Method in Selecting Vehicles for Interventions and Surveillance of Navigation Safety at Sea

Abstract

Selecting a vehicle to monitor navigational safety is an important goal, especially in search and rescue operations. It depends on the criteria set and the type of equipment in use. The study aimed to select the optimal vehicle according to the optimal criterion. In the decision-making, the AHP method was used to analyze and rank the selection criteria and vehicle types. As the most important criterion, the results point to reliability in different weather conditions and the SAR vessel as the first choice for interventions and monitoring navigational safety in Croatia. In the selection process, the AHP methodology pointed directly to the significant inconsistency of the expert group and indirectly to the need for more careful selection of members, additional training, and a broader selection of criteria and equipment.

1. Introduction

The safety of navigation at sea has a significant part in the protection of life, environment, and economic interests. Surveillance and fast reaction in the event of emergencies at sea are of key importance. That is why it is useful to consider and apply methods that will facilitate decision-making to increase the level of navigation safety. This paper explores the application of the Analytic Hierarchy Process (AHP) as a decision-making tool in the context of surveillance of navigation safety in the Republic of Croatia. The AHP is a multi-criteria method that enables systematic and structured consideration of different quantitative and qualitative aspects in decision-making, providing valuable guidelines for the optimization of emergency procedures. It is used in different areas and has a wide scope of applications. This method is closest to the human intuitive analysis of complex issues, breaking them down into simpler patterns. It can be understood without a great need for a more detailed comprehension of mathematical models, enabling a relatively simple interpretation of results [1].

The AHP is often used and published. In 2006, ref. [2] identified 150 publications that were published in international scientific journals between 1983 and 2003 concerning this methodology. According to [3], planning and locating resources in the field of interventions and surveillance of sea navigation safety should be carried out based on scientific criteria. Multi-criteria analyses have already been used in the selection of search and rescue units at sea [4], and the AHP method was used in selection of a ship domain to reduce the safety risk of navigation [5]. Compared to the established criteria when making decisions in the aforementioned papers, as well as in [6], which are primarily technical, the criteria in this paper are principled and adjusted so that they are comprehensible to a wider group of stakeholders involved in the decision-making process. They are also the objects of analysis in this research, and the majority of them have a component relating to safety at sea.

Other methods can be used for the same purpose. TOPSIS may be a simpler decision-making method, but it depends on AHP or another method to determine the weighting of the criteria [7]. In the MACBETH method, which uses linear programming, there is a problem in the mathematical background in the calculation of qualitative criteria and quantitative criteria that do not have a recognized and accepted metric, a constraint known as the condition of order preservation (COP). It is considered that the analysis with this method is weaker and more linear than systemic, as in AHP [8]. The PROMETHEE method has the advantage of simplicity and can process more complex criteria. However, this can make it difficult for some experts, as the decomposition of complex problems into simpler ones, as in AHP, is closer to the human mind. In addition, the weighting in the method is based on the individual, whereas this function is methodically arranged in AHP [9]. The TODIM method has an advantage over AHP because it is simpler, does not process the variables in pairs, and is not limited by the number of criteria. It is considered that the number of criteria in AHP should not exceed nine in order not to jeopardize the judgment procedure and consistency [10].

In this paper, the AHP method is used to explore the optimal criteria for the selection of the means for intervention and the surveillance of navigation at sea in the Republic of Croatia, as well as the most appropriate means based on the established criteria. By examining and comparing the defined criteria and types of equipment, the aim is to prioritize among the offered alternatives to support decision-making in administrative bodies for maritime transport and navigation safety at sea. The value of the AHP method in navigation safety tasks will be tested. The selection of equipment for navigation safety in the Adriatic Sea using the AHP method has never been researched or published. The same has never been investigated based on an established set of criteria.

