Unraveling the Usage Characteristics of Human Element, Human Factor, and Human Error in Maritime Safety

Abstract

Human-related issues have become a popular topic in maritime safety research, with an increasing number of relevant research articles being published annually. However, a persistent problem in this field is that three terms, namely “human element”, “human factor”, and “human error” are used interchangeably in the literature. This issue poses questions regarding the characteristics of their usage; do these three terms have the same meaning? Herein, we conducted systematic research on the three terms by analyzing official information and published research using a collecting–classifying–summarizing policy. The results show that “human error” is easier to identify than “human element” and “human factor”, while the latter two terms have intersecting contents. These contents prompt the user to decide which term to choose depending on the situation. Herein, we aim to help scholars accurately distinguish these terms.

1. Introduction

The disaster of the motor vessel “Herald of Free Enterprise” in 1987 highlighted the human element in maritime safety over technical factors. Later, the International Maritime Organization (IMO) laid out its vision, principles, and goals regarding the human element by adopting Resolutions A.850(20) [1] and A.947(23) [2] in 1997 and 2003, respectively. People began to focus on human-related influences from different perspectives, especially those related to accident prevention. Subsequently, the International Safety Management (ISM) code was created and enforced as Chapter 9 of the Safety of Life at Sea Convention [3]. The fundamental principle of this code is to manage maritime safety, environmental quality, and security measures, which are all human-related issues. The ISM code has been revised many times, including Resolution A.788(19) [3] in 1995, A.913(22) [4] in 2001, A.1022(26) [5] in 2009, A.1071(28) [6] in 2013, and A.1118(30) [7] in 2017. Evidently, human influence is crucial to maritime safety.

Despite this, among the factors that can cause maritime accidents, the human factor remains the dominant one. The annual overview report of marine casualties and incidents (2021) [8] released by the European Maritime Safety Agency (EMSA) indicates that 89.5% of safety issues are affected by the human element (Figure 1). This percentage is 89.1% for cargo and passenger ships is, while it is 89% for fishing vessels. Note that being affected by the human element does not mean that the human element is solely responsible for casualties or incidents but that it is present. Strictly speaking, maritime accidents are not caused by one factor but by a series of factors. However, only human factors are considered herein as the EMSA report reveals that the human element deeply impacts ships’ safety. Currently, the maritime industry is pushing to improve safety and efficiency [9], which may be another reason why the human element has recently come into focus.

In the 1990s, a study observed that “human factor” and “human error” were used interchangeably as general terms [10]. Recently, Wrobel [11] discovered that the use of “human element”, “human factor”, and “human error” was confused. These two articles suggest that the interchangeability of these terms has become complicated rather than being resolved. Moreover, “human reliability” will be involved in this mix-up in the near future (refer to Section 3.4).

1.1. Significance

The abovementioned three terms have different origins. “Human element” was introduced by the IMO in the 1990s [1] to cope with human influence in the maritime domain. Its background implies that it is a domain-specific term. The term “human factor” has been used since World War II, and it is an interdisciplinary subject involving psychology, sociology, engineering, physiology, biomechanics, anthropometry, and user interface design [12]. The term “Human factor” has been utilized in various fields; thus, it differs from “human element” because of their different backgrounds. “Human error” first appeared in journal articles in the 1960s. It followed a path of “topic after working (the 1970s)→concept during working (the 1980s)→central of the human factor (the 1990s)” [13].

