Optimization of Liner Operations and Fuel Selection considering Emission Control Areas

The continuous expansion of shipping trade has brought about increasingly serious marine pollution problems. In the context of emission reduction in the global shipping industry, this paper focuses on the operation optimization of container ships inside and outside the emission control area (ECA). From the dual perspectives of shipowners and the general public, models in the annual operating cycle are established to study the economic and environmental benefit differences between traditional fuels, i.e., heavy fuel oil (HFO) and low-sulfur fuel oil (LSFO), and alternative fuels, i.e., liquefied natural gas (LNG) and methanol. Sensitivity analysis was carried out for the proportion of ECA and ship speed. The results show that, in the current situation of high natural gas prices, the use of HFO after the installation of scrubbers is still the most cost-effective option in the short term, followed by the use of LSFO and methanol. LNG is no longer an attractive option, while LSFO and methanol are the best options for both cost and the environment. With the tightening of ECA regulations, methanol will become the optimal choice when the ECA ratio is higher than 47%. By reducing the speed of the ship, the pollutant emission can be effectively reduced, but it will also lead to an overall decrease in profits. Considering the future “zero carbon” emission targets, slow streaming is only suitable as a short-term response measure, while switching to green power energy is a choice that is more in line with the long-term development strategy.


Introduction
International shipping, which serves as the main conduit for products moving across borders, transports more than 90% of goods traded globally and has made a great contribution to the growth of the world economy. However, due to the diverse, protracted, and fexible nature of ship transportation activities, the resulting large amount of air pollutants will cause serious environmental pollution in ports and coastal areas [1]. In 2018, the total emissions of the international shipping industry reached 1.056 billion tons of carbon dioxide equivalent, accounting for about 2.89% of the annual greenhouse gas emissions [2]. If no measures are taken, the total emissions are projected to increase by 50% to 250% in 2050 [3].
In this context, as the main governance bodies, various international organizations and sovereign countries have successively issued relevant policies to limit emissions from the shipping industry. Te International Maritime Organization (IMO) has been dedicated to reducing carbon emissions in the global shipping sector since 2011. Te most infuential one is the MARPOL 73/78 Convention, which has been ratifed by more than 160 countries. Te convention includes six technical annexes, each of which contains detailed provisions on specifc categories of pollution from ships. Among them, Annex VI imposes restrictions on the emission of pollutants such as nitrogen oxides (NOx), sulfur oxides (SOx), and greenhouse gases (GHGs) [4]. IMO has also set the Baltic Sea, the North Sea, North America, and the Caribbean Sea as emission control areas (ECAs), stipulating that from January 1, 2020, the global sulfur limit for ships' fuel oil will be reduced from 3.5% to 0.5%, while the upper limit within the ECA will be reduced from 0.5% to 0.1%, which are clearly listed in Table 1.
In addition, there is intense pressure on the shipping sector to reduce carbon emissions. With an initial strategy to reduce greenhouse gas emissions from ships, the IMO's Marine Environment Protection Committee has ambitiously joined the global quest for a path to decarbonization, with the following vision: International shipping's carbon intensity will decrease by at least 40% by 2030 compared to 2008, with a goal of decreasing by 70% by 2050. By 2050, the greenhouse gas emissions from international shipping will be reduced by at least 50% compared with 2008 and will reach the peak of emissions [5]. At the same time, the Chinese government also formally put forward the carbon peaking and carbon neutrality goals at the United Nations General Assembly in September 2020, striving to achieve carbon peak by 2030 and carbon neutrality by 2060 [6]. Under such increasingly stringent emission requirements, it is of great signifcance to study the choice of fuels in the context of dividing ECA. In order to provide a reference for the decision-making of various stakeholders and to some extent encourage the sustainable development of the global shipping sector, this paper evaluates alternative fuels with both economic and environmental benefts.
Te rest of the article is organized as follows: Section 2 lists the current major emission reduction measures in the shipping industry and introduces several major types of exhaust gas after-treatment technologies and common marine fuels. Section 3 establishes a mathematical model of pollutant emissions, private costs, and social costs of different fuels in the annual operating cycle from the dual perspectives of shipowners and the public. Section 4 selects a specifc route to conduct an example analysis. Section 5 shows a sensitivity analysis conducted on the proportion of emission control areas and the speed of ships. Section 6 summarizes the paper and provides a set of concluding remarks.