2. Actuality of Surveillance of Navigation Safety

Considering the vessels and aircraft that operate in the service of monitoring the safety of navigation in the Republic of Croatia, as well as trends and technological development, it is clear that the equipment significantly affects the overall system of monitoring the safety of navigation. Technological support improves the ability to react quickly and precisely locate vessels and persons in distress, coordinates search and rescue operations, reduces the risk of accidents, and makes the maritime space safer for all participants [11]. Optimal allocation of assets is a standard procedure in the maritime safety monitoring system [12], including optimization in the selection of required equipment. For example, helicopters are expected to play an increasingly important role in search and rescue operations at sea, which requires adequate scientific support [13]. Apart from their humanitarian nature, search and rescue operations can affect the well-being of the local community and even the economy of the region or country [14]. Yearly, about 600 maritime accidents occur on the high seas, of which about 250 ships sink [15]. Navigation safety monitoring began to develop rapidly after the Second World War, and the development of technology significantly increased the efficiency of search and rescue operations. Failures in the form of slow response or mysterious disappearance of a ship or aircraft at sea, as well as the increasingly dense maritime traffic, make this issue still relevant [16]. The system for monitoring navigation safety in the Mediterranean has recently played a significant role in the migrant crisis, further updating the issue [17]. International cooperation is very important for successfully monitoring the safety of navigation and search and rescue operations, although conflicts sometimes arise, especially on issues of responsibility and obligations [18]. The development of an operational search and rescue plan is a complex, multi-criteria problem. For this purpose, numerous complicated, multi-phase algorithms are created and proposed [19,20]. It is not entirely clear how effective such algorithms are in practice and how much decisions differ from those made by simple, classical methods such as AHP. Such studies in maritime traffic surveillance and search and rescue at sea, especially in crisis situations, aim to reduce the costs of crisis management, improve the coordination of SAR operations, improve the efficiency of operations, and reduce the number of deaths [21]. The term safety of navigation is used in this paper in the same context as the expressions safety at sea and maritime safety, and those terms are not differentiated or excluded. They are usually combined with the terms surveillance, search and rescue, intervention, operation, maritime traffic management, etc. [22]. Due to worsening meteorological conditions of navigation caused by climate changes and also due to political tensions and cross-border incidents, increased demand for these activities and services is expected, especially in the domain of cross-border cooperation [23]. According to [24], cross-border cooperation and joint operations improve search and rescue (SAR) services.

In this paper, the AHP method is used to explore the optimal criteria for the selection of the means for intervention and the surveillance of navigation at sea in the Republic of Croatia, as well as the most appropriate means based on the established criteria. By examining and comparing the defined criteria and types of equipment, the aim is to prioritize the offered alternatives to support decision-making in administrative bodies for maritime transport and navigation safety at sea. The value of the AHP method in navigation safety tasks will be tested. The selection of equipment for navigation safety in the Adriatic Sea using the AHP method has never been researched or published. The same has never been investigated based on an established set of criteria.

3. Methodology

3.1. Theoretical Foundations of the AHP

According to Saaty, the inconsistency factor makes a difference between the AHP and other multi-criteria methods [25]. When item A is more favorable than item B and item B is more favorable than item C, it is concluded that item A is more favorable than item C. Such consistency of deductive logic is not necessary for the AHP method, which compares alternatives and determines priorities based on the set criteria. It enables the simplification of a complex problem in a multilevel hierarchical model, as presented in Figure 1 [26].

In a hierarchical model in which paired variables are compared each with each other, the decision maker actively participates, independently or with the help of an expert, influencing the values of the weighting factors and expressing his preferences using Saaty’s scale for determining relative importance. Based on knowledge, experience, and evidence, it is decided which of the paired criteria has a greater importance compared to the alternative and to what extent (

Table 1. Saaty’s scale of relative importance.

Importance Intensity	Descriptive Degree of Importance	Explanation
1	Equal importance	Two selected elements equally contribute to the aim.
3	Moderate importance	Based on the experience and assessments, one element has a moderate advantage compared to the other.
5	Strong importance	Based on the experience and assessments, one of two elements is strongly favored.
7	Very strong importance	Very strong favoring of one element is demonstrated in practice.
9	Extreme importance	The evidence favoring one element over another has been confirmed with the greatest certainty.
2, 4, 6, 8	Intermediate value	Interrelationship between previous and next values.

) [1].

According to [27], there are certain axioms on which the AHP method is based, providing a better understanding of the methodology. Those axioms represent well-established statements that do not need to be proven further, and those are the Axiom of dependence—comparison of elements of one level concerning an element of a higher level is allowed, that is, comparison of a lower level element depends on an element of a higher level; and Axiom of homogeneity—comparison of elements makes sense only if the elements are comparable. It ensures that comparisons between elements are consistent and allows for avoiding illogical and contradictory comparisons, which can lead to unreliable or unrealistic results of the AHP method; Axiom of expectations—every time there is a change in the hierarchy, there is a need to recalculate the priorities in the new hierarchy; Axiom of reciprocity—comparing element A with element B, where element A is x times more important than element B, then element B is 1/× times more important than element A, i.e., if element A is twice as important as element B, then, when comparing elements A and B, element B is one half more important than element A.