“Human factor” and ”human error” were not initially developed in the maritime domain, but have been recently adopted. The IMO adopted the term “human element” for use in the maritime field but did not differentiate the three terms properly. The mix-up in the use of the three terms has become common in research articles such as that conducted by Qiao et al. [14] who evaluated the human factors influencing maritime accidents, while human error appeared throughout the article. Other studies have shown a similar situation [15,16,17]. However, some scholars have addressed this confusion, noting that it could compromise the accuracy of human reliability assessment (HRA) [10,11,18]. Other scholars have distinguished between “human factor” and “human error” [19]. Unfortunately, research on the similarities and boundaries among the three terms is limited. This may be because only a few studies have considered differentiating among the terms to avoid complications [18]. However, the necessity to correctly identifying each term depends on the development, pain points, and future trends in the maritime domain. Since the implementation of the ISM code, the “human factor” has been a research focus, with an increasing number of human-related articles, dissertations, and conference papers published. If we just assume that the three terms express the same meaning, then one might misunderstand the authors’ intentions or the meanings they are attempting to express. For example, one study [20] claimed that marine causalities or safety incidents caused by human factors might not be because of human error but because of internal pressure. For pain points, numerous maritime accidents are related to human factors (EMSA report), indicating that the human factor remains a pain point in the maritime safety field. For future trends, it will likely be divided into manned and unmanned navigation. For manned navigation, similarly to present navigation, human factors will remain an important issue. For unmanned navigation, the vessel will be driven automatically or controlled remotely by shore-side drivers. In remote control situations, research has shown that human error remains the main factor influencing navigation safety [19]. Automatic navigation could reduce the chance of manual operations, thus minimizing navigational accidents (e.g., grounding, collision, and contact), but non-navigational accidents (e.g., fire, blackout, and cyber security hijacks) will increase significantly compared with manned ships [20]. Moreover, human error will still exist in the design stage or the operational period [21]. Therefore, it is necessary to distinguish the three terms.

1.2. Objective and Scope of the Study

Research on human-related issues is popular, and many articles have been published in various journals, such as Safety Science, Reliability Engineering and System Safety, and Ocean Engineering.

While studying the literature, we observed that many scholars used the three abovementioned terms interchangeably within a single article to express similar meanings. However, the authors of each article are not always the same, and it is not accurate and scientific to presume that these articles express the same meaning as different authors cannot always have the same opinion concerning a vague topic. Therefore, the following questions arise: what are the differences and connections between each term? Is it necessary to identify their usage, and how can this be done? A review of different databases (e.g., Web of Science, ScienceDirect, and IEEE Xplore) showed that almost no systemic studies on this topic exist.

Herein, we aim to compare the use of the terms “human element”, “human factor”, and “human error” to identify their differences and boundaries and help end the consistent mix-up of these terms from an academic perspective. This is accomplished through a systemic survey of the three terms in peer-reviewed articles and official information on maritime transportation.

2. Materials and Methods

Systematic research was performed on the three terms to conduct the identification process. A large amount of information, articles, and dissertations were collected for analysis, and the procedures are shown in Figure 2. The information sources included IMO documents, dissertations, journal articles, and conference papers. The collected information was divided into four categories: human element; human factor; human error; and HRA. The principle behind dividing the terms is found in Section 2.2. Subsequently, we analyzed the definition and research status of each part and discussed the relevant boundaries. Finally, a comparative analysis was conducted based on the above information.

2.1. Information Sources

The primary information sources were official information and academic articles. Official information from the IMO website and relevant documents (circulars, resolutions, and conference documents) were included. The IMO documents provided guidelines about the three terms. Furthermore, information from influential institutions, such as the British Nautical Institute, was also collected. Articles, reviews, conferences, and dissertations were the primary information sources obtained from academia. Articles, reviews, and conferences were collected from three databases, namely Web of Science, ScienceDirect, and IEEE Xplore. Dissertations were obtained from the ProQuest Digital Dissertation Library (PQDD) database. The keywords are listed in Section 2.2.

2.2. Selection Principle

To uncover related articles, several keywords were selected. These included human element, human factor, human error, human reliability analysis, human error probability, maritime accident, marine accident, maritime safety, marine safety, marine transportation, and maritime transportation. A publication period of 10 years was used from 1 January 2011 to 31 December 2021. The authors manually checked all the articles to determine whether they focused on the topic of human contribution to maritime safety. Otherwise, the article was excluded. However, only documents in the English language were included. The results of this screening were named “primary articles”. The references from the primary articles that concentrated on human-related issues formed the “secondary articles”. Subsequently, tertiary and quaternary articles were similarly collected. Duplication was avoided throughout the selection process.

