Understanding Seafarers’ Acceptance of the Transition to Alternative Fuels in Shipping Through the Technology Acceptance Model

Abstract

According to the International Maritime Organization, transitioning to alternative fuels is essential to achieving net-zero carbon emissions. Seafarers are at the frontline of this transition, and in this study, their attitude toward this strategy is analyzed using the technology acceptance model. The alternative fuels analyzed are liquefied natural gas (LNG) and methanol as short-term options and hydrogen and ammonia as long-term options. The analyzed seafarers are from South Korea, where alternative fuels are actively incorporated into shipbuilding and training. Across all fuels, perceived ease of use (PEOU) positively affected perceived usefulness (PU). PEOU and PU positively influenced attitude toward using (ATT). ATT and trust (TRU) significantly increased behavioral intention (BI), a finding that aligns with those of prior studies, while operational safety risk (OSR) also showed a positive effect on ATT. This indicates that seafarers became more aware of the need to use alternative fuels and expected improvements in managing related risks. Unlike OSR, environmental risk (ER) negatively affected ATT for hydrogen, consistent with prior risk perception studies. These findings suggest that to encourage alternative fuel use during the shipping industry’s energy transition, operational ease, enhanced risk management systems, and basic competency training and incentives are necessary to positively shape seafarers’ attitudes and behavioral intentions.

1. Introduction

The international shipping industry is navigating tremendous changes in response to regulatory pressures and societal demands for decarbonization [1,2,3,4,5]. The International Maritime Organization (IMO) revised its greenhouse gas (GHG) strategy by setting a target to achieve net-zero GHG emissions by 2050 [6]. While such policies and societal shifts are essential within the decarbonization framework, they must consider diverse challenges encountered across technological, economic, and safety dimensions [4,7,8,9]. In alignment with the IMO’s GHG strategy, the industry is currently reducing its carbon intensity through short-term efficiency measures such as route optimization, slow steaming, and the implementation of energy-saving devices. However, the long-term reduction targets would be difficult to achieve without transitioning to alternative fuels. Therefore, fuels such as liquefied natural gas (LNG), methanol, hydrogen, and ammonia are being considered to replace fossil fuels.

In the short term, LNG and methanol have certain advantages, as most ships are now built with the capacity to operate on these two fuels [10]. In the long term, ammonia and hydrogen are considered strong candidates for achieving complete carbon reduction. These four alternatives have unique characteristics when used as marine fuels. LNG exhibits cryogenic properties. In the event of a leak, it poses risks of explosion and fire, brittle fracture, and asphyxiation [11,12]. To prevent these hazards, storage tank insulation must be reinforced, and additional equipment is required to prevent low temperatures, fires, and explosions [13]. Methanol is a flammable liquid that poses a risk of explosion and is toxic when inhaled. Owing to these hazards, vessels must be adequately equipped with special internal coatings and ventilation systems, along with gas detection and shut-off devices, and flammable zones must be designated [14,15]. Ammonia has low flammability but is highly toxic and corrosive, and mass exposure can damage the hull and even cause loss of life [16]. Consequently, double barriers and piping are required, along with high-efficiency ventilation and detection equipment and automatic shutoff systems [17,18]. Hydrogen has an extremely wide flammability range, is ultralight, disperses rapidly into the atmosphere, and exhibits cryogenic properties, posing a high risk of explosion upon leakage. These characteristics necessitate the use of insulated tanks, gas detectors, and emergency ventilation systems; the isolation of hazardous zones; and the installation of explosion and safety sensors [19,20,21,22].

Each type of alternative fuel has unique advantages and disadvantages [23,24,25] and distinct characteristics, necessitating specialized vessel structures and equipment. Consequently, the associated additional capital expenditures and operating costs vary significantly. This complicates decision-making for several stakeholders in the shipping industry. In particular, during shipbuilding, maritime transport operators are faced with decisions that will profoundly affect their companies’ profitability for decades. Therefore, most research on alternative fuel conversion has largely focused on economic viability and regulatory compliance from the perspectives of shipowners, policymakers, and shipyards.

The transition to alternative fuels necessitates that seafarers, as on-site operators, demonstrate the understanding, proficiency, risk perception, and regulatory compliance required to safely introduce and use these fuels, as they are directly exposed to potential hazards during ship operation. To ensure the safety of crew members, the IMO has proposed various interim measures for handling alternative fuels [26,27,28,29]. Therefore, it is crucial that decision makers at the companies and organizations responsible for shipbuilding and operation understand the degree of crew acceptance and safety awareness regarding alternative fuel use. Nevertheless, studies analyzing the perceptions and attitudes of seafarers—the actual users at the forefront of the transition to alternative fuels—remain relatively scarce. Seafarers’ perceptions of and attitudes toward eco-friendly conversion are important. Developing strategies that include active participation and education can ensure the sustainability of decarbonization [30,31].

Specifically, how seafarers’ perceptions of fuels with varying technological maturity and risk profiles, such as LNG, methanol, ammonia, and hydrogen, differ and whether these differences correlate with resistance or facilitation of the adoption of alternative fuels have not yet been systematically determined.

Thus, in this study, the factors and relationships associated with seafarers’ acceptance of the transition to alternative fuels were empirically analyzed using an established model—the technology acceptance model (TAM)—pertaining to technology adoption. Differences between the fuel categories were examined through group comparisons, through which it was observed that the theoretical structure demonstrates different path strengths and directions depending on the fuel type. The TAM posits that users’ attitudes toward a specific technology are formed according to perceived usefulness (PU) and perceived ease of use (PEOU), which then influence user behavior [32,33]. The TAM is a perception survey targeting actual technology adopters, and most discussion has centered on practical technology users. For these reasons, the TAM has been widely used in information technology adoption research, where technology is accepted relatively easily at the general user level [34,35]. As information technology has advanced, this model has been extended and applied to healthcare-related research [36,37,38], commercial transaction studies related to IT development [39], and mobile-related research [40]. A recurrent theme in these studies is the tendency to prioritize empirical investigations of technology users. The technological transition in the shipping industry also necessitates an approach from this perspective. Seafarers can be considered the end users of new technologies in the shipping sector, rendering them suitable for the application of the Technology Acceptance Model (TAM). Consequently, an analysis of research related to the transition to alternative fuels in the shipping industry through a seafarer-centered Technology Acceptance Model (TAM) can be expected to yield structural similarity and scalability.