3.2. Mathematical Model of the AHP

The analytical hierarchy process uses a mathematical model for ranking and decision-making in multi-criteria situations [28]. In a hierarchical model, the set alternatives are A1, A2, …, An, ranked for the set criteria C1, C2, …, Cn, while the zero level is the set aim. The criteria and alternatives are compared based on different F_i factors, and weights that are w_i or w_j scores, assigned to each criterion and alternative, represent the decision maker’s subjective estimate. Given the obtained results, a matrix F = [f_ij] is formed in order to compare criteria and alternatives based on weights (scores). Each element of the matrix, F = [f_ij], is calculated using the formula [29]:

$${\mathrm{f}}_{\mathrm{ij}}=\frac{{\mathrm{w}}_{\mathrm{i}}}{{\mathrm{w}}_{\mathrm{j}}}, \mathrm{i},\mathrm{j} = 1, 2, \dots \mathrm{k}$$

(1)

where w_i is the weight or score of criterion C_i and w_j is the weight or score of the corresponding criterion C_j. This type of matrix is a positive reciprocal matrix, which must satisfy reciprocal values, as presented in the formula below [28]:

$${\mathrm{f}}_{\mathrm{ij}}=\frac{1}{{\mathrm{f}}_{\mathrm{ji}}}, {\mathrm{f}}_{\mathrm{ji}}>0, {\mathrm{f}}_{\mathrm{ii}}=1, \mathrm{i},\mathrm{j}=1, 2, \dots , \mathrm{k}$$

(2)

whereby element f_ij must be inversely proportional to f_ji, which means that f_ji must be larger than 0. The resulting matrix has the property of consistency when the following conditions are met [29]:

$${\mathrm{f}}_{\mathrm{ij}}={\mathrm{f}}_{\mathrm{ip}}{\mathrm{f}}_{\mathrm{pj}}, \mathrm{i},\mathrm{j}=1, 2, \dots \mathrm{k}; \mathrm{p}=1, 2, \dots \mathrm{k}$$

(3)

To confirm that the F matrix is consistent, its eigenvalue for a certain eigenproblem must have the value:

$$\mathrm{Fw}=\mathrm{kw}$$

(4)

where it is concluded that:

$${\mathsf{\lambda }}_{\max }=\mathrm{k}$$

(5)

whereby is λ_max the highest eigenvalue of the matrix F equal to the number of criteria and/or alternatives [29]. The next step includes the normalization of the weight vector w by dividing the corresponding element w with the total sum of weight factors according to the formula:

$${\overline{\mathrm{w}}}_{\mathrm{i}}=\frac{{\mathrm{w}}_{\mathrm{i}}}{{\mathrm{w}}_1+{\mathrm{w}}_2+\dots .{\mathrm{w}}_{\mathrm{k}}},\mathrm{i}=1,2,\dots ,\mathrm{k}$$

(6)

Then, consistency index (CI) is calculated according to the following formula [1]:

$$\mathrm{CI}=\frac{\mathsf{\lambda }\max -\mathrm{n}}{\mathrm{n}-1}$$

(7)

whereby λ_max is the largest eigenvalue of the matrix F, while n represents the number of elements (alternatives or criteria). Consistency Ratio as the final step is defined as:

$$\mathrm{CR}=\frac{\mathrm{CI}}{\mathrm{RI}}$$

(8)

whereby the previously obtained CI is divided by random index (RI). Random index RI is applied only if n ≥ 3, which can be seen in

Table 2. Random index (RI).

n	RI
1	0
2	0
3	0.52
4	0.89
5	1.11
6	1.25
7	1.35
8	1.4
9	1.45
10	1.49

[1]. When CR is less or equal to 0.10, comparison is considered acceptable. If CR is different from the stated number, the comparisons must be repeated to resolve the inconsistency of estimates [27].

According to [30], decision-makers who work with the AHP method have the advantage that this method has the feature of recognizing the inconsistency of decision-makers when comparing the elements of the AHP structure.

3.3. Alternatives and Criteria for Selecting the Optimal Equipment for Navigation Safety Surveillance in the Republic of Croatia

Equipment for monitoring navigation safety in the Republic of Croatia included in the research is the motor vessel M/V Pojišan, the helicopter Mil Mi-8 MTV-1, the drone Schiebel Camcopter S-100, and the plane Canadair CL-415.

