66 2.50
2008 52 920.97 430.00 1317.50 887.50 241.53  0.34 2.32
2009 52 528.08 367.50 657.50 290.00 84.68  0.28 1.80
2010 53 684.54 597.50 787.00 189.50 49.04 0.41 2.17
2011 52 946.41 788.00 1027.50 239.50 53.66  1.06 3.96
2012 52 954.15 822.50 1040.00 217.50 54.25  0.62 2.78
2013 52 902.69 812.50 1002.50 190.00 42.32 0.11 3.02
2014 52 818.44 512.50 902.50 390.00 98.16  1.65 4.88

All years 1314 409.49 0.00 1317.50 1317.50 308.23 0.76 2.28

mandatory regulatory framework. The selection of a competitive by the hire, but also the consumption at a given speed. This im-
ship is practically determined by the freight rate and the time- plies that owners have an interest to provide energy-efficient ships
charter hire offered. In other words, ceteris paribus, the lower the – i.e. ships with a lower unit cost of energy – given that these ships
freight rate, the higher the probability for voyage-based employ- consume the same bunkers. Consequently, the comparison of
ment. In the time-charter market, charterers are driven not only conventionally and LNG-fueled ships should be based on the
92 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96

transparent and, like conventionally fueled ships, the value is ex-

pected to vary significantly with market conditions. That said,
governments can also provide support – e.g. Norway established
the NOx fund (NHO, 2014) and many private funds have developed
finance models for investments in alternative or low-emission
fuels, including LNG-fueled ships.
As with most technologies, as more suppliers enter the market,
and more specialized and cost-effective technology is developed,
the fuel system costs are expected to decrease. The current pre-
mium that shipyards may charge to build an LNG-fueled vessel has
not been presented in academic or professional literature. How-
ever, a mark-up in the newbuild price should be expected to re-
flect the additional risk and expertise associated with construc-
tion, particularly for shipyards considered to be less flexible. In
general, the overall additional cost for an LNG-fueled vessel
(mainly applicable for merchant vessels such as bulkers, contain-
ers, and tankers) is 15–30% of the newbuild cost of a conventional
Fig. 6. PDF and ECDF of FOB as per Henry Hub Statistics.
vessel. The range is dependent on the vessel and tank size (DNV-
GL, 2014).
product of (1) the energy required and (2) the market price, for the
Although the implementation of the LNG technology will have
same speed, capacity, and voyage parameters.
an impact on operational expenses, the information released in
The cost structure of marine transportation is well analyzed in
the literature. As per Stopford (2009, Chapter 6), these costs can be academic and professional sources outlining the specific changes
broken down into: (1) capital expenses – i.e. expenses related to is limited. As the engine crew is determined on the basis of in-
the financing and acquisition of the ship; (2) operating expenses – stalled power (in HP or in KW), it can be assumed that no changes
i.e. expenses that deem the ship seaworthy, such as manning, lu- are expected in the number of hands dedicated to the propulsion.
bricants, spares, stores, administration, etc.; and (3) voyage ex- However, the provisos for maintenance and repair may be higher,
penses – i.e. costs related to the particular voyage or round trip, as the fuel storage and piping system is more complicated. Al-
such as port and canal dues, bunkers, and cargo-handling though all these considerations are plausible, it is expected that
expenses. the operational expenses will be more or less the same as those of
Capital expenses for LNG-fueled ships are expected to be higher conventional ships. Hence, for the purposes of this analysis, it is
than those of the conventionally fueled ships, due to the more assumed that the differences in operational expenses related to
expensive propulsion plant, associated technology, and procure- crewing, maintenance, and repair for LNG-fueled and con-
ment considerations. The biggest cost driver in the additional re- ventionally fueled ships are negligible.
quired investment for all vessels is the LNG tank. The additional As the cost of a particular voyage depends on the port and
capital expenses can vary between five to 20 million USD, de- channel dues, cargo-handling fees, and fuel expenses, substantial
pending on the tank and the engine size, as per market sources differences are expected in the cost structures of LNG-fueled and
(DNV-GL, 2014). The main factors that reduce the payback time
conventionally fueled ships. On assuming that cargo-handling fees
are: (1) the ECA exposure and (2) the LNG fuel price. Although the
are similar, the two other categories require attention. First, port
global sulfur cap in 2020 or 2025 will improve the business case by
dues are expected to be less for ships with reduced emission
making compliance necessary for the entire trip (see Fig. 3), the
profiles, as per Section 2. Second, the comparable price of energy
uncertainties surrounding the availability and pricing of LNG fuel
are ample cause for hesitation among ship-owners and operators. (in terms of TOE) for LNG-fueled ships is lower, as per the results
This reluctance to invest in LNG-fueled ships is further ex- of the analysis in Section 3 (note Fig. 5).
acerbated as pioneering owners face additional scrutiny from Therefore, the competitiveness of LNG-fueled ships vis-á-vis
traditional ship financiers, who express justifiable concerns about conventionally fueled ships depends on the differences in capital
the market value of ships with a limited second-hand market (i.e. expenses (CAPEX) and voyage expenses (VOYEX). Assigning the
the value of LNG-fueled vessels is taken as collateral if the ship- index 1 for conventionally fueled ships and 2 for LNG-fueled ships,
owner defaults on a non-performing vessel with increased capital the following calculations can be derived for similar ships if all
costs). Alternative finance methods exist (e.g. Sustainable Shipping other terms, such as capacity, voyage, etc. are to be equally com-
Initiative (SSI) (2014)), but the value of LNG-fueled ships is less petitive in the spot and time-charter markets:

Table 11
LNG plus supply cost, MDO, and MGO against HFO (Fujairah).

MDO/HFO MGO/HFO LNG0/HFO LNG20/HFO LNG50/HFO LNG100/HFO LNG120/HFO LNG200/HFO

2005 1.95 1.97 1.34 1.42 1.54 1.73 1.81 2.12

2006 2.03 1.99 0.86 0.92 1.02 1.18 1.25 1.50
2007 1.78 1.79 0.74 0.79 0.87 1.01 1.06 1.27
2008 2.04 2.00 0.69 0.73 0.79 0.89 0.93 1.08
2009 1.76 1.60 0.42 0.47 0.55 0.69 0.74 0.96
2010 1.52 1.53 0.37 0.41 0.48 0.58 0.63 0.80
2011 1.56 1.57 0.24 0.27 0.32 0.40 0.43 0.55
2012 1.57 1.55 0.16 0.19 0.24 0.31 0.34 0.46
2013 0.00 1.62 0.24 0.27 0.32 0.40 0.43 0.56
2014 0.00 1.71 0.31 0.35 0.40 0.49 0.53 0.67
 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96 93

Table 12
LNG plus supply cost, MDO, and MGO against HFO (Singapore).

MDO/HFO MGO/HFO LNG0/HFO LNG20/HFO LNG50/HFO LNG100/HFO LNG120/HFO LNG200/HFO

2005 1.84 1.86 1.32 1.39 1.51 1.70 1.77 2.08

2006 1.85 1.83 0.85 0.92 1.01 1.17 1.24 1.49
2007 1.67 1.70 0.74 0.80 0.88 1.01 1.06 1.28
2008 1.79 1.81 0.70 0.74 0.79 0.89 0.93 1.09
2009 1.39 1.42 0.42 0.47 0.56 0.69 0.74 0.96
2010 1.43 1.44 0.37 0.42 0.48 0.59 0.63 0.80
2011 1.44 1.45 0.25 0.28 0.32 0.40 0.43 0.55
2012 1.43 1.44 0.16 0.19 0.24 0.32 0.35 0.47
2013 n/a 1.51 0.24 0.27 0.32 0.40 0.44 0.57
2014 n/a 1.53 0.31 0.35 0.40 0.49 0.53 0.67

OPEX1 + CAPEX1 + VOYEX1 = OPEX2 + CAPEX2 + VOYEX2

CAPEX1 + VOYEX1 = CAPEX2 + VOYEX2 (3)

as OPEX1 = OPEX2 and, if CAPEX2 = (1 + k )*CAPEX1, where k is the

premium of the newbuild LNG-fueled ship, then Eq. (3) can be
rewritten as:

CAPEX1 + VOYEX1 = (1 + k )*CAPEX1 + VOYEX2 ⇒

⇒ VOYEX1 − k*CAPEX1 = VOYEX2 (4)

Eq. (4) implies that the voyage expenses of the LNG-fueled ship
should compensate for the acquisition premium. Breaking down
the voyage expenses into cargo handling, port dues, and fuel cost,
and assuming the following conditions:

VOYEX1 = Port1 + Fuel1 + Cargo1

VOYEX2 = Port2 + Fuel2 + Cargo2
Port1 = (1 − m)*Port2 Fig. 7. Unity value comparison of MDO, MGO, and LNG vs. HFO energy content
Fuel1 = (1 − n)*Fuel2 prices in Fujairah, based on calculated USD/TOE prices.

