Considerations on Prevention of Pollution from Ships in a Seaport

Abstract

Negative effects on the port environment can originate from ports’ hinterland, ports’ activities and operations, and from ships. According to the available literature and long-lasting experience of numerous ports presented in different sources, pollution from ships belongs to the group of environmental priorities in ports. After a theoretical introduction where the importance of ports, their development, and the challenges/risks faced by ports (with special attention to environmental risks) are analyzed, an overview of the literature from the domain of prevention of pollution from ships is presented. Based on the standard structure of the PDCA (Plan, Do, Check, Act) management cycle, in this paper, a process model of the management (sub)system of prevention of pollution from ships in a port is proposed. Key demands related to the modeled (sub)system and bases which directly determine those demands are made concrete through an analysis of a case study: the Port of Bar (Montenegro). A categorization of domains of risks of pollution from ships in the analyzed port is conducted using the Analytic Hierarchy Process (AHP) method: considerations are based on a three-level hierarchy model: “0”—goal; level “1”—criterion; and level “2”—choice. The results of the analyses show that a domain with the highest rank of risk of pollution from ships is variant (alternative) M1: unloading oil/oil derivatives from ships to tanks (and vice versa), with a composite weight of 0.5365 (53.65%). The results of considerations presented in this paper can be used in a process of optimization of the (sub)system of prevention of pollution from ships in a multipurpose port as well as a reliable base for further research in this domain.

1. Introduction

The high intensity of changes that ports face nowadays generates more and more complex requirements that must be fulfilled by them in order to maintain/improve the port’s market position. All of this is accompanied by numerous challenges, especially when considering smaller ports, which often face a lack of resources and limited investment opportunities in order to reach the necessary level of adaptation to the mentioned changes. The aforementioned is specifically visible in the domain of environmental protection in ports, where requirements for the reduction in emissions of gases that damage the ozone layer, reception and treatment of liquid and solid waste from ships, handling of hazardous waste, noise reduction, optimization of relations with the local community (elimination of the impact of port activities on populated areas), etc., day by day are becoming more and more strict.

One of the definitive priorities in the field of environmental protection in cargo ports open to international maritime traffic is the optimization of the prevention system against pollution from ships. This is one of the main initial motives of the author to write this paper, using an approach which, by the method of deduction, starts from the general importance of the port and is finalized with the identification of risk categories of sea pollution from ships in the concrete port, which is analyzed within a case study.

Ports have critical importance for the global supply chain [1,2,3] and have a significant impact on economic activities in the country to which they belong and their wider hinterland [4,5,6,7,8]. The capability of ports to confirm their crucial importance directly depends on the adequacy and intensity of their development. Port development appears as a research problem in many of the available references, where different aspects of port development are taken into consideration, as shown in Table 1.

Table 1. Port development—a literature overview.

Research Problem	Reference
Construction of future port development scenarios
- A tool for defining a group of possible “futures” for a port	[9]
- Potential changes in ports up to 2050 under four climate-based scenarios	[10]
- Identification of drivers of changes and definition of four scenarios of a concrete North American port development up to 2050	[11]
- Approaches for generating scenarios of seaport terminal development	[12,13]
- Models of port development according to the degree of private investors’ involvement	[14]
- Implications of different strategic choices of port development	[15]
- Modeling the strategic development of a port’s collaboration with local operators and logistic subjects from its hinterland	[16]
Correlation between port development and its competitiveness	[17]
Port connectivity as one of the key influential factors on port development	[18]
Different aspects of port land development	[19,20,21]
Port–city relations	[22,23]
Implementation of an adaptive model in managing port development	[24]
Correlation between port development and economic development of its hinterland	[25]
Port regionalization phase in port development	[26]
Development of a “smart port”
- Importance of the development of a smart port and port 4.0	[27]
- Future prospects of digitalization in ports	[28]
Development of a “green port” (port sustainable development)
- Introducing energy efficiency	[29]
- Circular economy transition of ports	[30]
- Development of a low-carbon port	[31]
- Use of climate technology, including carbon removal solutions	[32]
- An approach to the sustainable development of a port in circumstances characterized by nearby underexploited cultural and tourist capacities	[33]
- Incorporating advanced technologies connected with the fourth industrial revolution as a strategic direction towards the sustainable development of modern ports	[34]

(Source: author).

Port sustainability includes internal (port side) and external (ships and land transport) actions and measures [35]. In general, the concept called “green port” aims to balance economic and ecological port operations [36]. The European Green Deal [37] calls for a 90% reduction in greenhouse gas (GHG) emissions from transport, in order for the EU to become a climate-neutral economy by 2050, while also working towards a zero-pollution ambition. The way of reaching the defined targets is very challenging, as can be illustrated by an example. Following the 80th session of the IMO’s Marine Environment Protection Committee (MEPC) in July 2023, the revised strategy to reduce GHG emissions from ships includes a commitment to reaching net zero “by or around” 2050. The previous target was a 50% reduction in GHG emissions by 2050, compared to 2008 levels [38]. The disruption in the Red Sea and Suez Canal combined with factors linked to the Panama Canal and the Black Sea and leading to the rerouting of vessels through longer routes are causing vessel sailing speeds to increase [39], implying increasing GHG emissions for a round trip.

In their daily functioning and especially through the development process, ports are facing different challenges/risks: the existing and forecasted effects of climate change [9]; the changing nature of shipping (the size and the complexity of the fleet are increasing; energy trades are changing) [40]; growing risks; etc.

According to ISO 31000 (2018) [41], risk is the effect of uncertainty in achieving the objectives, often quantified as the likelihood of the occurrence of an event multiplied by its impact (L × I) [42]. The ISO 31000 provides principles and generic guidelines on risk management that are applicable not only to seaports but to any organization or individual [43]. Risk analysis in seaports plays a crucially important role in ensuring port operation reliability, transportation safety, and supply chain resilience [44].