The criteria that influence the decision-making on the choice of optimal equipment in navigation safety surveillance operations in this research are readiness, response time, reliability in different weather conditions, working range, directness of intervention, cost, maintenance, and impact on the environment. The process of defining individual criteria, i.e., the level of their influence (importance) concerning the investigated alternatives (equipment) in navigation safety surveillance operations, is the result of recent findings and consultations with stakeholders directly involved in the subject area. All the mentioned criteria for fulfilling the axiom of homogeneity are mutually comparable, and the comparisons between the elements are consistent.

To select the optimal equipment for undertaking interventions within navigation safety and data processing using the AHP method,

Table 3. Importance of each defined criterion in the selection of individual vehicles for the safety navigation control in Croatia.

Vehicle	Manufacturer	Type of Vehicle	Readiness	Response Time	Reliability in Diverse Weather Conditions	Operational Range	Direct Intervention	Cost	Maintenance	Environmental Impact
SAR vessel	Brodosplit shipyard (Split, Croatia)	M/V Pojišan	moderate	great	moderate	great	great	moderate	moderate	moderate
Helicopter	Mil Moscow Helicopter Plant (Russia)	Mil Mi-8 MTV-1	great	great	moderate	great	moderate	great	great	great
Drone	Schiebel (Vienna, Austria)	Schiebel Camcopter S-100	great	moderate	minor	moderate	moderate	minor	moderate	minor
Plane	Viking air (Sidney, British Columbia, Canada)	Canadair CL-415	great	great	moderate	great	moderate	great	great	great

shows the relevant criteria with the associated valorization of the level of influence of individual criteria.

All the defined criteria represent a certain set of standards, that is, conditions according to which the quality of service of each individual type of navigation safety surveillance equipment used in the Republic of Croatia is valued. The defined criteria serve as reference points or benchmarks for making decisions or scores. These guidelines were adopted in cooperation with an expert group.

The criterion of readiness refers to the degree of preparedness of the considered equipment for activities of navigation safety surveillance, the assessment of the availability of the required resources (primarily human and technical), skills, and knowledge, as well as the evaluation of planning and organization. For example, the criterion of readiness for the helicopter Mil MI-8 MTV 1 covers numerous prerequisites that must be examined and guaranteed in order for the helicopter to be ready for intervention. Criteria fulfillment contributes to crew and aircraft safety. The key readiness criteria for a helicopter are technical soundness, trained crew that includes the pilot and other cabin crew, weather conditions that must be acceptable in accordance with the operational limitations of the helicopter, etc. Given the 24-h duty of the team trained and qualified for working with a helicopter and its very quick response to a required action, the criterion of readiness for the helicopter Mil MI-8 MTV 1 is assessed as a high-importance criterion. The same arguments can be applied to planes, that is, drone preparedness. On the other hand, exclusively due to the dislocation of SAR ship crew, i.e., a lower level of vigilance and the absence of an immediate reaction in activating the intervention, the level of importance of the SAR ship’s readiness criterion is assessed as moderate.

The criterion of response time refers to the speed of a given means in the realization of required operations at sea within navigation safety surveillance. The relevant criterion refers to the time interval from a specific event or situation until the response. Due to certain endogenous, e.g., human factors, and exogenous reasons, e.g., meteorological conditions, which significantly affect the response time of the drone, the importance of this criterion for the drone is set at a moderate level.

The criterion of reliability in different weather conditions refers to the ability of the equipment in consistent and efficient operations in diverse weather conditions. This criterion includes resilience to extreme temperatures, moisture, snow and ice, storms and strong wind, reduced visibility, and corrosion. A moderate level of importance of the mentioned criterion was assigned for the use of almost all types of tested equipment, which achieve significant operational performance in adverse weather conditions but are nevertheless limited by strong winds and reduced direct visibility in operations at sea. It excludes the use of drones, so the reliability degree was assessed as low, primarily due to technical, technological, working, and operative limitations. Proper selection of equipment, taking into account weather conditions, can determine the final success of the search and rescue operation and minimize any type of risk.