Cargo1 = Cargo2 (5)

m*Port1 + n*Fuel1 = k*CAPEX1 ⇒
allows Eq. (4) to be rewritten as: k*CAPEX1 − n*Fuel1
m=
Port1 (7)
Port1 + Fuel1 − k*CAPEX1 = Port2 + Fuel2 ⇒ Due to the geographic nature of specific ECA regulations, it is
m*Port1 + n*Fuel1 = k*CAPEX1 (6) more plausible to expect that containerships engaged in feeder
services and ships operating along the coastlines of ECAs will use
Eq. (6) expresses the condition where both ship types are equally LNG as a marine fuel. Assuming a typical feeder-containership of
competitive. This expression could also drive policy-makers to 1500 TEU slots capacity, 20% acquisition premium (k parameter in
provide incentives, as the discount for ships with reduced emis- Eq. (4)), and 15% less energy cost (n parameter in Eq. (5)), and the
sion profiles calling at their ports m (Eq. (7)) is estimated and indicative values of Table 3, then Eq. (7) yields m = 11.25% .
determined based on both technology maturity – i.e. the premium The values provided for this example are based on market data,
paid for the acquisition of an LNG-fueled ship – and energy market such as Clarkson's (2015) and HANSA (2014), and are given in USD/
conditions – i.e. the price difference of bunkering LNG relative to day. However, the data from this example can also be used to test
conventional bunkers. various cases, such as k = 20% , n = 20% , and m = 11.25% , where the

Table 13
LNG plus supply cost, MDO, and MGO against HFO (Houston).

MDO/HFO MGO/HFO LNG0/HFO LNG20/HFO LNG50/HFO LNG100/HFO LNG120/HFO LNG200/HFO

2005 2.05 n/a 1.39 1.47 1.59 1.79 1.87 2.19

2006 1.86 n/a 0.88 0.95 1.05 1.21 1.28 1.54
2007 1.73 n/a 0.79 0.84 0.93 1.07 1.13 1.35
2008 1.88 n/a 0.71 0.75 0.81 0.91 0.95 1.11
2009 1.46 n/a 0.43 0.49 0.57 0.71 0.77 0.99
2010 1.52 n/a 0.39 0.43 0.50 0.61 0.65 0.83
2011 1.55 n/a 0.25 0.29 0.33 0.41 0.45 0.57
2012 1.54 1.59 0.17 0.20 0.25 0.32 0.35 0.48
2013 n/a 1.65 0.25 0.28 0.33 0.41 0.44 0.58
2014 n/a 1.71 0.32 0.36 0.41 0.51 0.54 0.69
94 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96