Application of the TAM has been limited in the maritime sector, where it is primarily used in studies on autonomous navigation [41,42], and there has been a lack of focus on technologies for converting ships to alternative fuels. Moreover, research involving seafarers—the actual users of the technology—has been insufficient. As seafarers’ safety awareness and risk perception are essential to the application of new technologies in the maritime sector, this research is particularly important [43,44,45]. Therefore, we incorporated risk factors that represent important variables in real ship-operating environments into the general TAM. We comprehensively investigated how the transition to alternative fuels affects the technology acceptance of seafarers using an expanded TAM that incorporates risk perception. In particular, the technical acceptance of alternative fuel systems by seafarers in the marine engineering field influences improvements in the technology’s reliability, safety, and efficiency. Moreover, since many alternative fuels require separate control interfaces, operating rules, or risk mitigation procedures, evaluating user-level acceptance can also contribute to the system’s engineering implementation. Therefore, this study provides theoretical insights into technology in this field and contributes to the adoption of timely, human-centered technology acceptance strategies. This enables strategic and policy improvements toward achieving net-zero goals.

2. Literature Review

2.1. Global Policy and Regulatory Frameworks for Alternative Fuel Transition

The IMO aims to achieve net-zero carbon emissions from shipping by 2050 through its revised GHG reduction strategy, which was adopted in 2023. To this end, the IMO views the global introduction of alternative fuels and energy sources for shipping as essential and expects all types of fuels to be considered [6]. Indeed, the IMO anticipates that the adoption of alternative fuels will account for 64% of the total carbon reduction achieved by 2050 [46]. The IMO recognizes that this target will not be accomplished through technical policy regulations alone. Thus, the need for measures that encompass technological and economic elements has increased. Accordingly, although the 83rd session of the Marine Environment Protection Committee initially intended to approve and implement the revised MARPOL (International Convention for the Prevention of Pollution from Ships) Annex VI—which introduces a net-zero framework incorporating economic measures—the decision was postponed for one year at the second extraordinary session, and the future implementation plan will be reconsidered [47]. Despite ongoing debates over the economic implications of these alternative fuel–related policies, the IMO is still expected to advance their implementation. Notably, this regulation extends the scope of carbon-emission measurements to include the production and supply stages of fuel, making it similar to the EU’s FuelEU Maritime Regulation. Therefore, it is expected to accelerate the shipping industry’s transition to alternative fuels for decarbonization.

In 2024, the EU began including all ships over 5000 gross tons in the EU Emissions Trading System, mandating the submission of carbon allowances and aiming to achieve 100% coverage of reported emissions in 2027 [48]. Driven by these regulatory changes, orders for newbuilds using alternative fuels increased by 38% in 2024 compared with the previous year, extending beyond LNG-powered vessels to include orders for ammonia-fueled ships [49].

2.2. Technological Measures and Seafarer Acceptance in the Transition to Alternative Fuels

In order to achieve the IMO’s net-zero goals, the shipping industry is implementing various measures more proactively than other sectors. This is partly because the efficient use of energy for carbon reduction can support economic efficiency in addition to helping achieve carbon targets. The implementation of these carbon reduction measures can be categorized according to the temporal parameters of the project.

First, there are measures that can be implemented in the short term. A prime example of a short-term measure is the introduction of optimization approaches that utilize digital technology. A range of big data analysis methods is being implemented to manage ship energy efficiency and predict pollutant emissions. The combination of intelligent algorithms and machine learning has been demonstrated to enhance predictive performance and improve energy efficiency [50]. It is anticipated that advancements in technological capabilities, encompassing these digital technologies, will engender efficiency gains in technical and operational domains, including operational efficiency [51], energy efficiency [52], power system efficiency [53], and hull efficiency [54]. The evaluation of such efficiency improvements is primarily conducted on a short-term basis because these measures can be implemented immediately. It is evident that the IMO’s short-term measures, EEXI and CII, also focus on this aspect.

In the medium term, measures to implement include the application or ongoing development of liquefied natural gas (LNG), methanol, biofuels, support for wind power, and the electrification of ships. The analysis of long-term measures is primarily focused on the utilization of hydrogen, ammonia, nuclear power, and carbon capture technologies [55]. The rationale behind this assertion is that the carbon reduction capacity of short-term measures is limited. Consequently, in addition to technical and operational measures, the mid-to-long-term transition to alternative fuels is an essential means of achieving the IMO’s net-zero goal [56].

Among alternative fuels, LNG, methanol, hydrogen, and ammonia are considered to be of particular significance. Nevertheless, there are still deficiencies in the realm of safety management technologies for these fuels, which have a direct impact on the well-being of crews and the preservation of the environment in the context of the transition to alternative fuels [57]. Although awareness of seafarers’ acceptance of technology is crucial in decisions related to this transition, there is a risk that this group may be marginalized when such decisions are made [58]. Indeed, an analysis of seafarers’ awareness regarding climate change revealed that they have a low level of understanding about international policy changes related to climate change, indicating that efforts are needed to increase their awareness [59]. In particular, even during the implementation of short-term energy efficiency measures such as SEEMP, seafarers experienced technological changes and demonstrated efforts to achieve company goals, despite the fact that financial incentives and other benefits were found to be severely lacking [60]. Financial incentives for energy efficiency and low-carbon shipping have been identified as potential motivation for seafarers to actively participate in these policy changes. Educational level has also been identified as a significant factor in participation [30].

2.3. Components of the TAM

Anchored in the concepts of PU and PEOU, the TAM is a framework for understanding technology acceptance. PU refers to the degree to which a user believes that using a particular system will enhance their performance, whereas PEOU refers to the degree to which a user believes that using a particular technology will require little effort [32]. PEOU directly influences PU, and they jointly influence attitude. Changes in attitude can alter BI, revealing the degree of technology acceptance. The TAM has generally been used to study the adoption of new, previously unavailable technologies, such as the Internet and digital services [61,62]. Recently, with the integration of fintech and technological changes in banking systems, the TAM has been widely applied to assess the adoption of digital banking system technologies [63,64]. Furthermore, with increasing environmental concerns and eco-friendly policy trends demanding diverse technological changes, the TAM has been applied to adoption of technologies related to energy use by various stakeholders [3,65,66,67,68,69,70,71,72,73,74,75].

In the shipping sector, the TAM has been applied to autonomous navigation—a next-generation technology [42]. It is also being applied to technological changes driven by digital transformation in the logistics sector, including shipping [76,77]. Efforts are being made to apply the TAM across all fields related to technological change to assess the degree of technology acceptance among users.