The criterion of the operational range of the navigation safety surveillance equipment, especially the search and rescue operation, refers to the distance or area that the stated equipment can cover with its efficient operation. This criterion’s accuracy and range depend on the specificity of the equipment that is being assessed. When it comes to drones, operational range means the maximum distance at which the drone can operate while fulfilling all of its functions without interruptions; therefore, when compared to the rest of the equipment, the importance level was marked as moderate. It also includes the maximum permitted distance between a drone and the operator, that is, the control station. Helicopters and planes are capable of considerable operational distance; however, their flight autonomy is often less than that of drones. Regardless of the above, and taking into account the coverage of a larger surveillance area and the speed of intervention, the criterion of importance for the highlighted equipment was evaluated as high. Ships have a large operational range at sea, and in this case, M/B Pojišan achieves navigation autonomy of 250 NM [31].

The criterion of direct intervention refers to a situation in which a decision is being made or a measure is taken directly and without mediation in order to resolve a specific situation or problem. This kind of intervention requires a quick and immediate reaction to achieve the desired result, in this case, the quickest and most efficient results possible regarding operations of navigation safety surveillance. Taking into account the above assumptions, the level of importance of the considered criterion was determined as high when using the ship in interventions at sea or moderate for the use of other equipment, i.e., drones, planes, and helicopters. The use of the ship enables direct intervention in emergencies, inspection, and other operations within navigation safety surveillance.

The criterion of cost represents an important factor used to evaluate the cost-effectiveness, competitiveness, or accessibility of equipment for various types of actions in monitoring the safety of navigation. This criterion covers all financial elements and resources needed for the realization of planned or sudden interventions. Cost can be expressed in different units such as money, working hours, or through other relevant measures. Cost analysis enables decision-makers to better understand the financial implications of the alternatives and direct resources towards the best options or strategies. The cost is often, even in this case, compared to other criteria in order to make relevant decisions. Considering the nature and complexity of the individual tested equipment, i.e., their operational implications, the level of importance of the cost for the use of drones is low, moderate when choosing a ship, and high for navigation safety monitoring operations using aircraft and helicopters.

The maintenance criterion refers to procedures and resources necessary for regular service and support for correct operation, efficiency, and durability of the equipment. This criterion has a key role in assessing long-term reliability and cost-effectiveness of the equipment because maintenance can have a significant effect on total operational costs and functionality. Drones usually require less maintenance compared to other options, as do ships, while the cost of maintaining aircraft (planes and helicopters) is determined to be high.

The criterion of environmental impact is extremely important due to maintaining a balance between operational needs and nature and environmental protection. This criterion assesses how the use of the equipment affects land, air, and water, as well as the whole ecosystem. When analyzing the aforementioned aircraft in the service of navigation safety surveillance, the most common and important is the assessment of harmful gas emissions and their potential impact on air quality and the environment. In drones, this aspect refers only to drones with internal combustion engines. An important factor is noise and its impact on the environment and society as a whole, especially with planes and helicopters. The impact of wastewater and potentially harmful liquid substances discharged by ships into the sea, as well as the impact of anti-fouling coatings and emissions, are assessed for vessels engaged in interventions at sea.

The choice of the optimum is complicated because it depends not only on the chosen criterion but also on the scenario investigated, the research design, and the weighting of the criteria [32]. The AHP hierarchy also depends on the wishes and needs of those involved in the decision-making process and their knowledge and judgments [33]. The criteria differ depending on the priorities of the decision-makers. For the weighting of the technical criteria, the so-called subjective ranking method is used, which depends mainly on the requirements of the decision-makers. Economic and ecological criteria are weighted according to the equal weighting method, according to which the influence of the decision-makers on the criteria is minimal [34]. In general, in the AHP method, decision-making depends on the subjective assessment of experts [35], and a successful analysis results from the proper development of the criteria hierarchy and the comparison of pairs of criteria, sub-criteria, alternatives, and priorities [36]. Therefore, the different priorities within the criteria, the selection of the criteria included, and their different weighting depend largely on the purpose of the decision and the specific conditions in each individual case [37]. The decision-making process is never completely devoid of subjectivity, regardless of the method used. However, within the AHP method, some indicators indicate the degree of subjectivity in research. Decision-making cannot be general but depends on the specific problem [28], and this requires an in-depth understanding of the actual factors [38]. Good decisions also depend on conditions in the future, so it is good to include a time component in the decision-making process [39]. The choice of the optimal decision depends on many parameters and criteria that have relative importance [40].