Table 1. Collected articles.

Literature	Number of Documents
Primary	40
Secondary	61
Tertiary	7
Quaternary	3

shows the result of the selection.

As discussed in Section 1, many academic papers have used the three terms interchangeably; thus, researching the terms would be complicated if all of the collected articles were typed into one group. However, as the articles were divided into several groups, a robust principle to distinguish between the articles was required. The straightforward choice was to use the titles of the articles. Articles that included “human element” in the title were placed into the human element group, and articles with “human factor” in the title were placed into the human factor group. A similar procedure was used for the human error and HRA groups.

2.3. Outcomes

Our main aim was to identify the usage principles of the three terms, the ambiguity of their boundaries, and the similarities among them. At last, we provide a reference to clarify the confusion regarding the use of the terms and to help others use these terms to express ideas clearly.

3. Results and Discussion

This section presents the results of the systematic research. The IMO does not limit the incidents involving human activities to an official name, so all three terms are applicable. This may be due to a lack of consensus. Nevertheless, on the IMO website, the section that concerns human-related issues is termed “Human element” [22].

3.1. Human Element

3.1.1. Definition

In Resolution A.849(20), the IMO defines the human element as follows [23]: “The human element is a complex multidimensional issue that affects maritime safety and marine environmental protection. It involves the entire spectrum of human activities performed by ship’s crews, shore-based management, regulatory bodies, recognized organizations, shipyards, legislators, and other relevant parties and they need to cooperate to address human element issues effectively”. However, this definition is specific and complex. The British Nautical Institute proposes a simpler and more concise definition [24]: “In the maritime context, the term Human Element embraces anything that influences the interaction between a human and any system aboard ship”. This second definition focuses on entities aboard the ship, whereas the first has a broader scope.

3.1.2. Research Status

In addition to these definitions, Figure 3 illustrates the connotation and extension of the human element. It indicates that the human element comprises six parts: people factors; ship factors; external influences and environment; working and living conditions; shore-side management; and organization on board. The details of these six parts are discussed elsewhere [25]. Some of these segments, such as people factors, shore-side management, and organization on board, can directly affect human behavior. According to Section 3.2, these segments can be viewed as human-related factors; the remaining categories indirectly affect human behavior.

To thoroughly analyze and study the human element, the IMO adopted the Human Element Analyzing Process (HEAP) to help regulatory bodies consider the human element when developing regulations. The details of the HEAP can be found elsewhere [25].

Few academic papers on the human element have been published in the maritime safety field. We collected seven articles (five journal articles and two conference papers) with “human element” in the title. The details are listed in Appendix A. Popa [26] introduced performance management systems, crew management, and human contribution to shipping companies. Juan et al. [27] addressed how human error influences shipping casualties, fluctuating the risk homeostasis in the shipping market. Furthermore, Paolo et al. [28] developed the semi-SRPM model to analyze human factors by studying maritime accidents. Other work has focused on the potential impacts of autonomous technology and human roles in maritime operations [29,30]. The authors of [31] studied the influence and benefits of the IMO member states and agencies on maritime safety and human factor issues. Barnett et al. [32] presented a thorough human element analysis from the perspective of its definition, connotation, and denotation.

3.1.3. Discussion

The IMO documents provide useful information concerning the human element but are still not enough to identify the abovementioned three terms. Only three of the seven collected articles focused on the human element; the others did not. For example, in the first two articles, one addressed human error issues, while the other addressed human factor affairs. Consequently, it was difficult to determine the relationship between the three terms. To address this issue, human factor and human error must be included in the analysis.

3.2. Human Factor

3.2.1. Definition

The “human factor” has been studied in many research fields, resulting in variable definitions, as discussed in Section 1.1. Herein, we focus on “human factors” in maritime safety and its definition.