2.4. Risk and Trust Constructs

Seafarers are exposed to various risks while onboard a ship and using fuel. Alternative fuels adopted for decarbonization carry greater risks than conventional marine heavy fuel oil in certain circumstances. Uncertainty regarding the risks associated with alternative fuels may require incorporating trust and risk factors into the TAM [39]. Trust has been used as a key factor influencing the acceptance of online commerce technologies, such as those relating to e-commerce and IT [78,79]. Similarly, trust is crucial to the adoption of new and sustainable energy technologies [66,80]. Therefore, trust factors need to be incorporated into the TAM to understand seafarers’ adoption of alternative fuel technologies. Users’ risk perceptions are also important. Models incorporating risk perception factors have primarily been applied to the Internet and mobile systems to analyze the perceived risks associated with potential hazards arising from using these technologies [39,73,75,81]. Variables related to each individual’s risk have been modified and used in each model. Generally, perceived risk is considered to indirectly influence acceptance intention through mediating variables [80]. This study aimed to construct a model by distinguishing risks rather than using integrated perceived risk perception.

2.5. Trends in Research on the Acceptance of Energy Conversion Technologies

Research on the acceptance of technologies related to energy conversion, including alternative fuels, has undergone various developments. In general, these studies have focused on areas in which consumers have relatively broad decision-making authority, with vehicle fuel conversion technologies being a representative example. Accordingly, studies applying the TAM have primarily analyzed factors influencing the adoption of new energy conversion technologies for vehicles [68,70,75]. Some of these studies have also extended the basic TAM constructs by incorporating esthetic, functional, and symbolic components that reflect the preferences of consumers [70]. Consistent with typical TAM findings, PU and PEOU were found to have positive effects on attitude and behavioral intention, whereas perceived risk exerted a negative effect.

Research on alternative fuel technologies has expanded beyond this consumer-oriented vehicle domain into energy management and infrastructure technologies, including hydrogen, biofuels, battery storage systems, and smart grid technologies [72,73]. While these studies demonstrated results similar to those attained with conventional TAM frameworks, some [73] suggested that risk perception should be interpreted based on consumer risk profiles rather than simply minimized.

Researchers have also begun to explore technology acceptance for alternative fuel applications within specific industries, such as agricultural machinery [74]. At the same time, the research has broadened to encompass energy conversion across the transportation sector [3,65]. These studies generally examine technology acceptance from the perspective of stakeholders involved in alternative fuel transitions rather than ordinary consumers. Notably, in the case of shipping research [3], unlike typical TAM findings, risk perception was found to positively influence PU—indicating that awareness of potential risks may enhance the perceived value of adopting alternative fuels by highlighting the scope for improvement within a specialized field. Building on these prior studies, this research focuses on seafarers, who comprise one of the key stakeholder groups in this specialized domain. The relevant studies and key factors influencing the acceptance of energy conversion technologies are summarized in Table 1.

Table 1. Summary of technology acceptance studies in energy conversion.

Ref.	Research Focus	Data Source	Key Determinants
[72]	Hydrogen fuel stations, biofuel production plants, and stationary battery storage facilities	German public	Perceived problems, Trust in industry, Trust in municipalities, perceived effects, environmental identity, affect, general acceptance, local acceptance
[3]	LNG and electricity as alternative fuels	Greek stakeholders	Relative advantage, compatibility, complexity, trialability, observability, safety procedures and risk, governance policy, infrastructure, bunkering and electrification, economic and financial concerns, port regulation, technological concerns, availability, public trust, perceived utility, perceived usefulness, perceived ease of use, attitude toward using, behavioral intention to use, actual use
[73]	Smart grid technologies	Romanian public	Perceived compatibility, perceived understanding, perceived power supply reliability, perceived electricity rate saving, perceived eco-environment, perceived cyber insecurity, perceived electromagnetic radiation, perceived ease of use, perceived usefulness, perceived risk, intention to use smart grid technologies
[75]	New energy vehicles (NEVs)	Chinese public	Perceived usefulness, perceived ease of use, perceived risk, attitudinal ambivalence, purchase intention
[70]	Electric vehicle (EV)	Korean and Chinese public consumers	Esthetic, functional, symbolic, perceived ease of use, perceived usefulness, environmental awareness, and willingness to buy
[74]	Alternative fuel machinery	Chinese farmers	Perceived usefulness, perceived ease of use, perceived behavioral control, subjective norms, perceived benefit, behavioral attitude, behavioral intention
[68]	Electric vehicles (EVs)	Consumers in Texas	Perceived usefulness, perceived risk, personal innovativeness, adoption intention
[65]	LNG as an alternative fuel in the transport sector	Stakeholders along the whole LNG value chain (European Rhine–Main–Danube axis)	Accessibility, attitude, safety, usability, usefulness, acceptance

3. Research Model and Method

3.1. Research Process

In this study, we constructed an extended TAM through the research process illustrated in Figure 1 below.

Figure 1. Research process for developing the extended TAM framework.

In the first step, we conducted a multifaceted review of prior TAM-related research to identify conceptual constructs and develop a research model for the study design. Hypotheses were formulated based on the derived constructs, referencing hypotheses validated in prior studies. The questionnaire referenced existing items measuring TAM constructs, while risk perception factors related to alternative fuel characteristics were developed by the research team based on prior studies. The survey was conducted online, and basic preprocessing for data analysis was performed.

In the second step, exploratory factor analysis (EFA) was performed to assess the heterogeneity of the newly added construct and refine the items. After confirming the suitability of the data for factor analysis via KMO and Bartlett’s tests, factors were extracted using Varimax orthogonal rotation. Subsequently, items with factor loadings below the threshold were removed.

In the third step, confirmatory factor analysis (CFA) was performed by integrating the construct concepts derived from the EFA with the existing TAM factors. The appropriateness of the construct concepts was verified through content validity, discriminant validity, and internal consistency. This process enabled use of the construct concepts in structural equation modeling.

In the fourth step, the adequacy of the model established in the first step was verified, and the significance of each path coefficient for hypothesis testing was assessed. This determined whether hypotheses were accepted or rejected, and academic and practical implications were derived based on the level of hypothesis acceptance.

3.2. Research Model and Hypotheses

This study adopted the TAM as a fundamental framework to explain seafarers’ acceptance of alternative fuels [32]. The TAM assumes that PU and PEOU influence an individual’s attitude toward the technology (ATT) and BI. Its explanatory power is enhanced by extending these core components with the addition of elements tailored to various conditions [79]. Therefore, referencing existing research, we expanded the existing model by adding trust and risk factors to establish our research model (Figure 2).

Figure 2. Proposed research model.

PEOU, a core component of the TAM, is positively correlated with PU, and both components are expected to positively influence ATT. Accordingly, we formulated hypotheses H1–H3 (Table 2). Regarding H1, the higher the PEOU of the system, the higher its PU. This is because the less additional training is required to switch to an alternative fuel, the higher the PU of the fuel is among crew members. Regarding H2, PU is a key determinant of ATT because the more crew members perceive the transition to alternative fuel as changing their work and as a good method of achieving carbon-reduction goals, the more positive their attitude will be. Furthermore, regarding H3, the higher the PEOU of an alternative fuel is among crew members, the more positive their ATT will be.