Risk analysis in a port, among others, takes into consideration such factors as [45] spill response equipment availability, tug assistance availability, fire-fighting service and equipment availability, depth of water in the port and its physical characteristics, ecological sensitivity of the coastal environment, safety and security considerations, etc. There are various groups of factors that could lead to port disruptions. One of these groups is environmental factors: pollution, seismic events, adverse weather, hydrological hazards, etc. [42].

Environmental risks can arise at all stages of the port and terminal life cycle [46] and involve [47,48,49,50,51,52] ship emissions, dust emissions, dredging, oil spills, chemical contaminants, ballast waters, noise pollution, alien species, etc. Forms of concretization of mentioned risks categories—negative effects of pollution could be very different. If the pollution from ships is focused, it is necessary to point out that contamination of sediments in the port water area is one of the principal negative consequences of that pollution.

From the general point of view, the port environment is negatively influenced by the ports’ hinterland, ports’ activities and operations, and ships [53,54,55,56,57,58,59].

The principal objective of this paper is to create/propose (respecting the relevant literature and verified methodology) a process model of the management (sub)system of prevention of pollution from ships in a port and to categorize the risks of pollution from ships determined with a group of influential factors.

After a general introduction, Section 2 analyzes the evolution of the environmental priorities in European Union ports. A literature overview on prevention of pollution from ships is given in Section 3, and a process model of the prevention of pollution from ships in a port, based on the PDCA management cycle, is structured in Section 4. Section 5 elaborates on the prevention of pollution from ships though a case study—the Port of Bar. Within Section 5, special attention is given to the basic characteristics of the Port of Bar and to factors that determine requirements related to the (sub)system of prevention of pollution from ships in that port as well as to the categorization of the domains of risk of pollution from ships, which is performed using the Analytic Hierarchy Process (AHP) method. In Section 6, key conclusions are given.

2. Evolution of the Environmental Priorities in European Union Ports

The evolution of the environmental priorities in European Union (EU) ports in the period from 2019 to 2023 is shown in the next table (Table 2).

Table 2. Evolution of the environmental priorities in European Union ports.

Priority	2019	2020	2020/ 2019	2021	2021/ 2020	2022	2022/ 2021	2023	2023/ 2022	2023/2019	Average Rank
AQ—air quality	1	1	const.	1	const.	2	+1	2	const.	+1	1, 4	1
EE—energy consumption	2	3 *	+1	3	const.	3	const.	3	const.	−1	2, 8	3
CC—climate change	3	2	−1	2	const.	1	−1	1	const.	−2	1, 8	2
NO—noise	4	4	const.	4	const.	4	const.	4	const.	const.	4	4
RLC—relationships with the local community	5	5	const.	5	const.	6	+1	7	+1	+2	5, 6	5
SW—ship waste	6	6	const.	7	+1	7	const.	6	−1	const.	6, 4	6
PW—garbage/port waste	7	8	+1	10	−2	8	−2	9	+1	+2	8, 4	8
PDL—port development (land-related)	8	10	+2	9	−1	9	const.	8	−1	const.	8, 8	9
DOs—dredging operations	9	9	const.	8	−1	10	+2	Port development (water-related)	-	-	-	-
WQ—water quality	10	7	−3	6	−1	5	−1	5	const.	−5	6, 6	7

(Source: author, based on [60].) * In the year 2020, priority “energy consumption” was renamed to “energy efficiency”.

In Table 3, environmental priorities in the ports of the European Union (EU) in 1996 and 2023 are compared (based on [60]). The key conclusion is that dramatic changes have happened in the 27-year period: only three among the top ten priorities from 1996 are still actual in the year 2023: port development—water-related (first in 1996, tenth in 2023); water quality (second in 1996, fifth in 2023); and port development—land-related (sixth in 1996, eighth in 2023).

Table 3. Comparison of environmental priorities in ports 2023/1996.

Priority—1996	Rank—1996	Priority—2023	Rank—2023
PDW—port development (water-related)	1	CC—climate change	1
WQ—water quality	2	AQ—air quality	2
DD—dredging disposal	3	EE—energy efficiency	3
DOs—dredging operations	4	N—noise	4
D—dust	5	WQ—water quality	5
PDL—port development (land-related)	6	SW—ship waste	6
CL—contaminated land	7	RLC—relationships with the local community	7
HL-D—habit loss/degradation	8	PDL—port development (land-related)	8
TV—traffic volume	9	PW—garbage/port waste	9
IEs—industrial effluents	10	Port development (water-related)	10

(Source: author, based on [60]).

Respecting the previously mentioned, it is an obvious fact that potential pollution from ships belongs to the group of environmental priorities in ports. It is closely connected with all the top 10 priorities analyzed in Table 2. This is a reason why pollution from ships in a port is taken as an object of further analyses in this paper.

3. Prevention of Pollution from Ships—A Literature Overview

The available literature from the domain of the prevention of pollution from ships is related to the very wide range of objects of research, as shown in Table 4.

Table 4. Prevention of pollution from ships—a literature overview.