Emission Reduction Measures
As the global climate continues to deteriorate, the corresponding emission reduction policies are gradually tightening, and the operations of shipping companies will be directly afected. Currently, the methods to deal with emission restrictions mainly include four types: improving technical design, optimizing operation methods, using green power energy, and introducing market mechanisms.
Technical design measures are primarily aimed at improving the hull design and optimizing the engine system, such as the bulbous bow hull design, waste heat recovery system, and exhaust gas after-treatment device so as to maximize propulsion efciency and reduce pollutant emissions. Among them, exhaust after-treatment devices mainly include scrubbers and selective catalytic reduction (SCR) devices, which can signifcantly reduce NOx and SOx emissions with little impact on engine performance and fuel economy [7]. SCR technology can achieve 90%-95% NOx reduction [8,9], making it the most efcient NOx emission reduction method, while the desulfurization efciency of scrubbers can be as high as 97% [10], allowing ships to continue to use heavy fuel oil (HFO). Scrubbers remain an attractive emission reduction option for shipowners given the current high price trends for low-sulfur fuel oil (LSFO) and liquefed natural gas (LNG). Some studies have comparatively evaluated the economic and environmental benefts of installing after-treatment devices and fuelswitching measures on ships [11][12][13][14], as well as changes in options as factors such as fuel prices [15,16], emission regulations [17], and government subsidies [18].
Te improvement measures of the operation plan primarily include the optimization of navigation routes, speed, fuel replenishment strategy, and feet structure. Many researchers have conducted studies from various perspectives and have extensively verifed the efectiveness of the deceleration ship scheduling method [19][20][21].
Low-carbon fuel is an efective way to reduce the carbon footprint, and zero-carbon fuel may be the main means to achieve carbon neutrality. Alternative fuels that have been considered in the research mainly include LNG, methanol, hydrogen, ammonia, liquefed biogas, and biofuels. In addition, a series of green power systems such as fuel cells, wind energy, solar energy, and nuclear energy have been gradually developed [22]. Although the most important fuel for shipping is still HFO, more and more shipowners have chosen to install or reserve alternative fuel systems for new buildings. In the frst quarter of 2022, a total of 61% of the total tonnage of newbuilding orders can use alternative fuels, of which 57% of the orders use LNG fuel and 3.4% of the orders use methanol fuel [23]. Compared with traditional marine fuels, the carbon emission reduction potential of LNG and methanol is about 20%-40% [24][25][26], and the overall potential of alternative fuels to reduce SOx and NOx emissions can reach 60-90% and 80-85%, respectively [9,26]. In addition, market mechanisms such as carbon tax and carbon trading can achieve the purpose of restriction by increasing the cost of carbon emissions, which will become one of the important ways for the low-carbon governance in the future.
Based on the above research, this study comprehensively considers the newbuilding orders in recent years and the physical and chemical properties of diferent fuels (see Table 2), retains HFO, MGO, and VLSFO as traditional fuel options, and selects LNG and methanol as alternative fuel options. Taking the following four emission reduction measures as evaluation scenarios, a mathematical model was established innovatively from the dual perspectives of shipowners and the public, and the annual pollution emissions and costs of various fuel-powered container ships were calculated, evaluating their economic and environmental benefts. Journal of Environmental and Public Health Scenario 1: MGO (0.1%) and HFO (3.5%) are used inside and outside the ECA, respectively, and scrubbers and SCR equipment are used throughout Scenario 2: MGO (0.1%) and VLSFO (0.5%) are used inside and outside the ECA, respectively, using SCR equipment throughout Scenario 3: LNG is used as power throughout the process, and SCR equipment is used Scenario 4: methanol is used as power throughout the process, and SCR equipment is used