Table 2. Summary of research hypotheses and structural paths.

Hypothesis	Path	Content
H1	PEOU → PU	PEOU has a positive effect on PU.
H2	PU → ATT	PU has a positive effect on ATT.
H3	PEOU → ATT	PEOU has a positive effect on ATT.
H4	OSR → ATT	OSR has a negative effect on ATT.
H5	ER → ATT	ER has a negative effect on PU.
H6	ATT → BI	ATT has a positive effect on BI.
H7	TRU → BI	TRU has a positive effect on BI.
H8	PU → BI	PU has a positive effect on BI.

The newly adopted alternative ship fuels present distinct safety and environmental risks compared with conventional fuels [82]. As previously discussed, models that incorporate perceived risk into the TAM framework defined and utilized various perceived risks according to the research topic [73,75,83]. However, previous studies did not identify an appropriate conceptual framework for the risks associated with alternative fuels. Therefore, this study categorized the risks arising from the use of alternative fuels according to two distinct elements: operational safety risk (OSR) and environmental risk (ER). These two concepts align with the IMO’s primary objectives of ensuring the safety of ships and human life and preventing marine pollution. Operational safety risks (OSRs) occur where alternative fuels directly impact seafarers. They encompass not only health hazards for seafarers but also operational risks stemming from damage to hulls and machinery. These factors are particularly important because seafarers may find it difficult to distinguish between the risk of damage and risks to human health associated with alternative fuels.

Beyond these direct risk concepts, indirect environmental risks also need to be considered. Although alternative fuels are highly environmentally friendly in terms of carbon emissions, they may not have favorable impacts on the marine environment and its organisms. Therefore, crew members may recognize that indirect environmental risks exist for some alternative fuels. Consequently, we considered the environmental risks (ER) of alternative fuels in relation to environmental pollution to reflect these aspects. While each concept can be expressed as a single perceived risk, understanding how each risk component influences ATT is crucial. Accordingly, each risk factor was expected to act as a deterrent to the formation of attitude. Thus, we formulated H4 and H5, pertaining to factors that exert a negative influence on ATT.

Based on the fundamental structure of the TAM, in which ATT is influenced by PU, PEOU, OSR, and ER directly mediates BI, ATT positively influences BI, leading to H6. The primary predictor of behavior is behavioral intention, which mediates the influence of attitude on behavior [84,85].

Additionally, referring to prior research, we expanded the concept of trust (TRU) to reflect concerns about the safety of alternative fuels [42,78,79,86,87,88]. Although various studies have indicated that trust influences the four components of the existing TAM (PU, PEOU, ATT, and BI), we established H7 based on the expectation that the actual BI will change directly based on the trust felt by individual users. We established H8 based on the expectation that PU, independent of mediating ATT, would directly influence BI if the actual utility of transitioning to alternative fuel was judged to be high, regardless of attitude.

3.3. Method

This study used a cross-sectional online survey to collect data for quantifying seafarers’ acceptance of alternative fuels. A purposive sampling design was considered suitable for the study’s research objectives and adopted as the sampling strategy. The study population was defined as Korean seafarers with onboard experience. The Republic of Korea is an optimal context for investigating seafarers’ perceptions of alternative fuels, given its status as a nation with a high proportion of green vessel construction and one of the largest operating fleets of such vessels. The survey was conducted from April to July 2025, resulting in a total of 232 responses. To ensure the reliability and integrity of the data, the collected questionnaires were rigorously screened. Completed surveys exhibiting patterns of careless responses, such as straight-line responses and inconsistent answers on reverse-scored items, were subsequently removed. Consequently, 219 valid samples were retained and utilized for analysis. The program used for the analysis was IBM SPSS Amos 30.

To scale the responses, all survey items except demographic questions were divided into alternative fuel categories (LNG, methanol, ammonia, and hydrogen) and measured using a five-point Likert scale (ranging from 1 = strongly disagree to 5 = strongly agree). Reverse-scored items were used to ensure reliability. The alternative fuel group reflected opinions of LNG and methanol, fuels that are actively used in the global shipping market in the short term, and carbon-free fuels (hydrogen and ammonia) that align with the IMO’s long-term GHG-reduction strategy. Consequently, the inclusion of LNG, methanol, ammonia, and hydrogen considered both short-term feasibility and long-term net-zero alignment. As this fuel portfolio represents industry adoption trends, infrastructure progress, and technology commercialization stages, it was a valid choice for comparatively analyzing seafarers’ TAM-based acceptance mechanisms. By structuring the survey items according to each fuel type, we compared and measured factors of seafarers’ technology acceptance for each fuel. This enabled us to assess seafarers’ technology acceptance levels for alternative fuels from the present to the medium and long term, along with their risk perceptions.

The model’s constructs were designed based on standardized questionnaires derived from existing TAM-related research. Key risk-related constructs were estimated based on the major risk factors associated with each alternative fuel, and exploratory factor analysis (EFA) was conducted to assess their appropriateness (see Appendix A for the details of the latent constructs, observed variables, and questionnaire items).

The prerequisites for the explanatory factor analysis were assessed using the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity (KMO ≥ 0.80 and Bartlett’s test of sphericity p-value < 0.05 across all fuel groups) [89]. Varimax rotation, an orthogonal method, was employed to maximize the variance of factor loadings, thereby facilitating clearer factor interpretation and minimizing cross-loadings [90]. This rotation was appropriate since the two factors used in the EFA are conceptually distinct and assumed to be uncorrelated. Kaiser normalization was also applied to reduce bias from items with high communalities and enhance the stability of factor loadings [91].

Confirmatory factor analysis (CFA) was conducted on the risk scale refined by EFA and all Technology Acceptance Model (TAM) constructs to assess convergent validity, discriminant validity, and reliability. Convergent validity was confirmed by checking the significance of the standardized factor loadings (λ ≥ 0.50 recommended), Average Variance Extracted (AVE ≥ 0.40), and Composite Reliability (CR ≥ 0.70) [92]. Discriminant validity was verified using the Fornell–Larcker criterion, a classical and rigorous method in Structural Equation Modeling (SEM) that ensures that each latent variable measures a distinct concept, requiring that the square root of the AVE for each construct is greater than its correlation with all other constructs [92,93].

The model fit for the CFA model was evaluated using the same indices as the SEM: the Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMR), with recommended cutoffs of CFI, IFI, TLI ≥ 0.90, RMSEA, and SRMR ≤ 0.08 [92,94].