Research Problem	Reference
International Maritime Organization (IMO) legal instruments
- Bibliography and index to explore the role of the IMO in facilitating the adoption and implementation of international legal instruments for the protection of the marine environment	[61,62,63]
- Marine pollution under the Law of the Sea Convention	[64]
- How international law has regulated the protection of the marine environment from pollution from ships	[65]
- The international legal framework of the prevention of vessel-sourced marine pollution	[66,67]
- The effectiveness of the MARPOL convention	[68]
- Some aspects of correlation between the survival of the human species and the maintenance of a clean and alive ocean	[69]
- A problem that developing countries, in general, have not successfully implemented IMO conventions	[70]
- Background of the MARPOL convention and elements related to its implementation	[71,72,73,74,75]
General aspects of sea pollution from ship-sourced pollution
- Vessel-sourced pollution as one of the major sources of marine pollution	[76,77]
- Delay in implementing strategies to address marine pollution	[78]
- Operating conditions of ship’s main pollution prevention equipment	[79]
- Importance of achieving the target for uniform international regulation for preventing and controlling ship-sourced pollution	[80]
- Key marine waste issues, taking into account sources, risks, transport pathways, legislation, current challenges, etc.	[81]
- Marine pollution caused by ship operations	[82,83,84]
- The preliminary feasibility of a particular waste recycling technology	[85]
Pollution from ships—oil
- Issues related to the identification and prevention of ship pollution, with special focus on oil pollution	[86]
- Issues related to pollution caused by oil spills into the sea, analyzing some major marine pollution incidents	[87,88]
- Various aspects of both land- and sea-based oil pollution	[89]
- Results of research based on a fact that over a million tons per year of oil is spilled from ships into the sea	[90]
- Adequacy level of the MARPOL convention implementation related to the pollution of the sea by oil from ships	[91,92]
Pollution from ships—solid and liquid waste
- The marine pollution caused by marine domestic waste	[93]
- Issues related to the implementation of international regulations (mainly MARPOL convention) for the control of pollution of the sea with solid and liquid waste from ships	[94,95,96,97]
Pollution from ships—emissions
- A fact that pollution from ships is not limited to the sea; it causes significant air and other forms of pollution	[98]
- The international regulations on ship emissions and their influence on level of SO2 emissions from Ocean Going Vessels (OGVs)	[99]
- Summarizing studies that address air pollution, with special focus on particulate matter from marine vessels	[100]
- Pollution of air by large ships in the hub ports (taking into account that low-grade marine fuel oil contains 3500 times more sulfur than road diesel)	[101]
- Public health and climate impacts of low-sulfur fuels in global shipping	[102]
- The status of pollution mitigation measures implemented to date in the shipping sector	[103]
- The contributions of ship emissions of NO2, SO2, PM10, and PM2.5 to air quality in the ports	[104]
- Annex VI (of the MARPOL convention) issues (influence of shipping industry on climate change)	[105,106]
- Initial strategy for the reduction in GHG emissions from ships in a port	[107]
- Air quality management in main Mediterranean ports	[108]

(Source: author).

The research whose results are shown in the analyzed literature covers a very wide geographical area: developing countries [61,70], Indonesia [65], India [71], Nigeria [74,84], China [76], Malaysia [77], the world ocean [78,86], the Arctic region [80], Cambodia [89], Vietnam, the USA, Australia [94], the Southern North sea [99], Hong Kong [101], the Mediterranean sea [104,108], Kenya [105], and the Caspian sea [107]. Obviously, the references analyzed in Table 4, as per the related geographical areas, cover almost all continents and the world ocean and indicate what the priorities are in the domain of the prevention of pollution from ships in those areas. A further classification of the analyzed references was performed, recognizing the following groups: group 1—related to the prevention of pollution from ships in a concrete port [101]; group 2—related to the prevention of pollution from ships in a concrete country [65,71,74,76,77,84,89]; group 3—related to the prevention of pollution from ships in a group of countries/regions [61,70,80,94]; and group 4—related to the prevention of pollution from ships in the world ocean or its segments [78,86,99,104,107,108].

The methods used to conduct the research in the analyzed literature were the Gaussian Plume Dispersion Approach [53], dynamic method [54], ratio-matching method [56], qualitative research methodology [65], statistical tools [68], multiple criteria decision making method [82], fuzzy Bayesian network [83], instrument of questionnaires [84], system analysis [86], normative juridical legal research method [87,105], and integrated atmospheric models and health risk functions [102].

4. Process Model of Prevention of Pollution from Ships in a Port

Based on the standard structure of the PDCA management cycle [109], Figure 1 proposes a process model of the management (sub)system of the prevention of pollution from ships in a port (the term “(sub)system” is used due to fact that it is a segment of a port’s overall environmental management system).

Figure 1. Process model of a (sub)system of prevention of pollution from ships in a port. (Source: author, based on [109]).

The key general demands related to the (sub)system of prevention of pollution from ships in a port are adequate organization, the availability of qualified personnel, the availability of adequate equipment, etc. In order to establish bases for enabling the optimal functioning of the analyzed management sub(system), it is of crucial importance to recognize factors which directly determine the mentioned demands.

Speaking from a general point of view, sustainable development is a path which can be achieved through environmental management systems (EMSs).

The best-known standards setting out the criteria for an environmental management system (EMS) are ISO 14001 [110,111] and the EMAS (recommended by the European Parliament and the Council of the European Union, 2009) [112]. ISO 14001 provides a clear management framework to reduce environmental impacts, to meet legal requirements, to build stakeholder trust [113], and to integrate environmental management practices by supporting preventive and corrective measures [114]. It specifies requirements for an organization to identify the environmental aspects of its activities, products, and services and correlated environmental impacts. ISO 14001 is designed to be compatible and harmonized with other recognized management system standards, including ISO 9001 [113,115]. ISO 14001 requires organizations to define and execute their own operational controls in a manner appropriate for the sector they operate in [116]. Standards are not port-specific [112].

Ports, depending on their size, tend to have a wide range of environmental (sustainability) management needs. Small ports do not often have sufficient resources to implement environmental effectiveness enhancing tools, even if they need them [117].

5. Prevention of Pollution from Ships. Case Study: Port of Bar

5.1. Basic Characteristics of the Port of Bar

The Port of Bar [118] is a landlord port. In the port area, two port terminal operators are currently operating. There are the following specialized terminals within the Port of Bar: Liquid Cargo Terminal, Dry Bulk Cargo Terminal, Container Terminal, General Cargo Terminal, Ro-Ro Terminal, and Passenger Terminal.

In accordance with national strategic documents, laws, by-laws, and elements of international regulation, necessary activities are carried out in the Port of Bar with the aim of developing the function of environmental protection in accordance with the principles of sustainable development.

5.2. Factors That Determine Requirements Related to the (Sub)System of Prevention of Pollution from Ships in the Port of Bar

In this section, factors which determine requirements related to the (sub)system of prevention of pollution from ships in the Port of Bar are identified and analyzed.

