1. Introduction
The container terminal plays an important role in both the transportation and trade sectors and also serves as visible indications of a country’s economic growth [
1]. This has, however, led to the domination of this sector in the world by Asian countries [
2]. Some of the largest container terminals in the world are competing to improve terminal productivity, some of the approaches conducted are with integrating their quay cranes with an internal truck [
3] and integrating collaborative of outsourcing strategies for yard truck assignment [
4]. Even in the modern container terminal, several measures have been performed to increase energy efficiency [
5,
6] and reduce emissions at the port [
7,
8]. There are complex work processes on a container terminal and they include stevedoring, delivery, and receiving, each of which is contributing to the unexpectedly high risk experienced by people or actors [
9], the environment, or properties and facilities at the port [
10]. Based on data released by the Hong Kong Maritime Department in 2016, there were 76 cases of work accidents at the loading and unloading section of the port. Those related to work consisted of 60 minor, 15 serious, and 1 fatal accident [
11]. Another data from the United Kingdom Health Safety Executive in 2010–2011 also recorded 392 cases due to container loading and unloading [
12] while Petros L. Palli’s study on port risk management in container terminals reported a total of 1498 victims due to work accidents at container terminals [
13] and 45 incidents of environmental pollution due to oil spills which were equivalent to 3.5 tons from 2008 to 2011 at Piraeus and Thessalonica Ports, Greece [
14].
In Indonesia, accident statistics data issued in 2016 recorded 4 accident cases with 23 fatalities in 2012 and 6 cases with 75 fatalities in 2013. The data from 2014 to 2016 recorded 7, 11, and 15 accidents with 26, 87, and 69 fatalities respectively [
15]. According to Darbra Casal’s study, the majority of accidents represented by 56.5% on the ports were due to cargo transportation, 14.9% by loading and unloading operations, and 15.9% because of human error [
16]. These data showed the potential risk at the container terminal is also high, therefore, there is a need for effective risk management with the ability to prevent and handle these potential risks.
Safety and security have been a concern in container shipments over the past few decades [
17] and this has led to the use of risk management in the decision-making process of health care, environment [
18], and physical infrastructure systems [
19]. Also, several scientific materials have also been implemented in decision-making to develop theories, methodologies, and equipment required to assess these risks [
20]. Moreover, there is a need to apply these efforts in handling safety at the port, especially at the dock, [
21] to reduce or eliminate the effects of certain environmental, weather, human, social, and economic factors [
22] such as compensation for damage or death, prevention of pollution, and other medical expenses [
23,
24]. The port provides information, costs [
25], and facilities required by consumers to use container loading and unloading services and also considers the accident rate and material damage [
26,
27].
A lot of studies have been conducted on risk management using different approaches. For example, the assessment of the multi-criteria risk of Analytical Hierarchy Process requires the use of an opinion approach to defining a list of key risks [
28,
29] while a fuzzy reasoning method was used to measure the risk and cost-benefit analysis of the control measures [
30,
31]. Moreover, the risk factors related to port operations and management have been described and evaluated in Iran [
32] while other studies have used the average value approach and stochastic dominance method to determine the level of risk factors at the port of Taiwan based on questionnaire survey data [
33,
34]. The event tree analysis approach was also reported to have presented a multi-agent model to solve the problems of port management and logistics chains at the ports of San Antonio and Valparaíso [
32,
35]. Also, the formal safety assessment (FSA) approach assessed the possibility of port risks with due consideration for several factors and their mutual effects as well as the evaluation of risk management by considering the economic effects of the Risk Control Option (RCO) [
36]. This approach was implemented in the ports of Piraeus and Thessalonica. Regardless of the risk approach used, several advantages are depending on the type of analysis—quantitative, qualitative, or hybrid which could be through the combination of quantitative-qualitative or semi-quantitative [
37]. From the existing literature review, there is much research on risk assessment in container terminals. From the existing research, most of the study focuses on risk analysis to assess the safety performance of the terminals. The risk identification is carried out within the limits of qualitative assessment, without investigating the root cause of the problem. In its application in the field, terminal operators need to know the root cause of the risk. Identification of the root causes and its control options will be a novel contribution to the terminal operator. Thus, there is still a research gap that focuses on the root cause of the risk analysis in the container terminal, the method that can be used is to investigate using Fault Tree Analysis.
The purpose of this study was to identify accidents and potential risks at the one of the major container terminals in Indonesia by assessing the causes using the Fault Tree Analysis (FTA). This approach provides a combination of causes underlying the main causes of risk. This research grouped the accidents occurring at the container into categories commonly used by port operators to ensure it is easily understood while the classification of the risk causes into taxonomies was to make the determination of the appropriate mitigation process easy to achieve more effective and efficient implementation.