Structural equation modeling (SEM) was conducted using the Maximum Likelihood (ML) method, applying Covariance-Based SEM (CB-SEM). All latent variables were modeled as reflective constructs. For scale identification, the factor loading of the first indicator (item) for each latent variable was fixed to 1. The path coefficients were reported as standardized estimates (β). Significance was determined according to the critical ratio and the p-value. The model fit for the SEM was evaluated using the same indices as the CFA: the CFI, TLI, RMSEA, and SRMR.

Hypothesis testing was performed based on the standardized path coefficients and the CR of the structural model. For the corresponding path of each hypothesis (H1–H8), β, standard error, CR, and p-value were reported. A hypothesis was deemed supported if the p-value was less than 0.05.

Furthermore, to investigate the perceived differences across fuel types, a separate SEM analysis was conducted for each fuel type (LNG, methanol (MET), ammonia (AMM), and hydrogen (HYD)) to compare the path coefficients.

4. Results

Table 3 presents the respondents’ demographic profiles. Male respondents constituted the majority at 214 (97.7%), with 5 female respondents (2.3%). This aligns with the male-dominated composition of the global seafarer workforce, which is consistent with the 1.2% proportion of female seafarers worldwide reported by the BIMCO (Baltic and International Maritime Council)/ICS (International Chamber of Shipping) in 2021 [95]. Those under 30 years of age comprised the largest age group, with 95 participants (43.4%), followed by 51 participants (23.3%) aged 30–39 years, 33 (15.1%) aged 40–49 years, 27 (12.3%) aged 50–59, and 13 (5.9%) aged 60 years or older. In terms of service experience, 74 individuals (33.8%) had 3–5 years of service, 49 individuals (22.4%) had less than 3 years, 40 individuals (18.3%) had 5–10 years, 31 individuals (14.2%) had 10–20 years, and 25 individuals (11.4%) had more than 20 years. Junior officers/junior engineers (operational level) constituted the largest group at 114 respondents (52.1%), while there were 23 captains (10.5%), 49 chief officers/first engineers (22.4%), 27 chief engineers (12.3%), and 6 others (2.7%). The sample primarily comprised officers and engineers responsible for the practical implementation of alternative fuel usage, making it appropriate for assessing their awareness of relevant practices. This sample has a limitation in that it does not reflect the perceptions of support-level seafarers regarding alternative fuels. However, it is better suited for investigating risk perception concerning alternative fuels as it is primarily composed of navigating and engineering officers, including master and chief engineers, who bear direct responsibility for vessel safety and marine pollution prevention. Furthermore, the ranks included in the sample are generally aligned with both the operational and management levels of most vessels, indicating a well-structured representation.

Table 3. Respondents’ demographic profile (N = 219).

Characteristics of Respondents		N	Frequency %
Gender	Male	214	97.7%
	Female	5	2.3%
Age	Under 30 years	95	43.4%
	30–39 years	51	23.3%
	40–49 years	33	15.1%
	50–59 years	27	12.3%
	Over 60 years	13	5.9%
Onboard experience	<3 years	49	22.4%
	3–5 years	74	33.8%
	5–10 years	40	18.3%
	10–20 years	31	14.2%
	More than 20 years	25	11.4%
Position	Master	23	10.5%
	Chief engineer	27	12.3%
	Chief officer and first engineer	49	22.4%
	Junior officer and engineer	114	52.1%
	Etc.	6	2.7%
Type of ship 1	Dry bulk carrier	52	23.7%
	Container carrier	68	31.1%
	Ro-Ro carrier	31	14.2%
	Passenger ship	20	9.1%
	Crude oil tanker	28	12.8%
	Chemical tanker	47	21.5%
	Liquid–gas carrier	52	23.7%
	Fishing vessel	15	6.8%

¹ Multiple responses were allowed.

The two risk constructs presented in this study—OSR and ER—are newly proposed concepts. EFA was conducted to measure the number of factors and the explanatory power of the construct concepts. The initial EFA was conducted using all observed variables for each fuel type. The analysis revealed that the Kaiser–Meyer–Olkin values for all four fuels exceeded the criterion KMO ≥ 0.80, indicating that the correlation among variables was adequately explained by common factors [89]. However, the items OSR1 and OSR7 were removed as they failed to meet the specified factor loading criterion (λ ≥ 0.50) [65]. Following the removal of these items and the execution of a factor analysis, the results were classified according to two concepts, as shown in Table 4. Items 1 and 2 in Table 4 represent the components separated by the factor analysis. These distinct components correspond to the two risk concepts proposed in the model. In the EFA by fuel type, the Kaiser–Meyer–Olkin (KMO) value ranged from 0.802 to 0.821 (KMO ≥ 0.80), and Bartlett’s test of sphericity was significant for all (p < 0.001), indicating suitability for factor analysis [89]. In factor analysis, the total variance should be explained by at least 60% of the rotation sums of the squared loadings [96]. This was confirmed across all fuel groups, indicating that the variables sufficiently explained the overall factors. An analysis to validate this concept suggested that in the context of introducing alternative fuels, the risks associated with alternative fuels comprise OSR, representing physical and human safety risks, and ER. This provided theoretical grounds for subsequent structural equation modeling to verify the relationship between each risk and ATT.

Table 4. Exploratory factor analysis results.

Latent Factor	Observed Variables	LNG		Methanol		Ammonia		Hydrogen
		1	2	1	2	1	2	1	2
OSR	OSR2	0.549	0.163	0.662	0.195	0.634	0.216	0.586	0.314
	OSR3	0.706	−0.015	0.664	0.103	0.563	0.248	0.694	0.054
	OSR4	0.757	−0.042	0.745	0.011	0.727	−0.084	0.808	−0.139
	OSR5	0.805	0.111	0.834	0.200	0.765	0.269	0.734	0.299
	OSR6	0.838	0.032	0.810	0.114	0.811	−0.036	0.844	−0.065
	OSR8	0.692	0.041	0.648	0.055	0.642	0.174	0.696	−0.011
	Cronbach’s α	0.817		0.831		0.795		0.822
ER	ER1	0.063	0.900	0.209	0.870	0.172	0.873	0.057	0.916
	ER2	0.027	0.921	0.176	0.888	0.159	0.901	0.050	0.922
	ER3	−0.014	0.865	0.018	0.812	0.012	0.832	−0.037	0.848
	ER4	0.156	0.830	0.118	0.844	0.226	0.841	0.117	0.809
	Cronbach’s α	0.905		0.887		0.902		0.904
KMO		0.802		0.821		0.813		0.808
Bartlett’s test		1101.6 ***		1071.4 ***		1053.5 ***		1178.5 ***
Rotation sums of squared loadings		3.232	3.140	3.296	3.022	3.006	3.194	3.234	3.277

* p < 0.05, ** p < 0.01, *** p < 0.001.