Mathematical Model
Te model established in this paper is based on the following assumptions: (1) Te auxiliary engine uses MGO throughout the process and does not switch with the main engine fuel. (2) Te cost of new container ships for LNG and methanol is 20% and 15% [35] higher than that of traditional fuel ships, respectively (3) Te ratio of the ship to the sailing mode and the port berthing mode is 9 : 1 [36], and the average speed in the sailing mode is 18 knots (4) In order to meet the strict NOx emission requirements, SCR equipment is required to be used throughout the process in all scenarios (5) Te average service life of ships and SCR equipment is 25 years, and the average service life of scrubbers is 15 years [37] 3.1. Fuel Consumption. Te fuel consumption of the ship is estimated by dividing it into two operating modes: sailing and port berthing. Te main engine load in the sailing mode is the cubic ratio of the actual speed and the design speed, while the auxiliary engine load is generally considered to be independent of speed, and the value in this paper is 0.5 [38].
In the port berthing mode, only one auxiliary engine is reserved. Te fuel consumption of the main and auxiliary engines is calculated by where F M and F A are the fuel consumption rate (t/h) of the main and auxiliary engines per unit time, respectively, when the ship is sailing; SFOC M and SFOC A are the specifc fuel consumption factors (g/kWh) of the main and auxiliary engines, respectively; according to the existing research, this paper takes 196 g HFO/kWh and 216.7 g MGO/kWh, respectively [39]; EL M and EL A are the load factors of the main and auxiliary engines respectively; and P design M and P design A are the design rated power (kW) of the main engine and the auxiliary engine, respectively.
When using MGO and HFO inside and outside the ECA, respectively, the fuel oil consumed throughout the voyage is calculated as follows: where F HFO and F MGO are the total consumption (t) of HFO and MGO, respectively; F M MGO and F A MGO are the MGO consumption (t) of the main and auxiliary engines, respectively, and the sum of the two is the total consumption of MGO; V d is the ship's design speed and V is the actual sailing speed (kn); D I and D O are the sailing distance (n mile) of the ship inside and outside the ECA, respectively, while D is the total sailing distance (n mile); n A is the number of auxiliary engines; and LHV HFO and LHV MGO are the low heating value (kJ/kg) of HFO and MGO, respectively, which can be used to convert the mass of diferent fuels under the same heat release.
Similarly, when VLSFO is used outside the ECA, the MGO consumption remains unchanged, and the fuel consumption of VLSFO can be expressed as When using alternative fuels such as LNG or methanol, MGO consumed by the auxiliary engines remains unchanged, and the fuel consumption of the main engine can be expressed as where F VLS and F ALT are the consumption of VLSFO and alternative fuel (LNG or methanol), respectively (t), and LHV VLS and LHV ALT are the low heating value (kJ/kg) of the corresponding fuel.

Pollutant
Emissions. Te ten pollutant emissions reported by the IMO include CO 2 , CH 4 , N 2 O, CO, SOx, NOx, PM, NMVOC, and BC [2]. Tis paper selects three main greenhouse gases CO 2 , CH 4 , and N 2 O and three atmospheric pollutants SOx, NOx, and PM as emissions for quantitative evaluation. Pollutant emissions are represented by the product of fuel consumption and the corresponding emission factor, namely, where E ij is the mass (t) of the emission i produced by the combustion of the fuel j, F j is the consumption (t) of the fuel j, LHV j is the low heating value of the fuel j (kJ/kg), and EF ij is the emission factor (g/MJ) for the emission i of the fuel j. Tis paper assumes that the use of SCR and scrubbers can reduce NOx and SOx emissions by 90% and 95%, respectively. In addition, due to the limitations of current research on alternative fuels, some emission factors for methanol are not available. Since methanol does not contain nitrogen oxides and sulfur, we assumed that this part of the emissions is zero in the subsequent calculation. Te emission factors for various fuels are shown in Table 3.

Annual Cost and Proft.
Te number of ships operating on a route in a year can be expressed as where the mathematical symbol [] Floor means the foor function. Te costs involved in the operation of ships mainly include three parts: capital cost, operating cost, and fuel cost. Te capital cost mainly includes the construction and installation cost of the new building and related equipment. Generally, the depreciation method of the average service life of the ship is adopted; i.e., the total investment cost is divided by the service life. Te operating costs are the costs incurred by shipping companies to maintain normal shipping services, including equipment operating costs, crew wages, maintenance costs, and insurance and management fees, and port fees, loading and unloading fees, anchoring fees, etc., are generally 15%-50% of the capital cost [45,46], and the compromise value in this paper is 30%. Te fuel cost is afected by various conditions such as ship size, speed, and fuel and can be expressed as the product of fuel consumption, price, and number of sailings. For shipowners, the choice of an emission reduction plan mainly depends on the trade-of between capital expenditure (CAPEX) and operating expenditure (OPEX) [47].
Te private cost and its subdivision cost ($) can be calculated as where PC is the private cost of the shipping company, C FUEL is the fuel cost, and C CAPEX and C OPEX are the capital cost and operating cost, respectively, which can be seen in Table 4; C ship and C scrubber are the investment cost of the new building and scrubber, respectively; F and P are the annual consumption (t) and price of the fuel ($/t); s is a 0-1 variable; if 0 is taken, it means that the scrubber does not need to be installed in this scenario, and if 1 is taken, it means that the scrubber needs to be installed. Te net proft of a single voyage can be expressed as the diference between the total revenue and the total private cost: where n is the number of voyages in one year as required above, CL is the container ship's load capacity (TEU), and r is the utilization rate of the space. In the situation where it is difcult to get one cabin, this paper takes the value of 1.
According to the statistics of USDA [50], we take the average freight for all kinds (FAK) in the frst half of this year as 1270 $/TEU.