Table 2. Definitions of human factors.

Ref.	Definitions
IMO Resolutions A.884(21) [33]	Human factors which contribute to marine casualties and incidents may be broadly defined as the acts or omissions, intentional or otherwise, which adversely affect the proper functioning of a particular system, or the successful performance of a particular task.
HSE [34]	Human factors refer to environmental, organizational and job factors, system design, task attributes and human characteristics that influence behavior and affect health and safety.
B. Woodcock [35]	Human factor is the discipline that seeks to understand the contribution of the human operator within the system and to influence design and operation accordingly.
IEA [36]	Human factor is the scientific discipline concerned with the understanding of interactions among humans and other elements of a system, and the profession that applies theory, principles, data and methods to design in order to optimize human well-being and overall system performance.

shows the relevant definitions. HSE’s definition refers to environmental, organizational, and job factors, as well as human characteristics. Woodcock and IEA’s definitions, given from the perspective of scientific discipline, are relatively general and macroscopic. The IMO’s definition emphasizes human actions or omissions (intentional or unintentional) that have caused certain maritime casualties or safety incidents. The commonality between the four definitions is difficult to extract, implying that there is no consensus on the definition of human factor from an academic perspective.

3.2.2. Research Status

Most scholars understand what the human factor refers to, its contents, and how to analyze these contents. However, it is unclear where the content boundary lies and what the differences are compared with the other two terms. Therefore, the contents studied by different scholars are not the same.

The collected articles are listed in

Table 3. Collected articles with human factor in the title.

Category	Number of Documents
Journal article	30
Conference paper	11
Dissertation	2

based on the principle set out in Section 2.2. Their detailed information is provided in Appendix B. Among them, six articles (five journal articles and one conference paper) were review articles involving human factors, while the others were research articles. Notably, human-related factor [17], human influencing factor [20], and human contributing factor [37] were equal to human factor, according to the relevant articles.

Research Article

These articles often used newly proposed or modified models to conduct research. The popular model, HFACS, has many revised versions: HFACS-FFTA [38]; HFACS-PV [39,40]; HFACS-Coll [41]; HFACS-MA [42], HFACS-BN [43]; and HFACS-PEHCA [44]. Different versions may produce different results, according to their research field. However, they all lie within the four categories of the framework: unsafe acts, preconditions for unsafe acts, unsafe supervision, and organization influences. Chen et al. [45] used multidimensional association rules to explore the causation chains of human factors in ship accidents. Because this method comes under the framework of the reason model, the human factors are within four layers: organizational factors, organizational prevention and supervision, the premise of unsafe behavior, and unsafe behavior. Cordon et al. [46] proposed a structural equation-modeling technique with exploratory factor analysis and confirmatory factor analysis to identify human factors in seafaring. They identified five factors: situation awareness, adaptability, group skills, self-knowledge, and drive. Furthermore, Maya et al. [47] proposed the card-sorting model, where human factors are classified into commercial pressure; the effect of environmental and external factors; improper design, installation, and working environment; inadequate leadership and supervision; lack of communication and coordination; lack of, improper, or late maintenance; lack of training; safety culture; safety management system; and unprofessional behavior. Coraddu et al. [37] proposed an approach to determine the most important human factors in maritime accidents, collecting 94 contributing factors. They identified the top 10 factors but did not clarify whether the contributing factors can be equated to human factors. Excluding the top 10 human factors, it was ambiguous whether the remaining contributing factors were included. Cai et al. [48] presented a dynamic Bayesian network (BN) combined with a pseudo fault tree to assess human factors. Three categories of human factors were identified: individual factor, organizational factor, and group factor. Furthermore, considering human factors, Sotiralis et al. [49] developed a BN-TRACEr model to quantitatively analyze ship operation risk. They identified 398 performance-shaping factors (PSFs), which were divided into six categories: aspects of communication/information; internal/external environment; organizational factors; training/competence; and personal factors. However, whether PSF equates to human factors remains unknown. Fan et al. [50] developed a BN-TOPSIS model to explore maritime accident prevention strategies from the human factor perspective. They contributed to recommended prevention strategies but gave no information regarding the definition and scope of human factors. The same situation was found in other works [51,52].