The results of the confirmatory factor analysis conducted to validate each factor structure used in the analysis are presented in Table 5. The model incorporates the construct concepts PU, PEOU, ATT, BI, and TRU from the extended TAM alongside the risk concepts OSR and ER, which were identified through exploratory factor analysis. The beta values (standardized regression weights) for each construct concept, along with the average variance extracted (AVE), construct reliability (CR) for confirming convergent validity, and Cronbach’s alpha values for confirming internal consistency reliability, are presented for each of the four fuels. While most variables from the existing TAM fit their respective components, some variables from PEOU lacked sufficient validity and were therefore removed. Cronbach’s values for reliability measurement showed PEOU at 0.674 for methane fuel, while all other components demonstrated high reliability above 0.7. This value is above the commonly used threshold of 0.6, so it was used in the analysis [97]. The average variance extracted (AVE) values also exceeded 0.5 for all components except OSR, and construct reliability (CR) was adequate at 0.7 or above for all components [93]. Although the AVE value for OSR was somewhat low, CR was above 0.6. As the AVE criterion is relatively strict, an evaluation solely based on CR suggests that it is acceptable as a component [93,98,99]. The model demonstrated an acceptable level of fit, with all fit indices meeting the recommended thresholds.

Table 5. Confirmatory factor analysis results.

Latent Construct	Observed Variables	LNG				Methanol				Ammonia				Hydrogen
		$\mathsf{\beta }$	AVE	CR	α	$\mathsf{\beta }$	AVE	CR	α	$\mathsf{\beta }$	AVE	CR	α	$\mathsf{\beta }$	AVE	CR	α
PU	PU1	0.832	0.725	0.894	0.927	0.840	0.764	0.911	0.941	0.815	0.759	0.908	0.939	0.740	0.689	0.873	0.915
	PU2	0.773				0.815				0.812				0.794
	PU3	0.879				0.913				0.928				0.882
	PU4	0.916				0.911				0.904				0.858
	PU5	0.852				0.887				0.890				0.868
PEOU	PEOU4	0.754	0.555	0.744	0.712	0.734	0.510	0.728	0.674	0.804	0.573	0.771	0.724	0.806	0.584	0.771	0.732
	PEOU5	0.736				0.694				0.707				0.720
ATT	ATT1	0.903	0.787	0.932	0.936	0.935	0.801	0.936	0.940	0.899	0.804	0.937	0.941	0.912	0.786	0.930	0.935
	ATT2	0.930				0.945				0.928				0.907
	ATT3	0.847				0.831				0.873				0.839
	ATT4	0.866				0.864				0.885				0.887
BI	BI1	0.925	0.669	0.870	0.860	0.891	0.614	0.850	0.818	0.940	0.665	0.876	0.849	0.922	0.657	0.872	0.841
	BI2	0.793				0.807				0.837				0.841
	BI3	0.722				0.629				0.642				0.644
TRU	TRU1	0.959	0.860	0.943	0.945	0.971	0.870	0.942	0.951	0.974	0.886	0.950	0.958	0.967	0.864	0.942	0.948
	TRU2	0.959				0.927				0.948				0.967
	TRU3	0.861				0.898				0.900				0.850
OSR	OSR2	0.425	0.451	0.825	0.817	0.621	0.467	0.836	0.831	0.595	0.406	0.801	0.795	0.505	0.454	0.830	0.822
	OSR3	0.600				0.569				0.556				0.617
	OSR4	0.737				0.629				0.556				0.764
	OSR5	0.725				0.875				0.792				0.662
	OSR6	0.851				0.794				0.688				0.824
	OSR8	0.608				0.546				0.604				0.623
ER	ER1	0.904	0.706	0.905	0.905	0.904	0.667	0.887	0.887	0.893	0.698	0.901	0.902	0.940	0.706	0.904	0.904
	ER2	0.937				0.929				0.941				0.956
	ER3	0.764				0.655				0.695				0.741
	ER4	0.737				0.748				0.792				0.691

Table 6 presents the correlations confirming the degree of association between the components and their discriminant validity. Although some variable correlations were relatively high, they corresponded to relationships already theoretically established in the TAM. Furthermore, as each correlation coefficient was smaller than the square root of the AVE diagonal matrix, discriminant validity could be presumed to exist [93,100].

Table 6. Correlation test results.

LNG								Methanol
	PU	PEOU	ATT	BI	TRU	OSR	ER		PU	PEOU	ATT	BI	TRU	OSR	ER
PU	0.851							PU	0.874
PEOU	0.549	0.746						PEOU	0.508	0.714
ATT	0.820	0.649	0.888					ATT	0.812	0.642	0.896
BI	0.684	0.631	0.770	0.817				BI	0.624	0.678	0.770	0.783
TRU	0.669	0.694	0.810	0.738	0.927			TRU	0.686	0.672	0.796	0.727	0.932
OSR	−0.053	−0.32	−0.081	−0.056	−0.105	0.672		OSR	−0.113	−0.392	−0.081	−0.129	−0.203	0.683
ER	−0.145	−0.143	−0.151	−0.107	−0.139	0.113	0.840	ER	−0.133	−0.156	−0.135	−0.094	−0.060	0.370	0.817
Ammonia								Hydrogen
	PU	PEOU	ATT	BI	TRU	OSR	ER		PU	PEOU	ATT	BI	TRU	OSR	ER
PU	0.871							PU	0.830
PEOU	0.413	0.756						PEOU	0.434	0.764
ATT	0.804	0.579	0.897					ATT	0.779	0.619	0.887
BI	0.621	0.587	0.820	0.816				BI	0.568	0.619	0.768	0.811
TRU	0.661	0.627	0.808	0.760	0.941			TRU	0.601	0.707	0.762	0.728	0.930
OSR	−0.075	−0.433	−0.089	−0.134	−0.192	0.637		OSR	−0.080	−0.469	−0.073	−0.143	−0.217	0.674
ER	−0.204	−0.173	−0.254	−0.199	−0.162	0.399	0.835	ER	−0.118	0.032	−0.170	−0.092	−0.029	0.097	0.840

Note: The square roots of the AVE appear on the diagonal, whereas all other positions indicate correlations.

Table 7 presents the model fit indices for each alternative fuel model. The ${\chi }^2/df$, Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR) are absolute fit indices that comprehensively verify the goodness-of-fit of the model. The ${\chi }^2/df$ values ranged from 1.885 to 2.142 and were all below the conventional threshold of 3 [101]. The RMSEA measures the magnitude of the population approximation error while considering model parsimony. The SRMR measures the average of the model’s standardized residuals. RMSEA values ranged from 0.66 to 0.72, and SRMR values ranged from 0.06 to 0.08, indicating an adequate fit. The Incremental Fit Index (IFI) and the Comparative Fit Index (CFI) were used to evaluate the degree to which the model developed in the study is an improvement compared to the most restricted null model (baseline model) that assumes zero correlation among all observed variables. The Tucker–Lewis Index (TLI) aids in finding a parsimonious model by imposing a penalty for model complexity. The IFI, TLI, and CFI values were all above 0.9 [94,102,103].