To select the optimal equipment for the needs of monitoring the safety of navigation in the Adriatic Sea, the methodological conditions for weighting the criteria are applied in accordance with the requirements and priorities of the specific environment. With an area of 31,479 km² and numerous islands that reduce the sea distance between the coast and the state border, the coastal sea of the Republic of Croatia does not represent a limitation in the categories of operational range and response time. It is a closed sea where no extreme meteorological conditions are to be expected, but for the same reason, there is an increased ecological sensitivity. With a GDP of 73% of the European Union average in 2022 [41], the Republic of Croatia is still materially limited in the procurement and maintenance of maritime safety surveillance equipment, but even under these conditions, full operational readiness and directness in maritime safety surveillance and search and rescue operations are required. The Republic of Croatia is traditionally a maritime country with extensive maritime experience, and maritime safety surveillance has so far been carried out mainly by SAR vessels.

Based on all the assumptions defined in the methodological section, which primarily arise from the need for the AHP method and the selected research problem, a flowchart of the subject research was created. The elements of this hierarchical structure can be divided into objectives, defined criteria, and examined alternatives (Figure 2).

The process of data gathering to carry out this research was based on the development and dissemination of the questionnaire. The survey questionnaire consisted of two parts: comparison and valorization of pairs of basic criteria to determine their importance and mutual comparison of pairs of alternatives according to each of the criteria. Based on the assessment of the relative importance of the criteria according to the corresponding level of the hierarchical structure of the problem, the local weights of the criteria and the priorities of the alternatives are calculated. The calculation of the total priorities of alternatives is based on the weighting of local priorities with the weights of all nodes and their total sum [42].

The questionnaire was forwarded to a previously defined expert group, which included stakeholders who are directly and indirectly involved in different aspects of navigation safety surveillance. The expert group is composed of representatives of the Ministry of the Sea, Transport and Infrastructure (Navigation Safety Administration), the National Center for Coordination of Search and Rescue (MRCC), the National Center for Monitoring and Management of Maritime Traffic (VTS Croatia), port authorities, harbor masters, representatives of the academic community, and other stakeholders. The questionnaire was completed by 15 out of a total of 35 stakeholders, accounting for 43% of realization success. Therefore, the sample can be characterized as representative. It is important to point out that in the process of synthesizing the judgments of different stakeholders, during the valorization of criteria and alternatives, the arithmetic mean technique was applied, especially in the phase of synthesizing the comparison of pairs, and according to the propositions established in [43]. For the needs of carrying out the empirical part of the research, the AHP Excel tool was used to calculate the total weight of the criteria and the total priorities of alternatives according to [44].

4. Results

Using Saaty’s scale of relative importance (Table 1), after determining the aim and hierarchical setting of the research problem, the survey group evaluates the importance ratios among the defined criteria. It is followed by the calculation of the weight of each criterion from their estimated ratios using the approximate procedure for calculating the maximum eigenvalues and eigenvectors. In addition, the method includes the calculation of the consistency index, consistency ratio, and λ_max (the largest eigenvalue of the matrix).

Table 4. Calculation of the relative weight of each criterion and the overall consistency ratio (CR), consistency index (CI), and eigenvalue of the matrix (λ_max).

	Weights/Ponders (w)	Weighted Sum Vector	λmax	CI	CR
Readiness	0.139	1.130	8.104	0.014	0.011
Response time	0.131	1.060
Reliability—in diverse weather conditions	0.229	1.855
Operational range	0.168	1.358
Direct intervention	0.189	1.533
Cost	0.036	0.289
Maintenance	0.062	0.499
Environmental impact	0.047	0.380

shows the results of the AHP.

The data show that the relative weight, or the evaluation factor, is the highest for the criterion reliability in different weather conditions and the lowest for the criterion cost. The weights of individual criteria are shown in Figure 3.

The next level of hierarchy referred to the comparison of alternative pairs by each defined criterion.

Table 5. Relative weights of alternatives according to individual criteria.

Weighted Sum Vector
	Readiness	Response Time	Reliability—in Diverse Weather Conditions	Operational Range	Direct Intervention	Cost	Maintenance	Environmental Impact
SAR vessel	0.404	0.271	0.370	0.199	0.454	0.245	0.212	0.258
Helicopter	0.275	0.273	0.327	0.319	0.268	0.186	0.219	0.220
Drone	0.195	0.293	0.106	0.109	0.082	0.388	0.362	0.296
Plane	0.126	0.163	0.197	0.373	0.196	0.181	0.207	0.226
λmax	4.108	4.041	4.029	4.024	4.026	4.117	4.045	4.003
CI	0.036	0.014	0.010	0.008	0.009	0.039	0.015	0.001
CR	0.040	0.016	0.011	0.009	0.010	0.044	0.017	0.001

shows alternative priorities based on individual criteria given the evaluation factor (relative weight) and calculated values CI, CR, and λ_max.