Literature Review

The six review articles covered the development or research status of human-factor-related issues. Chauvin C. [41] reviewed popular models and concepts of human factors and proposed a three-layer category of human factors in maritime accidents: cognitive factors, interpersonal factors, and organizational factors. None of the other five papers defined the specific content of human factors. Nevertheless, these review articles are still essential to the present research as they were used as references to check if any paper types were missing. They were also used as a source to find secondary articles, as discussed in Section 2.2.

3.2.3. Discussion

The status of the literature provides an overview of the scope of human factors. Because there are no agreements on this point, we aimed to conduct a systematic analysis and provide judgment to define its boundary and scope.

The conclusions drawn from Cordon, Cai, Sotiralis, and Chauvin are neutral, while those from the HFACS group, Chen, and Beatriz’s comments are biased toward the negative aspects of the human factor. However, the IMO document identified the negative aspects (adversely affecting the proper functioning of a particular system) and the positive aspects (enabling the successful performance of a particular task). We believe that the human factor is a neutral term in maritime safety, expressing both positive and negative aspects. The scope and relationship of Cordon, Cai, Sotiralis, and Chauvin’s judgments were studied and are indicated in Part A of Figure 4. Part B shows the results of the HFACS group, as well as Chen and Beatriz’s comments. Part C displays the judgment of this research. The connections between the boxes in Parts A and B represent similar (or the same) meanings.

3.3. Human Error

This section can be started by citing one sentence [36]: “The goal of human factors is to reduce human error, increase productivity, and enhance safety and comfort with a specific focus on the interaction between a human and the thing of interest” This suggests that human factor and human error are not the same.

3.3.1. Definition

In addition to the IMO’s definition, some scholars have also defined human error.

Table 4. Definitions of human error.

Ref.	Definitions
Lorenzo, D.K. [53]	Any human action, or lack thereof, that exceeds or fails to achieve some limit of acceptability, where limits of human performance are defined by the system.
Sanders et al. [54]	Human error is an inappropriate or unacceptable human decision or action that degrades efficiency, safety, or system performance.
IMO resolutions A. 884(21) [33]	Human error is a departure from acceptable or desirable practice on the part of an individual or group of individuals that can result in unacceptable or undesirable results.
Do-Hoon Kim [55]	Human error is defined as the failure to perform a prescribed task or performing a prohibited activity, with consequences that can result in serious injury and property loss, as well as near-miss incidents.
DiMattia et al. [15]	Human error is treated as a natural consequence arising from a discontinuity between human capabilities and system demands.
Read et al. [13]	Human error is a non-observable construct, used to make causal inferences, without clarity on the mechanism behind causation.
Gordon, RPE [10]	Human error refers to acts which are judged by somebody to deviate from some kind of reference act.

presents the definitions. The authors of [15,33,53,54] emphasize inappropriate actions or decisions and unwanted results. Furthermore, the authors of [10,13,55] focus on its structure and function, rather than mentioning human performance and consequences. From the reader’s viewpoint, the former four definitions are more specific and easier to understand, while the latter three are more abstract and macroscopic. These two types of explanations are not contradictory, but they do interpret one factor differently.

3.3.2. Research Status

The research here is similar to that in Section 3.2.2. The numbers of collected articles are displayed in

Table 5. Collected articles titled human error.