Table 7. Summary of model fit measures.

Fuel	${\mathit{\chi }}^2/\mathit{d}\mathit{f}$	IFI	TLI	CFI	RMSEA	SRMR
LNG	1.953	0.937	0.927	0.936	0.066	0.060
Methanol	1.885	0.942	0.934	0.942	0.064	0.064
Ammonia	2.142	0.928	0.917	0.927	0.072	0.067
Hydrogen	1.978	0.934	0.925	0.934	0.067	0.080

Table 8 presents the path coefficients and critical ratios for the structural equation model for each fuel in the hypothesis. When a path coefficient is statistically significant, it indicates a substantial influence of the path between the corresponding constructs. First, H1, H2, and H3 showed significant positive effects across all fuel types, consistent with the results of previous TAM-related studies. For H1, the coefficients of methanol (0.726) and LNG (0.707) were notably large: as experience with these two relatively short-term fuels accumulates, their PEOU strongly influences their PU. Therefore, the practical usefulness of short-term alternative fuels is expected to increase as procedures are simplified and requirements, such as crew training, are reduced. H2 and H3 showed positive relationships, but H3’s coefficient (0.607–0.684) was higher than that of H2 (0.352–0.414). Thus, ease of use exerts a greater influence on seafarers’ attitudes toward adopting eco-friendly technologies than usefulness. In particular, LNG and methanol, despite being relatively familiar, significantly influenced attitude improvement based on ease of use.

Table 8. Structural equation modeling results.

Hypotheses	Path	Path Coefficients				Critical Ratio				Result
		LNG	MET	AMM	HYD	LNG	MET	AMM	HYD
H1	PEOU →PU	0.707 ***	0.726 ***	0.669 ***	0.578 ***	7.953	7.830	7.625	7.184	Supported
H2	PU →ATT	0.404 ***	0.352 ***	0.366 ***	0.414 ***	5.806	4.636	5.591	7.005	Supported
H3	PEOU →ATT	0.607 ***	0.668 ***	0.680 ***	0.684 ***	6.621	6.465	7.122	8.177	Supported
H4	OSR →ATT	0.104 *	0.187 **	0.237 ***	0.290 ***	1.963	3.182	3.554	4.122	Opposite is supported
H5	ER →ATT	0.010	−0.033	−0.102	−0.147 **	0.211	−0.627	−1.825	−2.959	Partially supported
H6	ATT →BI	0.386 **	0.560 ***	0.682 ***	0.566 ***	3.127	4.676	6.126	5.301	Supported
H7	TRU →BI	0.329 ***	0.314 ***	0.285 ***	0.344 ***	3.770	3.506	3.545	4.414	Supported
H8	PU →BI	0.149	−0.045	−0.117	−0.080	1.595	−0.488	−1.495	−0.992	Not supported

* p < 0.05; ** p < 0.01; *** p < 0.001.

Regarding H4, operational safety risk showed a significant positive influence on attitude toward all alternative fuels, yielding results contrary to the hypothesis—that is, perceived level of risk has a positive effect on attitude. This finding is at odds with the majority of extant research; however, it is in alignment with several prior studies suggesting that heightened risk awareness may engender positive attitudes, safe behaviors, and information seeking driven by the expectation that risks will be mitigated [3,38,104,105]. Thus, seafarers may tend to adopt a more positive attitude, as they likely assume that the higher the risk, the greater the necessity of using that fuel. Notably, the coefficients for long-term fuels, such as hydrogen and ammonia, for which risks may be uncertain, were higher. This could be interpreted as the result of higher risk perception increasing expectations for improved procedures and equipment. Regarding H5, change in attitude related to ER was not significant for most fuels, with only hydrogen showing a negative effect, thus supporting this hypothesis. Although hydrogen is generally associated with clean energy, seafarers first recognize concerns about the environmental damage caused by explosive combustion due to hydrogen leaks, which can negatively affect their attitudes.

H6 and H7 significantly influenced BI. Ammonia exhibited the highest coefficient in H6, indicating strong future BI following an attitude change. Conversely, LNG, the most widely adopted alternative fuel, had a low coefficient of 0.386, suggesting a minimal impact on the formation of attitude-based BI. Regarding the influence of trust on BI in H7, similar levels of impact were observed, although hydrogen had the highest coefficient at 0.344. Similarly to H5, although seafarers perceive hydrogen as somewhat hazardous, increased trust could lead to the most significant change in their BI. Finally, contrary to our hypothesis, none of the factors in H8 were significant; thus, usefulness alone does not translate into actions on the part of seafarers and must be accompanied by attitude changes and perceived reliability.

5. Discussion

This study assessed seafarers’ levels of technology acceptance of alternative fuels based on a modified TAM, which identified perspectives on technology adoption. Specifically, our model analyzed and compared various alternative fuels (i.e., LNG, methanol, ammonia, and hydrogen), which have been introduced in response to recent IMO carbon and air pollution regulations. The IMO is implementing short-, medium-, and long-term measures to reduce carbon emissions. The transition to alternative fuels is considered the most effective method in the medium to long term. The most suitable fuel for each period may change; however, the development of associated technologies and deployment are gradually progressing. This leaves seafarers with no choice but to adopt technologies in stages. Owing to the difference in characteristics between alternative and conventional fuels, seafarers’ participation in their adoption is crucial for future carbon-reduction objectives.

Our results showed that all basic pathways of the TAM yielded significant results, which is consistent with the findings of previous studies. In particular, for short-term transition fuels, such as methanol and LNG, ease of use was strongly associated with PU. This result is consistent with those of previous studies, showing that the more user-friendly a technology is, the more useful it is perceived to be by actual users [3,35,38,39,40,62,69,71,73,79]. As in existing prior studies, both PU and PEOU were found to have a positive influence on ATT [38,42,69,74,79]. In particular, while some studies [37,71] have reported that PEOU has no significant impact or even a negative impact on ATT, suggesting that it may not be decisive for technology adoption, this study revealed that PEOU exerts a greater influence on ATT than PU. Thus, continuous technological improvements are necessary for operational convenience.