higher hire might be compensated by lower voyage costs. This is

standard industry practice, as slightly larger ships might be char-
tered at a higher daily hire, consume less energy per unit, operate
at a slightly lower average speed, but envisage higher port dues,
vis-á-vis a smaller and slightly faster ship. Considering the sim-
plified formula for the EEDI, Eq. (1), one could conclude that the
nominator is sufficiently decreased; therefore, there is a sub-
stantial impact on the carbon footprint. If the denominator is not
affected, the reduction of the payload might be marginal and, gi-
ven the utilization of ships in weak markets, this consideration
may not be significant. The speed is largely determined by the
market and the ton-miles offered (see Ch. 2, Karakitsos and Var-
navides, 2014). However, speed affects the nominator and opera-
tion of the fleet at lower speeds, due to market conditions that
may remove the advantage LNG-fueled ships have against con-
ventional ones.
A point that has not been discussed so far, but which may have
financial consequences, is the loss of cargo space due to the in-
stallation of LNG tanks onboard. This consideration applies for new
ships, as well as for retrofits. Of the various ship types, the oil and
Fig. 8. Unity value comparison of MDO, MGO, and LNG vs. HFO energy content
prices in Singapore, based on calculated USD/TOE prices. chemical tankers segment provides the most flexibility for LNG
tank installations and retrofits, as no major modifications are re-
quired. LNG tanks could be placed on the deck, where plenty of
space is available. In existing ships, the main and auxiliary engines
need to be checked to ensure that they can be retrofitted to dual
fuel. Containerships and RoRo ships could also accommodate LNG
tanks relatively effortlessly. As in the case for oil and chemical
tankers, the installation of LNG tanks and retrofits are feasible, but
involve a penalty in terms of lost cargo slots. The LNG tank will
occupy container space, which is not desirable from a commercial
point of view, as it reduces the capacity available for utilization.
Bulk-carriers and general cargo-ships are faced with the biggest
challenge in terms of the LNG tank location. Tanks larger than
600 m3 will require significant modifications in the superstructure
to accommodate the tanks. Any retrofit will increase the cost and
the off-hire time (DNV-GL, 2014).
Finally, two ship categories that could interest policy-makers,
as they operate close to the coastline, are offshore vessels and
passenger ships. Offshore vessels will sacrifice a lot of cargo space
below deck where the tanks are normally placed, which will have
an impact on their financial performance. Moreover, in most off-
shore vessels, the installed diesel engines are four-stroke engines
Fig. 9. Unity value comparison of MDO, MGO, and LNG vs. HFO energy content and are very expensive to retrofit. In many cases, the cost of a
prices in Houston, based on calculated USD/TOE prices. retrofit might be close to the cost of installing a new engine, thus
making the investment far less attractive. For small passenger
LNG-fueled ship is more competitive than its peer (total cost of vessels, the location of the tank will be a challenge and, in some
15,160 USD/day) against the 15,400 USD/day for the con- cases, it might not be feasible to place a tank onboard. For larger
ventionally fueled ship. However, the original case without any passenger ships, deck space could be used for placing the LNG
reduction to the port dues, k = 20% , n = 15% , and m = 0% leads to a tanks, although at the cost of capacity or available space for the
slightly higher total cost of 15,580 USD/day. passengers.
Although it is not possible to generalize the results of a single In conclusion, due to the space requirements for the installa-
numerical example or apply the results to all scenarios, an implied tion of the LNG tanks, it is reasonable to expect that LNG-fueled
verification of the argument is expressed by Agnolucci et al. ships might be slightly larger than the conventionally fueled ships
(2014). Agnolucci et al. demonstrates that only part of the financial to accommodate the same cargo. Therefore, the energy efficiency
savings from energy efficiency accrues to ship-owners. offered through using LNG as a marine fuel is necessarily asso-
However, in general, LNG-fueled ships have higher running ciated with a capacity penalty. However, this penalty, although
expenses (CAPEX þOPEX) and lower voyage expenses. For time- evident, is not severe. Hence, a penalty has not been included in
charter markets, this implies an increased hire (USD/day), better the cost structure provided in Eqs. (3)–(7).
operational performance warranty given by the owners (con-
sumption and speed – i.e. tonne of LNG/day consumed at x knots), 4.2. Regulatory considerations
and potentially reduced port dues based on improved environ-
mental performance. Following the analysis of Sections 3 and 4.1, optimism for the
Time-charterers, when selecting a vessel, take into considera- wider market acceptance of LNG as a marine fuel is justified.
tion the given capacity, daily hire, and consumption at a given Nevertheless, it would be remiss to not briefly mention the well
speed, as well as other trade voyage parameters, such as the price known, yet often unresolved, challenges preventing the adoption
of bunkering at the expected ports of call. Therefore, a relatively of LNG as a maritime fuel:
 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96 95