Category	Number of Documents
Journal article	26
Conference paper	3
Dissertation	2

. Human error probability research is included in this section because it has the keyword “human error”. Three articles were literature reviews concerning human error, and the remainder were research papers. The detailed content of Table 5 can be seen in Appendix C. Notably, human-factor-related error and human-related error are equivalent to human error [56]

Research Article

The number of articles in this group was somewhat lower than those in the human factor group. Among the articles in this group, about 40% focused on human error probability (HEP). These studies seldom provide human error content but find the most probable human errors from the listed tasks. However, the remaining 60% discussed human error contents, even as a byproduct. Uğurlu et al. [57] used the AHP model to study the prevention of grounding accidents, presenting four categories of human error: team management errors, voyage management errors, application errors, and individual errors. Zhang et al. [16] developed a THERP–BN model to assess HEP on human autonomy collaboration. They identified 16 human errors and divided them into three groups: perception error, decision error, and execution error. Li et al. [56] proposed an association-rule Bayesian network (ARBN) method to analyze the influencing factors in collision accidents. Six human errors were identified and further divided into two categories: Type I refers to the occurrence of negligence errors, and Type II refers to the occurrence of judgment/operation errors. Akyuz et al. [58] developed a SOHRA model to quantify human error during bunkering operations. The identified errors included inadequate reports or monitoring, delayed or inadequate feedback, inadequate checking/inspection, and lack of proper execution. Ung S. T. [59] constructed an FTA-CREAM model to assess human error in oil tanker grounding accidents. Five types of human error were proposed as follows: planning failure, verification failure, individual crew failure, incorrect action, and monitoring failure. Furthermore, Kandemir et al. [60] proposed an MMOHRA model to study error-producing conditions in maritime operations. Two types of human error were identified: unsafe supervision (inadequate supervision, planned inappropriate operation, failure to correct known problems, supervisory violations) and unsafe acts (skill-based error, rule-based mistakes, knowledge-based mistakes, routine violations, exceptional violations).

Literature Review

The three review articles addressed the origins, state of science, and perspectives of human error in the maritime safety field. Wrobel’s research [11] involves the interchangeable usage of the three terms and their defects and was an influencing factor in why we carried out this research. This is similar to the situation in Section 3.2.2.

3.3.3. Discussion

Unlike the human factor, a neutral term, human error, as its name implies, focuses on errors. Section 3.3.2 summarizes a thorough study of the relevant articles. Part A of Figure 5 shows the collected research models and the contents of human error. Moreover, there are few connections between the different models. This research intends to pass judgment without omitting human error content. Part B of Figure 5 displays the results.

3.4. Human Reliability Analysis

In the maritime domain, articles about HRA primarily refer to models or approaches for studying human-related factors or errors from qualitative or quantitative perspectives. Typically, these papers are not titled using the three terms; however, if these articles were excluded, some related information would be missing, and the results of the present study would be affected. In contrast, for articles focusing on HRA, human factors and human error were involved in the research process, so treating them as studies of human factor or human error is not accurate. This is why this section on HRA exists.

Research Status and Discussion

HRA is an alternative expression of HEP. A state-of-the-art HRA survey has been published [61].

Table 6. Collected articles on HRA.

Category	Number of Documents
Journal article	24
Conference paper	7

shows the number of collected articles. Furthermore, detailed information on these articles is presented in Appendix D. Only one review article was found in this group [61]. Hence, the studies of the collected articles in Table 6 may pro-vide some human factors, human errors, or human element information but not in-volve their boundaries. This section could be a supplement to the contents of Section 3.1, Section 3.2 and Section 3.3.