(A)

Regulative: national and international

Basic elements of the national (Montenegrin) legislation related to the prevention of pollution from ships in the ports are the following: the Law on the Protection of the Sea from Pollution from Ships [119]—regulates the protection of the sea from pollution from ships that sail or are located in the internal sea waters and the territorial sea of Montenegro and the reception and handling of waste in ports, as well as responsibility and compensation for damages in case of pollution—and the Law on Ports [120]—regulates the legal status, division of ports, management, fees, concessions, order, inspection, and other matters of importance, including basic provisions on environmental management.

The important conventions which regulate at the international level the prevention of pollution from ships are the following: MARPOL—Convention on the Prevention of Pollution from Ships; AFS—Convention on the Control of Harmful Systems Against Ship Fouling; OPRC—Convention on Preparedness, Response and Cooperation in case of oil pollution; BWM—Convention on the Monitoring and Management of Ship’s Ballast Water; and the Nairobi Wreck Removal Convention.

In this context, it is important to point out the directives and regulations of the European Parliament and the Council related to the prevention of pollution from ships [121]: Directive 2002/84/EC (amending the directives on maritime safety and the prevention of pollution from ships); Regulation EC No2002/2099 (establishing a Committee on Safe Seas and the Prevention of Pollution from Ships); Directive 2000/59/EC (on port reception facilities for ship-generated waste and cargo residues); Directive 2001/96/EC (establishing harmonized requirements and procedures for the safe loading and unloading of bulk carriers); etc.

(B)

National response plans related to the prevention of pollution from ships and documented risk scenarios referring to sea pollution in ports from ships

In the document Risk Assessment of Disasters in Montenegro [122], general risks related to maritime traffic are analyzed. It is stated there that the set of potential risks with potential to cause disruptions to maritime transport and port activities in Montenegro also include those related to the Port of Bar and the ships that use this port: damage and accidents on ships that are berthed in the port, spillage of oil derivatives and harmful substances in the port water area, etc. In the mentioned document [122], some scenarios of unwanted events are elaborated.

(C)

Port capacity (including port terminal operators) to protect the sea from pollution from ships

The main activities of port terminal operators in the domain of the protection of the sea from pollution from ships are [118] the monitoring of activities in the domain of the protection of the sea from pollution from ships which, according to the corresponding contract, are entrusted to the operator for the management of liquid and solid waste from ships at the port area.

(D)

Competencies of the operator for the management of liquid and solid waste from ships in the port

The operator for the management of liquid and solid waste from ships in the port is authorized by the competent state authorities, and the port terminal operators have signed commercial agreements on mutual rights and obligations in relation to preventive measures (installation of a protective dam, acceptance of solid and liquid waste from ships, etc.) and corrective activities (procedure in case of environmental incidents, etc.).

(E)

Sources of income for financing activities in the field of protection from pollution from ships in the port

Port terminal operators finance activities in the field of protection from pollution from ships from their own revenues. In this context, it is important to mention that the Law on Ports [120], among others, defines additional sources of financing the protection of the sea from pollution from vessels in Montenegrin ports.

(F)

Waste management plan

The waste management plan is one of the basic documents that determine the goals and provide the conditions for sustainable waste management in the area of port terminal operators. According to the Law on Waste Management [123], the plan fulfills one of the important prerequisites for minimizing the amount of waste, thus reducing business costs through the optimized use of resources and reduction in waste disposal costs.

(G)

Material Safety Data Sheets for dangerous goods

The Material Safety Data Sheet (MSDS) represents one of the bases for the organization of the environmental protection system from pollution from ships during the (un)loading of hazardous materials that are transported to/from the port. The content of the MSDS is defined by the Montenegrin Law on Chemicals [124] and the Rulebook on the Content of the MSDS for Chemicals [125].

(H)

Cargo handling technologies in the relation from ship to shore (and vice versa)—in the function of preventing pollution from ships

Documented cargo handling technology [118] is also an important basis for the organization of sea protection against pollution from ships.

(I)

Number of ships

In Table 5, the number of ships which called at the Port of Bar JSC (one out of two terminal operators) in the period from 2019 to 2023 is shown.

Table 5. Number of ships which called at the Port of Bar JSC in the period from 2019 to 2023.

Parameter	Values of Parameter per Year
		2019	2020	2021	2022	2023
Number of ships		549	476	400	363	544

(Source: author, based on [118]).

(J)

Volume of oil/oil derivatives handled in the relation from ships to tanks (and vice versa)

In the following table (Table 6), systematized data related to the volume of oil/oil derivatives handled in the Port of Bar in the relation from ships to tanks (and vice versa) in the period from 2019 to 2023 are shown.

Table 6. Volume of oil/oil derivatives handled in the period from 2019 to 2023.

Parameter	Values of Parameter per Year
		2019	2020	2021	2022	2023
Throughput		221,592.94 t	167,147.94 t	234,538.98 t	258,785.60 t	244,660.45 t
Relative ratio—compared to 2019		1	0.75	1.06	1.17	1.10
Time of stay of ships by the quay, based on the average value of productivity (t/h)		44.32 days	33.43 days	46.91 days	51.75 days	48.93 days

(Source: author, based on [118]).

(K)

Volume of harmful liquid substances in bulk (chemicals) handled in the relation from ships to tanks (and vice versa)

Table 7 shows systematized data on the volume of harmful liquid substances in bulk (chemicals) handled in the Port of Bar in the relation from ships to tanks (and vice versa) in the period from 2019 to 2023.

Table 7. Volume of harmful liquid substances in bulk (chemicals) handled in the period from 2019 to 2023.

Parameter	Values of Parameter per Year
		2019	2020	2021	2022	2023
Throughput		11,000.88 t	0	0	15,010.86 t	0
Relative ratio—compared to 2019		1	-	-	1.36	-
Time of stay of ships by the quay, based on the average value of productivity (t/h)		1.34 days	-	-	1.83 days	-

(Source: author, based on [118]).

(L)

Volume of packaged dangerous goods handled in the relation from ships to warehouses (and vice versa)

Table 8 shows data on volume of packaged dangerous goods handled in the Port of Bar in the relation from ships to warehouses in the period from 2019 to 2023.