1. bunkering infrastructure 5. Conclusions

2. availability of LNG as a fuel
3. after market In conclusion, the paper presents the feasibility and commer-
4. regulatory uncertainty. cial considerations of LNG-fueled ships, filling the gaps in the lit-
erature about the cost of LNG as bunker, as well as the potential
There is substantial evidence that these considerations should benefit to operators and a methodology that enables policy-ma-
be withdrawn. In Europe, there is a political decision to make LNG kers to promote these ships.
available in the main ports (European Commission, 2012a, 2014), The advent of LNG as a marine fuel is triggered by the efforts of
and in North America, there are many terminals sufficiently policy-makers to abate environmental burdens from the exhaust
equipped (ABS, 2015). As per Directive 2014/94/EU, LNG bunkering gases of ships. A thorough outline of the regulatory environment
infrastructure should be available till 2025 at ports of the core of triggering the use of alternative fuels in ships and, therefore,
Transport Transeuropean Networks (TEN-T) and till 2030 for all maritime networks, was presented. It seems evident that air
inland ports. Therefore, at least in the littoral states of ECAs in the emissions regulations will continue to demand that ships reduce
northern hemisphere, it seems that there is sufficient infra- emission and increase operational efficiency. Hence, policy and
structure or planned infrastructure to support LNG as a marine technology evolution and market initiatives are expected, which is
fuel. Should the European infrastructure plans materialise, then all a promising set of conditions for the market acceptance of LNG-
major European ports and the Mediterranean will soon be fully fueled ships.
covered with LNG bunkering spots, thus mitigating pertinent risks, Based on the statistical analysis of the cost of energy, the
limiting relevant concerns, and enhancing the range of trades ac- benefit of using LNG, in terms of the energy-price relationship, is
cessible to LNG fueled ships. The concerns that the supplies of LNG clear: The energy-price discount is sufficient to warrant the ac-
will not be sufficient for the needs of maritime operators, parti- quisition premium for LNG-fueled vessels. Relevant operational
cularly in areas not importing or exporting LNG, is also raised. As and regulatory risks should be considered further; however, most
the quantities of LNG required for marine fuel in comparison to the of them are fully or, for the time being, partly alleviated.
quantities of LNG required for natural gas electricity generation The derived Eq. (7) provides a condition to assess the compe-
and utilities is almost trivial, the high price differential between titiveness of LNG-fueled ships, given the accepted cost structure of
European markets and spot LNG delivered to Asia may redirect commercial ships. By using this formula or approach, stakeholders
LNG to Asia, leaving European terminals with a very low stock may assess or estimate the incentives required to promote LNG as
level (Semolinos et al., 2011, pp. 19–20). This risk also seems to be a marine fuel in a transparent way, which can be adjusted to ac-
overestimated, as there are many reports claiming sufficiency of commodate various ship types and trades. Given the inherent
LNG, at least in major hubs and ECA ports (Lloyd's Register, 2014). quandary regarding the evolution of global regulation, local in-
The risk of a lack of an after-market for the ships is also par- itiatives for gold plating should be considered to accelerate the
tially alleviated, as the installation and use of dual-fuel engines reduction of externalities from air emissions as regulatory acu-
permits the consumption of HFO, MGO, MDO, and LNG with pilot puncture enables more sophisticated strategies and collaborative
fuel. So, a premium or a discount for the installed dual-fuel pro- schemes to be deployed.
pulsion plant should be expected, given the market conditions. Finally, an overview of the challenges prohibiting the wider
The risk of regulatory uncertainty should be considered further. market adoption of the LNG technology was provided to reiterate