Groth et al. [62] developed a performance-influencing factor (PIF) hierarchy to address the problems in HRA. PIF, also known as PSF, is a common expression used to describe the causes affecting human performance in different systems. Classified PIFs can be split into five groups: organizational-based factors, team-based factors, person-based factors, situation/stressor-based factors, and machine-based factors. The first three groups can be found in the scope of the human factor in Section 3.2.3. Machine-based factors can be found in the scope of the human element (ship factors). Furthermore, situation/stressor-based factors or situation-based factors are actually within the scope of the human factor (preconditions for personal/group/organizational factors). Zhou et al. [63] proposed an FBN-CREAM model to study HRA. The recognized common performance conditions were organized into three groups: management-related factors, environment-related factors, and work-related factors. Compared with the analyses in Section 3.1, Section 3.2 and Section 3.3 management-related factors fall within the scope of the human factor (organizational factors), while the other two fall under the human element description. Bicen et al. [64] used a SOHRA approach to calculate the HEP in a crankshaft overhauling operation. The most common errors were unreliable instruments, poor feedback, inadequate checking, and insufficient crew knowledge. Unreliable instruments fall within the range of the human element, and the other three are in the scope of human error, according to Section 3.3. Similar research can be found in Kandemir et al. [65] and Akyuz et al. [66]. Furthermore, other papers have reported HRA research with different models, such as the HEART [67], AHP-CREAM [68], and HENT [17] models.

3.5. Usage Characteristics of the Three Terms

The usage characteristics of the three terms were obtained by integrating information from the IMO documents and collected articles. Figure 6 shows the results. The connecting lines represent the former segments contained in the latter parts. Detailed information can be found in Section 3.1, Section 3.2 and Section 3.3.

Human element, human factor, and human error have six segments, five segments, and five segments, respectively. Human error is easier to identify when compared with the other two terms because its five parts have distinct characteristics. Perception errors, decision errors, and execution errors are the three stages of errors when performing tasks. Individual errors are supplemental in case some personal errors are missing. Team management errors refer to errors in organizational management.

Human element and human factor are more challenging to identify because there are intersections in their contents, especially those marked by red dotted boxes. Initially, the people factors (human element) and the personal factors (human factor) look alike, but little differences exist. People factors can be factors of a person, a group, or a team, whereas personal factors refer to factors of one person. However, group factors and team factors are also parts of the term human factor. Hence, distinguishing them from this viewpoint is challenging. The three parts (ship factors, external influences and environment, and working and living conditions) are easily identifiable factors. Therefore, if it is clear that a document will involve the above three parts, the term human element is suggested; otherwise, when facing ambiguous preconditions, the term human factor is preferred. The last two parts are shore-side management and organization on board, which have similar meanings to the organizational factors. Consequently, identifying the human element and the human factor from these two parts is impossible. Drive is a segment only for the human factor. Therefore, the human factor must be the first preference for documents focusing on the drive.

Hence, human error should be identified first. The human element and human factor should be determined according to different situations. If the documents face ambiguous preconditions, the term “human factor” should be the first choice; otherwise, “human element” is preferred. For documents exploring the drive, the term human factor is preferable. If the documents focus on people factors, shore-side management, or organization on board, the authors should decide which term they prefer to use.

4. Conclusions

Recently, extensive research has been conducted on the three terms. Existing information (documents, articles, and dissertations) suggests that when referring to any of the three terms in maritime safety, our attention tends to focus on adverse effects. This focused attention promotes the confused usage of the three terms. Herein, we conducted systematic research to identify the three terms. Figure 6 presents their usage characteristics. Furthermore, the results are presented in Section 3.5. Each term has its contents, and we can use the corresponding terms of the segments where the document is mainly involved. For the document that involved the segments in red dotted boxes in Figure 6, the authors should decide on the right term to use. Although other studies may produce different results, the main framework remains the same. The ambiguity of these three terms has existed for over three decades, a fact that is not easy to change. Notably, the results here are not perfect. For example, for the contents in the red dotted box in Figure 6, the authors must decide on the correct term according to various situations. Nevertheless, the other usage characteristics of the three terms are clearly defined. In academic papers, some words are often replaced with synonyms; we recommend that these three terms should not be substituted.

One limitation of this research is that the data used here are only in English; information from other languages was excluded. This could be improved by adding more information from other languages to enhance the credibility of the research findings. Another limitation is that the collected articles came from Web of Science, ScienceDirect, IEEE Xplore, and PQDD databases but did not include Scopus, Google Scholar, and other databases. The last limitation is the authors’ limited knowledge, which could cause some important information to be omitted. Future research on this topic will be conducted continuously.