2. Risk Analysis Methodology
The risk analysis method performed in risk management will depend highly on data availability, information sources and the purpose of the risk analysis carried out. This study uses the Formal Safety Assessment (FSA) method with a Fault Tree Analysis (FTA) risk assessment. The FSA is a structured, systematic methodology designed to enhance safety at sea through risk analysis and cost assessments, including the protection of life, the environment and property. The FSA method was introduced by International Maritime Organization (IMO) as a rational and systematic process for accessing risks related to maritime safety and port environmental protection and for evaluating costs and verifying options to reduce risk. Based on the recommendations from IMO, the FSA process consists of five phases: identification of hazard, risk-assessment, risk-control options, cost-benefit evaluation, and decision-making recommendations. The explanation of each stage in detail is explained in the following sections.
2.1. Data Collection
The Formal Safety Assessment (FSA) method was combined with the Fault Tree Analysis approach to assessing the risks at the operation terminals 2 and 3 of Port of Tanjung Priok (PTP), Jakarta, Indonesia. The focus was on the cargo activities including gate and yard operations in the container yards to the delivery/receiving section. The data obtained consisted of primary data and secondary data. Primary data is data obtained by observing and research directions in the field. Data obtained from interviews and direct observations in the field. Secondary data is data that cannot be collected directly by researchers. This data can consist of company documents or other supporting data needed to complete the research. This data is an accident data that occurred at the container terminal for the last 5 years (2013–2017).
The interview process is conducted by two interviewers who have an engineering background and the interviewees are terminal operator employees. The duration of the interviews is about one month on the weekdays. The number of interviewees is about three to five employees depend on the container handling locations. There are at least four profiles of interviewees namely administrative officers, container yard operators, cranes operators, and truck operators. The interview process is carried out as follows, before conducting the interview the interviewer has collected the historical accident data in the terminal. The data is then grouped into several question segments that lead to the causes of accidents, there are six categories of causes of accidents which will be discussed in the next section. The questions given are in the form of open questions, where the interviewee is free to express anything related to the cause of the accident. From the result during the interview, the interviewer then groping the answer into the data presented in this paper.
The results of data collection in the field in the form of hazard identification contained in the container terminal as many as 23 data. There are 6 primary data, namely construction project fires, damaged bridge scales, dug-up trucks, docks damaged when the ship rests, containers received are damaged, workers are electrocuted. For secondary data, the results of hazard identification are included in the category of fire, loading, and unloading accidents, traffic accidents, non-loading accidents, work accidents, the environment. From the whole data is processed and presented in the next section.
2.2. Risk Identification
This stage involves the identification and classification of all relevant hazards and risks using work accident data at the container terminals between 2013–2017 as the basic guideline. The categories are as follows:
Categorizing the hazards identified into several general categories by considering who and what is losing. Also, the percentage of accidents causes was calculated based on the consideration that if 1 accident is caused by 1 cause, then the value of the cause is 1 but if 1 accident is due to n causes of an accident, then the value of each cause is 1/n. This calculation method is a percentage of the causes of accidents using the basic concepts of fractions and frequency distribution.
2.3. Severity
Severity is the result of an incident usually stated as a loss of risk and assessed using a numerical table. This study only considered severity concerning safety and environment while the other variables were ignored, therefore, a simple method was applied, and the parameters used are as shown in
Table 1.
2.4. Likelihood
Likelihood describes the frequency level of risk that might occur for 5 years or reaching 10,000 incidents depending on the number of workers and assets as well as the breadth of the area, and others.
Table 2 shows the parameters to assess the frequency level of risk.
2.5. Risk Matrix
Analytically, the 5 × 5 Risk Matrix reflects potential variations in frequency greater than the consequences. However, the facilitation of the ranking and validation of the ratings requires the consequences and frequency index are defined on a logarithmic scale. Moreover. the “risk index” is formed by combining the Frequency index and the Severity Index using the following equation:
2.6. Risk Assessment
The risk assessment aimed to comprehensively investigate the causes and consequences of the identified scenarios. The FTA method used describes and analyzes the cause of risk up to the root and this is the reason it is mostly applied as a tool for analysis, system evaluation, and decision making. It consists of several diagrams to show the information contained therein.
2.7. Risk Control Options
The risk control options in this study were to propose effective and practical risk control options (RCO). The first step was identifying and controlling potential risk while the second was evaluating the effectiveness of RCO in reducing the risk.
2.8. Results and Recommendations
The results obtained were used to make relevant recommendations in order to improve safety at the Container Terminal as indicated in the final chapter of this paper.