The adoption of the International Code of Safety for Ships using the concerns expressed by maritime operators and how, in some
Gases or other Low-flashpoint Fuels (IGF Code), along with cases, these concerns are, or will be, resolved.
amendments to make the Code mandatory under the International
Convention for the Safety of Life at Sea (SOLAS), is definitely a step
toward regulatory stability, as the code clarifies many technical References
issues International Maritime Organization (2015). In light of the
evolution of air emission regulation (see Section 2), it is possible ABS, 2015. Bunkering of Liquified Natural Gas-fuelled Marine Vessels in North
that new rules regulating other pollutants could emerge, making America. Technical Report 1.
Acciaro, M., 2014. Real option analysis for environmental compliance: {LNG} and
LNG technology obsolete or inadequate as a marine fuel. Although
emission control areas. Transportation Research Part D: Transport and En-
LNG technology seems to be less exposed to this risk, the safety, vironment. Emission Control Areas and their Impact on Maritime Transport, 28,
security, and operational regulatory considerations cannot be ex- pp. 41–50.
Agnolucci, P., Smith, T., Rehmatulla, N., 2014. Energy efficiency and time charter
cluded from future assessments. With respect to the global reg-
rates: energy efficiency savings recovered by ship owners in the panamax
ulatory risks and the associated predicaments in the evolution of a market. Transp. Res. Part A: Policy Pract. 66, 173–184.
global regulation, it seems that ports and regional policy-makers Algell, J., Bakosch, A., Forsman, B., 2012. Feasibility Study on LNG Fuelled Short Sea
can affect the market acceptance of LNG-fueled ships and, in and Coastal Shipping in the Wider Caribbean Region. TC/1208-05-2100, SSPA.
Alizadeh, A., Kavussanos, M., Menachof, D., 2004. Hedging against bunker price
general, the proliferation of low-emission ships at a faster pace. It fluctuations using petroleum futures contracts: constant versus time-varying
is a kind of regulatory acupuncture, where local action might have a hedge ratios. Appl. Econ. 36, 1337–1353.
global effect. This is evident in Europe and elsewhere, where local Bazari, Z., 2012. Energy Efficiency Design Index – An Update (Presentation at Royal
Institution of Naval Architects). Lloyd's Register.
policy initiatives have impacts on trade and ship traffic (see at the Brynolf, S., Magnusson, M., Fridell, E., Andersson, K., 2014. Compliance possibilities
end of Section 2). Eq. (7) offers a basis for more sophisticated for the future ECA regulations through the use of abatement technologies or
schemes that enable policy-makers to promote LNG-fueled ships, change of fuels. Transportation Research Part D: Transport and Environment 28,
6–18. Emission Control Areas and their Impact on Maritime Transport.
given the relative costs of conventional ships. Buhaug, Ø., Corbett, J., Endresen, Ø., Eyring, V., Faber, J., Hanayama, S., Lee, D., Lee,
Such local initiatives might be coupled with more sophisticated D., Lindstad, H., Markowska, A., Mjelde, A., Nelissen, D., Nilsen, J., Pålsson, C.,
policies, especially those that relate to carbon footprint, as there is Winebrake, J., Wu, W., Yoshida, K., 2009. Second IMO GHG Study. Technical
Report. International Maritime Organization (IMO). London, UK.
a need and substantial potential for improvement. Gold plating is
Burel, F., Taccani, R., Zuliani, N., 2013. Improving sustainability of maritime trans-
possible, especially in environmental regulation that applies re- port through utilization of liquefied natural gas (LNG) for propulsion. Energy
gionally, and the sulfur directive offers a useful example. In that 57, 412–420.
respect, LNG-fueled ships can be promoted justifiably and fairly by Chang, D., Rhee, T., Nam, K., Chang, K., Lee, D., Jeong, S., 2008. A study on availability
and safety of new propulsion systems for LNG carriers. In: Proceedings of the
considering the economics of the current conventional fleet or its 17th European Safety and Reliability Conference. Reliability Engineering &
segments. System Safety, 93, 1877–1885.
96 O. Schinas, M. Butler / Ocean Engineering 122 (2016) 84–96