Table 8. Volume of packaged dangerous goods handled in the period from 2019 to 2023.

Parameter	Values of Parameter per Year
		2019	2020	2021	2022	2023
Throughput		0	3317.68 t	3486.24 t	2322.82 t	5797.39 t
Relative ratio—compared to 2019		1	/	/	/	/
Time of stay of ships by the quay, based on the average value of productivity (t/h)		-	0.46 days	0.48 days	0.32 days	0.81 days

(Source: author, based on [118]).

(M)

Results of categorization of the domains of risks of pollution from ships

5.3. Categorization of the Domains of Risks of Pollution from Ships

The results of the literature overview confirm that none of the available references take into consideration the categorization of the domains of risks related to pollution from ships in a port. In order to contribute to the completion of the group of bases for the optimization of the (sub)system of prevention of pollution from ships in the analyzed port (Port of Bar), within this section, a categorization of the domains of risks related to pollution from ships is conducted using the Analytic Hierarchy Process (AHP) method.

The usual approach to risk identification and categorization includes investigating historical data on previous incidents, in addition to a structured brainstorming process with practitioners/professionals/experts for conceivable risks [48]. Due to its suitability, the AHP method is chosen for the realization of the mentioned brainstorming process for risk categorization in the domain of the prevention of pollution from ships.

5.3.1. Methodology

The AHP method is used to study production systems, software, supplier selection, construction method selection, warehouse selection, technology evaluation, etc. Numerous references are available where research results (based on the application of the AHP method) are presented: port selection [126], port competitiveness [127], cargo handling equipment selection as part of the investment process [128], etc.

The AHP is a method intended for solving complex problems at different hierarchical levels, where the goal is at the top, the middle levels are criteria and sub-criteria, and the lowest level is alternatives (choice) [129]. In principle, the AHP is a general measurement theory, which is used to define the scale of ratios for both discrete and continuous paired comparisons [130]. These comparisons may be taken from actual measurements or from a fundamental scale which reflects the relative strength of preferences. The AHP method has found its widest applications in multi-criteria decision making, planning, and resource allocation [131].

The basic steps in using the AHP method for the categorization of the domains of risks of pollution from ships in the Port of Bar are as follows (based on [131,132]):

(A)

Defining the goal

The goal is to identify the domains of the risks of pollution from ships in the Port of Bar, taking into account the elements analyzed in the previous sections of this paper.

(B)

Defining variants (alternatives) and selection criteria

The following variants (alternatives) are considered:

• • Variant (alternative) 1—domain of risks M1: unloading oil/oil derivatives from ships to tanks (and vice versa);
  • Variant (alternative) 2—domain of risks M2: unloading harmful liquid substances in bulk (chemicals) from ships to tanks (and vice versa);
  • Variant (alternative) 3—domain of risks M3: unloading packaged dangerous goods from ships to warehouses (and vice versa);
  • Variant (alternative) 4—domain of risks M4: liquid and solid waste from ships/emissions from ships.

The following selection/comparison criteria were chosen:

• • Criterion 1—K1: workforce (number, profile, and level of education of personnel, etc.); the results of the research show that specific attention in this domain has to be given to the qualification, training, and attitude of the involved workforce [133].
  • Criterion 2—K2: equipment.
  • Criterion 3—K3: costs.
  • Criterion 4—K4: period of exposure to the risks.

(C)

Defining the hierarchical analysis model

The general form of the hierarchical analysis model is shown in Figure 2.

Figure 2. General form of the hierarchy model of analysis (source: author).

(D)

Pair-wise comparison and consistency test

In order to perform pair-wise comparisons and associated consistency tests, the following activities should be performed [132]—Table 9.

Table 9. Activities necessary to perform pair-wise comparisons and consistency tests.

Step	Description
Create a reciprocal (comparative) matrix based on the number of items for comparison, n, and Saaty’s fundamental scale of absolute numbers	Number of items for comparison
	Number of things				1		2		3		4			5		6		7		n
	Number of comparisons				0		1		3		6			10		15		21		n(n − 1)/2
	Saaty’s fundamental scale of absolute numbers
	Intensity of importance		Definition									Explanation
	1		Equal importance									Two activities contribute equally to the objective
	2		Weak or slight
	3		Moderate importance									Experience and judgement slightly favour one activity over another
	4		Moderate plus
	5		Strong importance									Experience and judgement strongly favour one activity over another
	6		Strong plus
	7		Very strong or demonstrated importance									An activity is favoured very strongly over another; its dominance is demonstrated in practice
	8		Very, very strong
	9		Extreme importance									The evidence favoring one activity over another is of the highest possible order of affirmation
Summing each column of the reciprocal (comparison) matrix
Dividing each element of the reciprocal (comparison) matrix with the sum of its columns and obtaining the normalized relative weights
Calculating the normalized principal eigenvector (Priority Vector), w, by averaging across the rows of the matrix
Calculating the principal eigen value λmax obtained from the summation of the products between each element of the eigen vector and the sum of columns of the reciprocal (comparison) matrix
Calculating the consistency index, CI, as a measure of deviation or degree of consistency using the following equation:	CI = (λmax − n)/(n − 1)																		(1)
	where λmax—principal eigen value; n—number of items for comparison.
Calculating the consistency ratio, CR, based on the values of the consistency index, CI, and random consistency index, RI (values from the table in the right column), according to the number of items to be compared; for the analyzed case, n = 4 ⇒ RI = 0.9, as per Equation (2) (if the value of the consistency ratio is under 0.10, then the evaluation is consistent)	CR = CI/RI																		(2)
	n	1		2		3		4		5			6		7		8		9		10
	RI	0		0		0.5		0.9		1.1			1.2		1.3		1.4		1.4		1.49
Calculating the overall composite weight of each alternative based on the weights of level 1 and level 2; in fact, the overall weight is just the normalization of the linear combination of multiplication between the weight and priority eigenvector;
Calculating the overall consistency of the hierarchy by summing all the levels, with the weighted consistency index, CI, in the nominator and weighted random consistency, RI, in the denominator, based on Relation (3) (if the value of the overall consistency ratio is under 0.10, the evaluation is consistent)	CR = (ΣwiCIi)/(ΣwiRIi)																		(3)

(Source: [132]).