Social Cost.
Under the same conditions, shipowners and the public tend to prefer diferent optimal emission reduction options. Driven by economic interests, shipowners usually only pay attention to the private cost when they choose schemes, and their goal is to maximize the total profts on the basis of meeting the minimum emission requirements, while for the public, they are more concerned about whether the option reaches the best balance between economy and environment. Terefore, they need to consider the social cost of emissions, which can be expressed as the  Journal of Environmental and Public Health sum of the emissions of various pollutants multiplied by the corresponding cost factors: where SC is the social cost ($), E ij is the emission i of the fuel j calculated above (t), and CF ij is the social cost factor for the emission i of the fuel j ($/t). Te social cost factors of different pollutants are shown in Table 5.  Table 6.

Case Analysis
In 2020, the average price of LNG will remain around 400$/t, which is about 80% of the price of MGO fuel. Compared with expensive low-sulfur fuel, LNG has great price advantage and is considered to be a better emission reduction solution than other fuels in terms of economic and environmental efect [35,43]. However, since the second half of 2021, the market price of natural gas has risen rapidly and the supply has become tight, and the confict between Russia and Ukraine in 2022 further exacerbated this trend. Within a year and a half, the average price of LNG has risen to four times the original price, and the peak price even reached more than 3,000 $/t. At these prices, it is worth considering whether LNG ships are still a credible option to reduce emissions.
Based on the statistics of Ship & Bunker, this paper takes the average price of HFO, VLSFO, MGO, and LNG in Rotterdam Port from January to June 2022 as the benchmark price, which are 664, 884, 1120, and 1783 $/t, respectively; based on the statistical data of Methanex, this paper takes the mean value of the nondiscounted reference price of methanol in the same time period as the benchmark price, i.e., 630 $/t, and substitutes it into the mathematical model established above.

Pollutant Emissions.
As can be seen from Figure 1, before the application of exhaust after-treatment equipment, the emission of pollutants from alternative fuels is signifcantly lower than that of traditional fuels. However, it is worth noting that when LNG is used as fuel, there is serious methane leakage, and the methane emission can be nearly 300 times that of ordinary fuel oil and more than 1000 times that of alternative clean fuels. Terefore, LNG may not be an efective scheme to mitigate climate change under comprehensive consideration.
Compared with HFO, alternative fuels can reduce SOx emissions by more than 99%, but the CO 2 emissions of LNG and methanol do not have obvious advantages compared with traditional fuels. After switching to LNG and methanol, CO 2 emissions can only be reduced by about 21.7% and 7.5%, respectively. After the application of scrubbers and SCR equipment, the SOx and NOx emission levels of traditional fuels are reduced to the emission range of alternative fuels, which is an efective emission reduction measure. Terefore, for shipowners, the choice of the best emission reduction scheme will depend on the trade-of between the installation and operating costs of diferent emission reduction equipment and the cost of switching fuel.

Cost.
For shipowners, when making the choice of the best emission reduction scheme, the ultimate reference is the private cost. It can be seen from Figure 2 that, due to the absolute advantage of traditional fuel oil in price, the use of MGO and HFO inside and outside the ECA, respectively, is still the best choice for shipowners under the condition that exhaust after-treatment equipment is installed to meet the emission requirements, followed by the use of MGO and VLSFO inside and outside the ECA and the use of methanol. Te skyrocketing price of LNG makes it no longer an attractive option.
When making decisions, the public should comprehensively consider private costs and social costs and make choices that have both economic and environmental benefts. It can be seen that the best solution at present is to use MGO and VLSFO inside and outside the ECA, followed by methanol as power. LNG can gain cost advantages over VLSFO only when the price drops by more than 40% and can obtain comprehensive advantages over methanol and VLSFO when the price drops by about 14 and 30%, respectively, which may not be possible in the short term.  Journal of Environmental and Public Health 5