Chang, Y.T., Roh, Y., Park, H., 2014. Assessing noxious gases of vessel operations in a Menachof, D., Dicer, G., 2001. Risk management methods for the liner shipping
potential emission control area. Transp. Res. Part D: Transp. Environ. 27, 12–18. industry: the case of the bunker adjustment factor. Marit. Policy Manag. 28,
Clarkson's, 2015. Shipping intelligence network. Retrieved (April 2015) from 〈www. 141–155.
clarksons.net/markets〉, Clarkson's electronic database. Miola, A., Ciuffo, B., 2011. Estimating air emissions from ships: meta-analysis of
Cockett, N., 1997. Neil Cockett on Bunkers. LLP Professional Publishing, ISBN-13: modelling approaches and available data sources. Atmos. Environ. 45,
978-1850449720. 2242–2251.
Corbett, J., Thomson, H., Winebrake, J., 2014. Natural Gas for Waterborne Freight Miola, A., Marra, M., Ciuffo, B., 2011. Designing a climate change policy for the in-
Transport: A Life Cycle Emissions Assessment with Case Studies. Technical ternational maritime transport sector: market-based measures and technolo-
Report. U.S. DOT Maritime Administration, Washington, DC. gical options for global and regional policy actions. Energy Policy 39,
Corbett, J., Wang, H., Winebrake, J., 2009. The effectiveness and costs of speed re- 5490–5498.
ductions on emissions from international shipping. Transp. Res., Part D 14, MPA Singapore, 2015. Maritime Singapore Green Initiative.
593–598. Myhre, G., Shindell, D., Bréon, F.M., Collins, W., Fuglestvedt, J., Huang, J., Koch, D.,
Corbett, J.J., Winebrake, J.J., Green, E.H., Kasibhatla, P., Eyring, V., Lauer, A., 2007. Lamarque, J.F., Lee, D., Mendoza, B., Nakajima, T., Robock, A., Stephens, G.,
Mortality from ship emissions: a global assessment. Environ. Sci. Technol. 41, Takemura, T., Zhang, H., 2013. Anthropogenic and natural radiative forcing, in:
8512–8518. http://dx.doi.org/10.1021/es071686z, PMID: 18200887. Stocker, T.F., Qin, D., Plattner, G.K., Tignor, M., Allen, S.K., Baschung, J., Nauels, A.,
DNV-GL, 2014. LNG as Ship Fuel. Technical Report 1. Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science
Doudnikoff, M., Lacoste, R., 2014. Effect of a speed reduction of containerships in Basis. Contribution of Working Group I to the Fifth Assessment Report of the
response to higher energy costs in sulphur emission control areas. Transp. Res. Intergovernmental Panel on Climate Change. Cambridge University Press.
Part D: Transp. Environ. 27, 19–29. NHO, 2014. The business sector's NOx fund.
European Commission, 1999. Directive 1999/32/EC of the European Parliament and Panagakos, G.P., Stamatopoulou, E.V., Psaraftis, H.N., 2014. The possible designation
of the Council. Official Journal of the European Union. of the mediterranean sea as a seca: a case study. Transp. Res. Part D: Transp.
European Commission, 2005. Directive 2005/33/EC of the European Parliament and Environ. 28, 74–90, Emission Control Areas and their Impact on Maritime
of the Council. Official Journal of the European Union. Transport.
European Commission, 2011. Pollutant emission reduction from maritime transport Psaraftis, H., 2008. Environmental risk evaluation criteria. WMU J. Marit. Aff. 7,
and the Sustainable Waterborne Transport Toolbox. Commission Staff Working 409–427. http://dx.doi.org/10.1007/BF03195142.
Paper. Psaraftis, H. (Ed.), 2016. Green Transportation Logistics; The Quest for Win-Win
European Commission, 2012a. Connecting Europe; the energy infrastructure for Solutions. volume 226 of International Series in Operations Research & Man-
tomorrow. agement Science. Springer International Publishing. 1st edition.
European Commission, 2012b. Directive 2012/33/EC of the European Parliament Psaraftis, H., Kontovas, C., 2013. Speed models for energy-efficient maritime
and of the Council. Official Journal of the European Union. transportation: a taxonomy and survey. Transp. Res., Part C 26, 331–351.
European Commission, 2013. Communication from the commission to the eur- Psaraftis, H.N., Kontovas, C.A., 2010. Balancing the economic and environmental
opean parliament, the council, the European economic and social committee
performance of maritime transportation. Transp. Res. Part D: Transp. Environ.
and the committee of the regions clean power for transport: a European al-
15, 458–462.
ternative fuels strategy. COM/2013/017 final.
Psaraftis, H.N., Kontovas, C.A., 2014. Ship speed optimization: concepts, models and
European Commission, 2014. Directive 2014/94/EU of the European Parliament and
combined speed-routing scenarios. Transp. Res. Part C: Emerg. Technol. 44,
of the Council of 22 october 2014 on the deployment of alternative fuels in-
52–69.
frastructure. Official Journal of the European Union, OJ L 307/71 Brussels.
Schinas, O., 2013. The issue of air emissions: policy and operational considerations,
European Commission, 2015. Regulation (EU) 2015/757 of the European Parliament
In: Xu, J. (Ed.), Contemporary Marine and Maritime Policy. Nova Science
and of the Council of 29 April 2015 on the monitoring, reporting and ver-
Publishers.
ification of carbon dioxide emissions from maritime transport, and amending
Schinas, O., 2015. Directive 1999/32/EC relating to a reduction in the sulphur
Directive 2009/16/EC. OJ L 123, 19.5.2015, p. 55–76.