(E)

Calculation of the global weights

The overall composite weight for the analyzed variants (alternatives)—domains of risks, M_i—is calculated based on the following relation:

M_i = Σ(relative weight of the criterion c_j, from the compar. matrix with respect to the goal) × (relative weight of the variant (alternat.) − domain of risks M_i based on criterion c_j, from compar. matrix with respect to criterion c_j)

(4)

(F)

Final ranking of alternatives

5.3.2. Results—Final Ranking of Alternatives

(A)

Pair-wise comparison and consistency test

In line with the defined phases of the methodology (previous section of this paper), reciprocal matrices were created and all related parameters were calculated, all in order to define the categories of the domains of risks of pollution from ships in the Port of Bar, among defined variants (alternatives), M_i, based on the adopted selection criteria, K_j.

• (A.1)

  Pair-wise comparison and consistency test—level 1 of the hierarchy framework

Pair-wise comparison was performed by an ad hoc established group of 10 professionals engaged in different forms in the field of the prevention of pollution from ships (besides the author of this paper, members of the group were from the following institutions: Montenegrin Ministry of Traffic and Maritime Affairs, Montenegrin Directorate for Maritime Safety and Port Management, Harbor Master Office Bar, and Adriatic University). In comparison matrices, the rounded mean values of ratings from each member of the group were inserted.

A reciprocal (comparison) matrix—level 1 of the hierarchy framework (Figure 2)—with respect to the goal—is shown in Table 10, which was formed considering the number of items for comparison, n = 4, and Saaty’s fundamental scale of absolute numbers. The values from each column of Table 10 were summed (bottom row of the table), and, after that, each element of the table was divided with the sum of the belonging column in order to obtain the normalized relative weights. In the next step, the normalized principal eigen-vector (Priority Vector), w, is calculated by averaging across the rows of the matrix.

Table 10. Paired comparison matrix—level 1—with respect to the goal.

Criterion	K1	K2	K3	K4	Priority Vector
K1	1.00	2.00	0.33	0.50	0.1735
K2	0.50	1.00	0.33	0.50	0.1230
K3	3.00	3.00	1.00	0.50	0.3274
K4	2.00	2.00	2.00	1.00	0.3760
sum	6.50	8.00	3.66	2.50	1.000

(Source: author).

In order to carry out the consistency test of the evaluation, it is necessary to calculate the consistency index, CI, and consistency ratio, CR. If the value of consistency ratio, CR, is under 0.10, the evaluation is consistent.

The calculated values related to the matrix shown in Table 10 are as follows:

Principal eigenvector: λ_max = 4.2505; consistency index, CI = 0.0835; random value of consistency index, RI = 0.9; and consistency ratio, CR = 0.0928. The consistency ratio is under 0.10, which means that the performed evaluation is consistent.

• (A.2)

  Pair-wise comparisons and consistency tests—level 2 of the hierarchy framework

Reciprocal (comparison) matrices—level 2 of the hierarchy framework (Figure 2)—with respect to the selection criteria are presented in Table 11, Table 12, Table 13 and Table 14. All previously mentioned characteristic parameters are calculated as per the previously defined steps of the procedure (Table 9).

Table 11. Paired comparison matrix—level 2—with respect to criterion C1.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	7.00	5.00	3.00	0.5668
M2	0.14	1.00	0.20	0.33	0.0598
M3	0.20	5.00	1.00	0.50	0.1644
M4	0.33	3.00	2.00	1.00	0.2090
sum	1.67	16.00	8.20	4.83	1.000

(Source: author).

Table 11. Paired comparison matrix—level 2—with respect to criterion C1.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	7.00	5.00	3.00	0.5668
M2	0.14	1.00	0.20	0.33	0.0598
M3	0.20	5.00	1.00	0.50	0.1644
M4	0.33	3.00	2.00	1.00	0.2090
sum	1.67	16.00	8.20	4.83	1.000

(Source: author).

Principal eigenvector: λ_max = 4.2606; consistency index, CI = 0.0869; random value of consistency index, RI = 0.9; and consistency ratio, CR = 0.0965. The consistency ratio is under 0.10, which means that the performed evaluation is consistent.

Table 12. Paired comparison matrix—level 2—with respect to criterion C2.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	7.00	5.00	3.00	0.5668
M2	0.14	1.00	0.20	0.33	0.0598
M3	0.20	5.00	1.00	0.50	0.1644
M4	0.33	3.00	2.00	1.00	0.2090
sum	1.67	16.00	8.20	4.83	1.000

(Source: author).

Table 12. Paired comparison matrix—level 2—with respect to criterion C2.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	7.00	5.00	3.00	0.5668
M2	0.14	1.00	0.20	0.33	0.0598
M3	0.20	5.00	1.00	0.50	0.1644
M4	0.33	3.00	2.00	1.00	0.2090
sum	1.67	16.00	8.20	4.83	1.000

(Source: author).

Principal eigenvector: λ_max = 4.2606; consistency index, CI = 0.0869; random value of consistency index, RI = 0.9; and consistency ratio, CR = 0.0965. The consistency ratio is under 0.10, which means that the performed evaluation is consistent.

Table 13. Paired comparison matrix—level 2—with respect to criterion C3.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	7.00	5.00	3.00	0.5668
M2	0.14	1.00	0.20	0.33	0.0598
M3	0.20	5.00	1.00	0.50	0.1644
M4	0.33	3.00	2.00	1.00	0.2090
sum	1.67	16.00	8.20	4.83	1.000

(Source: author).