Interestingly, operational safety risks had a significant positive effect across all fuels. This finding contradicts the majority of studies indicating that risk perception negatively affects attitudes or intention [39,73,75,83,86]. However, some studies suggest that as perceived risk increases, the expectation that technological risks will be mitigated can have a positive influence on attitudes [3,38,104,105]. The perception of risk itself, rather than the risk level, drives positive attitude changes because, as ship technology generally evolves conservatively, the existence of risks may be perceived as necessitating special usage requirements for a technology’s application. Additionally, the risks associated with alternative fuels, which demand higher levels of education and knowledge, may induce proactive attitudes, such as the incentive to participate in additional tasks [106]. In other words, from the seafarer’s perspective, risk perception can actually have a positive influence on attitude.

However, changes in attitude due to ERs were mostly insignificant, with only hydrogen negatively affecting attitude. Seafarers generally perceive all alternative fuels to be environmentally friendly, which makes it difficult to observe significant changes in their attitude towards environmental pollution and the damage it causes. However, heightened awareness of risks, such as hydrogen explosions, has led to negative attitudes towards the environmental pollution caused by such incidents [107]. As expected, ATT had a positive influence on BI [37,38,42,65,69,71,78,79]. Therefore, efforts are needed to encourage positive changes in attitude toward the adoption of technology. Furthermore, TRU had a positive influence on BI, which aligns with prior studies and confirms that trust is crucial to technology adoption [42,79,80]. Therefore, establishing transparent trust in technology is vital for institutional change. Finally, PU did not significantly influence BI, an outcome that differs from prior research [35,37,39,40,42,62,65,66,67,76,79]. This means that the behavior of seafarers, as the end users of alternative fuels, did not change due to the perceived usefulness of the fuels. In other words, the usefulness of alternative fuels can be considered to affect behavioral intent not by changing it directly but by mediating behavior through changes in attitude. This suggests that since seafarers have limited choice in using alternative fuels, technology adoption occurs within restricted parameters through changes in attitude rather than behavioral intent.

6. Conclusions

6.1. Summary of Key Findings

This study examined seafarers’ acceptance of alternative fuel technologies following the introduction of alternative fuels, which is one of the most significant changes in the shipping industry at present. To this end, the TAM, which assesses the acceptance of novel technological changes, was used to compare and analyze seafarers’ perceptions of each alternative fuel, yielding the following results.

First, the basic pathways influencing TAM attitudes (H1, H2, and H3) were significantly validated for all fuels analyzed. Notably, the influence of ease of use on attitude formation was greater than that of usefulness. This indicates that, during the process of accepting alternative fuels, crew members may adjust their attitudes based more on the convenience of actual operation than on the usefulness of a specific alternative fuel. Changes in attitude based on ease of use require crew members to perceive the transition to alternative fuels as straightforward. However, many alternative fuels require additional training due to regulations, such as the IGF Code. This is likely to act as a barrier to the adoption of alternative fuel technology. Therefore, foundational training related to these regulations must be prioritized at basic maritime training institutions. Providing early foundational training will lower barriers to subsequent advanced training. Standardizing foundational technologies for alternative fuel conversion would reduce the training burden on companies and institutions responsible for advanced education while also contributing to the effective allocation of resources.

Second, OSR had a positive influence on attitude; recognizing and accepting risks as manageable contributed to the formation of positive attitudes among crew members. This phenomenon was particularly pronounced for long-term fuels, such as hydrogen and ammonia, which have relatively low usage levels but potentially uncertain risks. This finding suggests that risk perception may enhance expectations of safety, which in turn can positively shape attitudes toward the adoption of new technology. The working environment of seafarers is highly hazardous, but they are accustomed to managing risk effectively. This remains true even for additional risks arising from the transition to alternative fuels. Seafarers have high confidence in risk management thanks to the meticulous identification of risk factors during the assessment phase. Therefore, current temporary risk management regulations must be strengthened to establish systems and procedures capable of identifying a broad range of risk factors. If seafarers perceive the risks of alternative fuel conversion to be manageable, their attitudes will become more positive.

Third, ER had a weak influence, in general, but a negative effect in the case of hydrogen. This indicates that despite the inherent cleanliness of hydrogen, the potential for emergency situations to result in environmental damage and harm negatively influences the attitudes of crew members. Therefore, to introduce hydrogen fuel, it is necessary to emphasize its cleanliness and provide education and technological improvements to minimize environmental damage and enhance crews’ accident response capabilities.

Fourth, attitude and trust, which are elements of the TAM, emerged as key determinants of BI. Notably, for ammonia, attitude exerted the strongest influence on BI, whereas for hydrogen, TRU had the most prominent effect. Thus, shifting attitudes and building trust are critical to the acceptance of these long-term fuels. Over the years, the IMO and governments worldwide have established a variety of safety systems and continuously revised regulations. This feedback has been documented and systematized, particularly with the introduction of the ISM Code. Through this process, seafarers have developed a great deal of trust in the system. Therefore, it is crucial that policymakers and educational institutions do not undermine this trust, which requires dedicating significant effort to education and promotion to maintain. After policy formulation, systems such as the port state control (PSC)’s concentrated inspection campaigns (CICs) should also be actively utilized.

Finally, PU itself was not directly related to BI. Thus, PU alone does not determine seafarers’ BI; rather, changes in ATT based on PEOU and PU are important.

In conclusion, strategies for the adoption of alternative fuel technologies are ineffective if they remain confined to promoting utility. Improving ease of use, developing risk-management systems and technologies, and ensuring reliability are crucial factors in shaping seafarers’ attitudes and BI. This has important implications for establishing practical policies in the shipping industry, such as designing education and training programs, strengthening safety devices and procedures, and establishing institutional frameworks to guarantee safety.

6.2. Limitations and Future Research

This study has some limitations. With regard to the adoption of alternative fuel conversion technology, the shipping industry involves various stakeholders. Although seafarers are the end users, the roles determining their actual usage can vary. It is necessary, then, to expand the research scope to include decision makers in shipping companies and shipyard personnel and to diversify the nationalities of the seafarers surveyed. Expanding the stakeholder base will also require incorporating additional components into the model. For example, changes may occur based on factors such as financial considerations, route decisions, cultural variables, and port selection. Future research should adopt a multidimensional approach to overcome these limitations. Specifically, conducting longitudinal studies over extended periods would enable detailed observation of how each stakeholder’s perception evolves in response to technological change. By conducting comparisons between countries based on the expanded sample, it is also possible to determine whether there are actual differences in awareness regarding the transition to alternative fuels among seafarers across nations.

Nevertheless, this study is significant in terms of identifying the degree of acceptance of alternative fuel conversion technologies by seafarers, who are the actual users of the technology. Since the pace of technology adoption by seafarers will determine the extent to which medium- to long-term goals for alternative fuel conversion are achieved, this research is fundamental to achieving carbon reduction targets. Through this analysis, a theoretical and empirical foundation for designing carbon reduction policies is established.