content of certain liquid fuels and amending Directive 93/12/EEC, In: Werner,
European Parliament, 2015a. Air quality: environment MEPS call for tougher new
M.J., Jessen, H. (Eds.), Brussels Commentary on EU Shipping Law. C.H. Beck
national caps on pollutants. Press Release.
Verlag.
European Parliament, 2015b. Air quality: MEPS approve new national caps on
Schinas, O., Bani, J., 2012. The Impact of a Possible Extension at EU Level of SECAs to
pollutants. Press Release.
the Entire European Coastline. Note PE 474.549. European Parliament,
Faber, J., Markowska, A., Nelissen, D., Davidson, M., Eyring, V., Cionni, I., Selstad, E.,
Kågeson, P., Lee, D., Buhaug, O., Lindtsad, H., Roche, P., Humpries, E., Graichen, J., Directorate General for Internal Policies; Policy Department B: Structural and
Cames, M., Schwarz, W., 2009. Technical support for European action to redu- Cohesion Policies; Transport and Tourism.
cing Greenhouse Gas Emissions from international maritime transport. Tech- Schinas, O., Stefanakos, C., 2012. Cost assessment of environmental regulation and
nical Report DG ENV.C3/ATA/2008/0016. CE Delft. options for marine operators. Transp. Res. Part C: Emerg. Technol. 25, 81–99.
Fagerholt, K., Gausel, N.T., Rakke, J.G., Psaraftis, H.N., 2015. Maritime routing and Schinas, O., Stefanakos, C., 2013. The cost of SOx limits to marine operators; results
speed optimization with emission control areas. Transp. Res. Part C: Emerg. from exploring marine fuel prices. Int. J. Mar. Navig. Saf. Sea Transp. 7, 125–131.
Technol. 52, 57–73. Schinas, O., Stefanakos, C., 2014. Selecting technologies towards compliance with
Groene, O., 2015. Alternate fueled engine development and experience. Global MARPOL Annex VII: the perspective of operators. Transp. Res. Part D: Transp.
Maritime Hub 〈www.globalmaritimehub.com〉. Environ. 28, 28–40, Emission Control Areas and their Impact on Maritime
Hamburg Port Authority, 2013. Pressemitteilung 29.11.2012. Transport, http://dx.doi.org/10.1016/j.trd.2013.12.006.
HANSA, 2014. Schiffsbetriebskostenstudie 2014. HANSA International Maritime Semolinos, P., Olsen, G., Giacosa, A., 2011. Total LNG as Marine Fuel: Challenges to be
Journal Supplement. Overcome. Technical Report. TOTAL.
Holmgren, J., Nikopoulou, Z., Ramstedt, L., Woxenius, J., 2014. Modelling modal Smith, T.W.P., Jalkanen, J.P., Anderson, B.A., Corbett, J.J., Faber, J., Hanayama, S.,
choice effects of regulation on low-sulphur marine fuels in northern europe. O'Keeffe, E., Parker, S., Johansson, L., Aldous, L., Raucci, C., Traut, M., Ettinger, S.,
Transp. Res. Part D: Transp. Environ. 28, 62–73, Emission Control Areas and Nelissen, D., Lee, D.S., Ng, S.and Agrawal, A., Winebrake, J.J., Hoen, M.,
their Impact on Maritime Transport. Chesworth, S., Pandey, A., 2014. Third IMO GHG Study. Technical Report.
IGU, 2012. Natural gas conversion guide. Technical Report. International Gas Union. International Maritime Organization (IMO).
International Maritime Organization, 2014. Guidelines on the method of calculation Stefanakos, C., Schinas, O., 2015. Fuzzy time series forecasting of bunker prices;
of the attained energy efficiency design index (EEDI) for new ships. nonstationary considerations. WMU J. Marit. Aff.
International Maritime Organisation, 2015. Gas and low-flashpoint fuels code Stefanakos, C.N., Schinas, O., 2014. Forecasting bunker prices; a nonstationary,
adopted by imo; maritime safety committee (msc), 95th session, 3–12 June multivariate methodology. Transp. Res. Part C: Emerg. Technol. 38, 177–194.
2015. Stopford, M., 2009. Maritime Economics, 3rd edition. Routledge, London.
IPCC, 2007. Fourth assessment report: climate change. Technical Report. Inter- Sustainable Shipping Initiative (SSI), 2014. Financing sustainable shipping: save as
governmental Panel on Climate Change (IPCC). you sail.
Karakitsos, E., Varnavides, L., 2014. Maritime Economics. Palgrave MacMillan. http: Tay, E., 2012. 2012 Guide to Singapore Government Funding and Incentives for the
//dx.doi.org/10.1057/9781137383419. Environment.
Livanos, G.A., Theotokatos, G., Pagonis, D.N., 2014. Techno-economic investigation Thomson, H., Corbett, J.J., Winebrake, J.J., 2015. Natural gas as a marine fuel. Energy
of alternative propulsion plants for ferries and roro ships. Energy Convers. Policy 87, 153–167.
Manag. 79, 640–651. Vinkoert, D., 2012. Green trophy; eco ships awarded. Port Shipping Industry, 1,
Lloyd's Register, 2014. LNG Bunkering Infrastructure Survey. Technical Report. pp. 38–43.
Ma, H., Steernberg, K., Riera-Palou, X., Tait, N., 2012. Well-to-wake energy and Wang, S., Meng, Q., 2012. Sailing speed optimization for container ships in a liner
greenhouse gas analysis of SOx abatement options for the marine industry. shipping network. Transp. Res. Part E: Logist. Transp. Rev. 48, 701–714.
Transp. Res., Part D 17, 301–308. Wärtsilä, 2015. Wärtsilä solutions for marine and oil and gas markets.
Mellin, A., Rydhed, H., 2011. Swedish ports' attitudes towards regulations of the Yang, Z., Zhang, D., Caglayan, O., Jenkinson, I., Bonsall, S., Wang, J., Huang, M., Yan,
shipping sector's emissions of CO2. Marit. Policy Manag. 38, 437–450. http://dx. X., 2012. Selection of techniques for reducing shipping NOx and SOx emissions.
doi.org/10.1080/03088839.2011.588261. Transp. Res., Part D 17, 478–486.