Ratio of ECA.
More and more countries and regions have begun to divide the ECA and expand the scope of the existing ECA. With the increasingly tightened emission regulations, it is necessary to consider the impact of the proportion of ECA on emission reduction options. Tis paper calculates the changes in pollutant emissions, costs, and benefts when the time proportion of ships sailing in ECA increases from 2% to 20% in increments of 2%. Since both LNG and methanol are used in the whole process, the ratio of ECA has no efect on them. Meanwhile, the change in the ECA proportion has no impact on the single voyage time of the ship, so the total operating income remains the same, and the scheme with the lowest private cost is the scheme with the highest proft.
In terms of private costs, traditional fuel has great advantages. As shown in Figure 3, the use of MGO and HFO inside and outside the ECA, respectively, is the best choice, regardless of the proportion of the ECA. But after considering the social cost, when the proportion of the ECA is less than 46%, the best choice is VLSFO, methanol, and HFO in turn. When it is higher than 47%, methanol will be the best choice, followed by VLSFO and HFO, but when it rises to 58%, HFO will become better than VLSFO. Only when the proportion of the ECA is greater than 82%, LNG can have certain advantages over VLSFO, but the cost is still far higher than that of methanol. However, if VLSFO ships are also equipped with scrubbers, the environmental benefts will far exceed the investment and operating costs of scrubbers, and it will always be the best choice for shipowners when the ECA ratio is not more than 85%.
Extending this conclusion to other ships, it can be seen that if a container ship of the same size carries out shortrange ocean transportation, when the proportion of its sailing in ECA exceeds 46%, it can be considered switching to use methanol as power. When the proportion is less than 85%, using VLSFO on the basis of adding scrubbers will bring the greatest proft.

Speed.
In order to see the impact of speed on operating costs and emissions more intuitively, the range of speed is set to be 12-20 knots, and the pollutant emissions and costs are calculated separately. As the speed decreases, the load of the ship's engine gradually decreases, the corresponding fuel consumption and the emission of various pollutants will be reduced, and the overall fuel cost and operating cost of the corresponding solutions will drop. Under the condition of constant investment cost, both private and social costs show a decreasing trend. However, due to the decrease in speed, the operating voyages in one year will be reduced, and the annual proft at diferent speeds is shown in Figure 4.
Under the balance between the reduction of voyage benefts and the reduction of private costs, although the overall proft shows a gradual downward trend, there is an optimal speed in each speed segment, which enables the shipowner to obtain the highest proft, which is about 18.9    Journal of Environmental and Public Health knots, 15.7 knots, and 12.6 knots, of which 18.9 knots can be used to maximize profts. For the public, after taking the emission cost into consideration, it can be seen from Figure 5 that VLSFO is the most economical and environmentally benefcial choice at diferent speeds. When the speed is greater than 14 knots, methanol is the second-best choice; but with the decrease in speed, the price advantage of HFO will be more prominent, and its total cost will be lower than that of methanol.

Conclusion
Shipowners must consider diferent emission reduction solutions when placing orders for new ships in order to comply with increasingly stringent international conventions and regulations, including continuing to use HFO after adding scrubbers and switching to low-sulfur fuel oil and cleaner alternative fuels such as LNG and methanol.
Tis paper assesses the pollutant emissions of various fuels in the annual operation cycle of ships, taking into account both the economic and environmental benefts. A private and social cost model was developed to assess the best mitigation options in the context of current fuel prices. Te study takes into account how compliance decisions made by shipowners and the general public alter under various ECA ratios and speed situations, and the following results are drawn: (1) VLSFO, LNG, and methanol can efectively reduce SOx emissions by more than 90% compared with HFO, but LSFO and methanol can only reduce CO 2 emissions by 2% and 8%, respectively, and have no obvious advantages in CO 2 emission reduction. (2) Te most economical option at the moment is to continue using HFO after installing scrubbers, but converting to LSFO will be preferable when taking into account the social cost of emissions. (3) When the ECA ratio is greater than 47%, methanol will become the best choice for both the environment and the cost. At the same time, reducing the speed of sailing is indeed one of the efective measures to reduce emissions in the short term, although it will reduce the annual proft.
Te service life of ships is often between 20 and 30 years, and the choice of a power system will have a long-term impact on the future environmental climate. Considering the number of new ship orders, we focused on container ships as research objects, but similar methods can be extended to other ship types, such as dry bulk carriers and oil tankers, providing useful reference for various stakeholders to make dynamic decisions through horizontal comparisons, reduce cost, and maintain market competitiveness in the context of tightening environmental policies and changing fuel prices. In addition, it should be pointed out that this paper substitutes the historical average data of fuel prices into the calculation, but considering the instantaneous changes in fuel prices, it will be possible to update fuel prices, fuel consumption, and other parameters in real time and expand the existing model into a dynamic decision-making model in the next step.
When it comes to choosing a long-term sustainable alternative fuel, there is no single answer as to which is the best choice. It is determined by a number of elements, including various operation modes, the scope of ECA laws, and the principal authority making the decision. However, it     is apparent that the transition to cleaner and more efcient fuels will be an unavoidable trend in the future.

Data Availability
Te data used to support the fndings of this study are included within the article.

Conflicts of Interest
Te authors declare that there are no conficts of interest.