Table 13. Paired comparison matrix—level 2—with respect to criterion C3.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	7.00	5.00	3.00	0.5668
M2	0.14	1.00	0.20	0.33	0.0598
M3	0.20	5.00	1.00	0.50	0.1644
M4	0.33	3.00	2.00	1.00	0.2090
sum	1.67	16.00	8.20	4.83	1.000

(Source: author).

Principal eigenvector: λ_max = 4.2606; consistency index, CI = 0.0869; random value of consistency index, RI = 0.9; and consistency ratio, CR = 0.0965. The consistency ratio is under 0.10, which means that the performed evaluation is consistent.

Table 14. Paired comparison matrix—level 2—with respect to the criterion C4.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	6.00	4.00	2.00	0.4918
M2	0.16	1.00	0.25	0.25	0.0626
M3	0.25	4.00	1.00	0.33	0.1527
M4	0.50	4.00	3.00	1.00	0.2929
sum	1.91	15.00	8.25	3.58	1.000

(Source: author).

Table 14. Paired comparison matrix—level 2—with respect to the criterion C4.

Criterion	M1	M2	M3	M4	Priority Vector
M1	1.00	6.00	4.00	2.00	0.4918
M2	0.16	1.00	0.25	0.25	0.0626
M3	0.25	4.00	1.00	0.33	0.1527
M4	0.50	4.00	3.00	1.00	0.2929
sum	1.91	15.00	8.25	3.58	1.000

(Source: author).

Principal eigenvector: λ_max = 4.1874; consistency index, CI = 0.0625; random value of consistency index, RI = 0.9; and consistency ratio, CR = 0.0694. The consistency ratio is under 0.10, which means that the performed evaluation is consistent.

(B)

Calculating overall composite weights (synthesizing results) and final ranking

The overall composite weight of each variant (alternative)—potential domains of risks of pollution from ships—is based on the calculated weights of level 1 and level 2. It is just the normalization of the linear combination of multiplication between the weight and priority eigenvector (Priority Vector).

The overall composite weight for the analyzed variants (alternatives)—domains of risks, M_i—is calculated based on Equation (4):

M1 = (0.1753 × 0.5668) + (0.1230 × 0.5668) + (0.3274 × 0.5668) + (0.3760 × 0.4918) = 0.5365

M2 = (0.1753 × 0.0598) + (0.1230 × 0.0598) + (0.3274 × 0.0598) + (0.3760 × 0.0626) = 0.0623

M3 = (0.1753 × 0.1644) + (0.1230 × 0.1644) + (0.3274 × 0.1644) + (0.3760 × 0.1527) = 0.1602

M4 = (0.1753 × 0.2090) + (0.1230 × 0.2090) + (0.3274 × 0.2090) + (0.3760 × 0.2090) = 0.2408

The overall composite weight of the analyzed variants (alternatives) is presented with the matrix—Table 15. This table is formed of following elements: column “K1” is equal to the column “Priority Vector” in Table 11; column “K2” is equal to the column “Priority Vector” in Table 12; column “K3” is equal to the column “Priority Vector” in Table 13; and column “K4” is equal to the column “Priority Vector” in Table 14; in the column “Composite weight”, previously calculated values of this parameter are inserted.

Table 15. Values of composite weights.

	K1	K2	K3	K4	Composite Weight
M1	0.5668	0.5668	0.5668	0.4918	0.5365
M2	0.0598	0.0598	0.0598	0.0626	0.0623
M3	0.1644	0.1644	0.1644	0.1527	0.1602
M4	0.2090	0.2090	0.2090	0.2929	0.2408

(Source: author).

The overall consistency of the hierarchy, CR, calculated based on the related equation in Table 9, is 0.0995 (under 0.10), which means that the complete evaluation (at level 1 and level 2) is consistent.

5.3.3. Discussion of Results

Based on the results systematized in the previous sections, the analyzed variants (alternatives)—domains of risks of pollution from ships in the Port of Bar—can be ranked:

-

Rank 1: Variant (alternative) 1—domain of risks M1: unloading oil/oil derivatives from ships to tanks (and vice versa)—with a general composite weight share of 0.5365 (53.65%);

-

Rank 2: Variant (alternative) 4—domain of risks M4: liquid and solid waste from ships/emissions from ships—with a general composite weight share of 0.2408 (24.08%);

-

Rank 3: Variant (alternative) 3—domain of risks M3: unloading packaged dangerous goods from ships to warehouses (and vice versa)—with a general composite weight share of 0.1602 (16.02%);

-

Rank 4: Variant (alternative) 2—domain of risks M2: unloading harmful liquid substances in bulk (chemicals) from ships to tanks—with a general composite weight share of 0.0623 (6.23%).

The presented results (categories of domains of risks) indicate a strong dependence of the ranks on the following parameters:

-

Unloaded/loaded cargo volume, which directly determines the requirements regarding the necessary workforce and equipment for protection from pollution from ships, having at the same time an influence on costs;

-

Productivity during the unloading/loading operations, which directly determines the time of ships staying at berth—the time of exposure to risks of pollution from ships.

Connecting the previous statements with concrete data given in Section 5.2 of this paper, some correlations can be established:

-

Rank 1 of Variant (alternative) 1—domain of risks M1—corresponds to the volume of oil/oil derivates handled in the analyzed period, which was the biggest in comparison with the two other considered cargo types (packaged dangerous goods and harmful liquid substances in bulk (chemicals)), and the time of ships staying at berth for unloading/loading oil/oil derivates, which was the biggest in comparison with the two other cargo types (packaged dangerous goods and harmful liquid substances in bulk (chemicals)), which is followed with the biggest exposure to risks of pollution from ships.

In order to enable the highest possible adaptability of the (sub)system of prevention of pollution from ships, it is necessary to periodically reconsider the defined ranks depending on the variations in related parameters.

At the same time, the results shown, in a specific way, indicate the importance of all the requirements defined by the MARPOL Convention (including all its annexes) being fully met. In this context, it is necessary to point out the importance of all the elements of national (Montenegrin) regulation in the field of sea protection against pollution from ships being fully harmonized with international regulations, especially with the mentioned MARPOL Convention (and its annexes). Adequate implementation of MARPOL (as well as other related international widely adopted conventions and directives) in national legislation enables the unification of policy environments in the field of the prevention of pollution from ships, which is of crucial importance for the effectiveness and efficiency of the protection system. The management model of prevention of pollution from ships proposed by this paper considers that the related national legislation is fully in line with international legislation. It means that the implementation of the proposed management model in ports in countries where relevant conventions and directives are not successfully implemented would be limited.

The defined risks categories are, among others, established bases for the adequate planning of necessary preventive and corrective actions connected with pollution from ships. The previously mentioned directly bring to the fore the necessity to have defined (recognized) adequate recovery strategies. This can be very clearly illustrated if the contamination of sediments in the port water area—due to the pollution from ships—is taken into consideration. Concrete recovery strategies of dredged contaminated marine sediments in ports have been developed [134], and concrete remediation techniques have been proposed for the remediation of the contaminated dredged sediments [135].

Variant (alternative) 1—domain of risks M1: unloading oil/oil derivatives from ships to tanks (and vice versa)—has the highest rank according to all four selection criteria:

* According to criterion 1—K1: workforce—for the protection from pollution from ships (number, profile, and level of education of personnel, etc.), with a weight share of 0.5668 (56.68%);

* According to criterion 2—K2: equipment—for the protection from pollution from ships, with a weight share of 0.5668 (56.68%);

* According to criterion 3—K3: costs—related to the realization of activities for the purpose of protection from pollution from ships, with a weight share of 0.5668 (56.68%);

* According to criterion 4—K4: time of exposure to the risk of pollution from ships— with a weight share of 0.4918 (49.18%).

Variant (alternative) 4—domain of risks M4: liquid and solid waste from ships/emissions from ships—is ranked second according to all four selection criteria, with the following weight shares:

* According to criterion 1—K1: workforce—for the protection from pollution from ships (number, profile, and level of education of personnel, etc.), with a weight share of 0.2090 (20.90%);

* According to criterion 2—K2: equipment—for the protection from pollution from ships, with a weight share of 0.2090 (20.90%);

* According to criterion 3—K3: costs—related to the realization of activities for the purpose of protection from pollution from ships, with a weight share of 0.2090 (20.90%);

* According to criterion 4—K4: time of exposure to the risk of pollution from ships—with a weight share of 0.2929 (29.29%).

The ranks of Variant (alternative) 4—domain of risks M4—as per the selection criteria used, are mainly determined with the time of exposure to risks from pollution with liquid and solid waste from ships and emissions from ships, which, on the other hand, depends on productivity during the unloading/loading operations.

Variant (alternative) 3—domain of risks M3: unloading packaged dangerous goods from ships to warehouses—is the third in rank according to all four selection criteria, with the following weight shares:

* According to criterion 1—K1: workforce—for the protection from pollution from ships (number, profile, and level of education of personnel, etc.), with a weight share of 0.1644 (16.44%);

* According to criterion 2—K2: equipment—for the protection from pollution from ships, with a weight share of 0.1644 (16.44%);

* According to criterion 3—K3: costs—related to the realization of activities for the purpose of protection from pollution from ships, with a weight share of 0.1644 (16.44%);

* According to criterion 4—K4: time of exposure to the risk of pollution from ships—with a weight share of 0.1527 (15.27%).

Variant (alternative) 2—domain of risks M2: unloading harmful liquid substances in bulk (chemicals) from ships to tanks—is the fourth in rank according to all four selection criteria, with the following weight shares:

* According to criterion 1—K1: workforce—for the protection from pollution from ships (number, profile, and level of education of personnel, etc.), with a weight share of 0.0598 (5.98%);

* According to criterion 2—K2: equipment—for the protection from pollution from ships, with a weight share of 0.0598 (5.98%);

* According to criterion 3—K3: costs—related to the realization of activities for the purpose of protection from pollution from ships, with a weight share of 0.0598 (5.98%);

* According to criterion 4—K4: time of exposure to the risk of pollution from ships—with a weight share of 0.0626 (6.26%).

The ranks of variants (alternatives) 3 and 2—domains of risks M3 and M2—are in line with the values of related determining parameters: the volume of unloaded/loaded cargo and productivity during the unloading/loading operations—time of ships staying at berth—and time of exposure to risks.

6. Conclusions

The considerations made in this paper strongly confirm that the prevention of pollution from ships in a port definitely belongs to the priorities from all relevant managerial points of view, especially when an objective to fulfill demands determined by the “green port” concept is kept in mind. In this sense, it is of crucial importance to act systematically, in a modeled way, strictly in accordance with the development trends of environmental management systems in ports, taking into account all specific characteristics of a concrete port.

The range of factors with influence in the (sub)system of prevention of pollution from ships in a port is very wide. Among them, the ranks of domains of risks of pollution from ships deserve special attention, which motivated the author to propose an approach for their identification using the Analytic Hierarchy Process (AHP) method. Through the analyses conducted in this paper, some important correlations between the level of risks of pollution from ships and the quantity of unloaded/loaded cargo and productivity during the unloading/loading operations became obvious. The intention of the author is to investigate these correlations and present the results in a further paper.

Furthermore, respecting the fact that the categorization of risks of pollution from ships in a port, conducted in Section 5.3 of this paper, is, mainly, connected with the data from the concrete five-year period and fully respecting the changing nature of determining parameters, within the scope of his further engagement in this domain, the author plans to work on creating a software for calculating risk levels, based on the considerations made/the model proposed in this paper.

Respecting the fact that this paper proposes an approach to modeling a (sub)system of prevention of pollution from ships in a concrete port (analyzed through a case study) which is based on the general PDCA management model, recommends directions for the identification (and analysis) of factors which determine requirements related to the mentioned (sub)system of prevention of pollution, and defines a model for the categorization of the domains of risk of pollution from ships correlated with identified determining factors, the results of the considerations made could be implemented and/or replicated through research/analyses in other very similar ports (multipurpose ports), as well as in other ports, with necessary adjustments.