Highlighted values (yellow) indicate the estimated ratios of alternative priorities given the individual criterion. Based on the above results, the highest rank according to the readiness, reliability in different weather conditions, and directness of intervention criteria was assigned to the SAR vessel. According to the criteria of response time, cost, maintenance, and environmental impact, priority was assigned to the drone. The use of planes in navigation safety surveillance is the first choice based on the operational range criterion.

After the calculation of weights of the criteria and priorities of alternatives for each criterion, the overall priorities of alternatives are calculated by weighting the local priorities with the weights of criteria.

Table 6. Local and overall priorities of alternatives.

Alternatives	Criteria and Their Weights								Overall Priorities of the Alternatives
	1	2	3	4	5	6	7	8
	0.139	0.131	0.229	0.168	0.189	0.036	0.062	0.047
SAR vessel	0.404	0.271	0.370	0.199	0.454	0.245	0.212	0.258	0.330
Helicopter	0.275	0.273	0.327	0.319	0.268	0.186	0.219	0.220	0.284
Drone	0.195	0.293	0.106	0.109	0.082	0.388	0.362	0.296	0.173
Plane	0.126	0.163	0.197	0.373	0.196	0.181	0.207	0.226	0.213

1—Readiness; 2—Time response; 3—Reliability-in diverse weather conditions; 4—Operational range; 5—Direct intervention; 6—Cost; 7—Maintenance; 8—Environmental impact.

shows the total alternative priorities.

The graphic representation of the results of the overall alternative priorities (Figure 4) can be used to rank the selection of dedicated equipment (means) for navigation safety surveillance interventions in Croatia. By carrying out the AHP method, the use of SAR vessels is marked as a priority for the stated purpose, while the lowest priority among the offered alternatives was assigned to the use of a drone.

5. Discussion

The AHP method can be used for various purposes, usually in decision-making for some future goals. In this paper, it helps to make decisions on the equipment selection for navigation safety. There are various digital programs for applying the AHP method. We used Excel, but others could also be utilized. The individual criteria used in this work within the AHP method are neither new nor unknown. The set of them and the influence level of each criterion are new and original. The design of each new research using the AHP method is different depending on the specifics of the task.

The research results suggest reliability in different weather conditions as the most important criterion for the selection of means for interventions and sea safety surveillance in the Republic of Croatia. The cost was assigned the lowest weight by the expert group that filled in the survey. Consistency ratio CR 0.011 < 0.1 fulfils the condition in order for the criteria analysis to be acceptable. A low consistency index (CI = 0.014) provides information regarding non-compliance with compared weights, and it can be interpreted with a non-homogenous group of stakeholders who took the survey, which significantly differs in terms of position, knowledge, and experience. This item has a strong effect on the value of research results as it reveals a high degree of incompetence of the surveyed group in deciding the task. The same item reveals and confirms the value of the AHP method in precisely locating the weakness in the process of decision-making, which indirectly implies the need for a more careful selection of expert group members, as well as additional and continuous education. It also highlighted the importance of correct interpretation of the results. In the model presented, three of the eight criteria (the first, the third, and the fifth) contribute 55.7% to the weight of the decision, while the eighth criterion accounts for less than 5% (4.7%). This imbalance weakens the model, but we have accepted the expert group’s decision as relevant.

The largest eigenvalue of the matrix λ_max would have to be approximately equal to the number of criteria and/or alternatives, and in this case, it is 8.104, which corresponds to the total number of the set criteria (eight).

Even though drones, as an alternative, are the best selection based on four set criteria, response time, cost, maintenance, and environmental impact, these criteria have the lowest weight score of importance ratio among the set criteria. Therefore, the importance of drones in final interpretation of the results is small, or at least less important, when compared to other alternatives. The analysis of individual criteria such as preparedness, reliability in different weather conditions, and directness of intervention clearly highlights SAR vessels as the priority alternative. The mentioned criteria have the highest weight score; given the evaluation factor, the SAR vessel has the highest value based on those criteria. The analysis of the operational range criterion showed that the plane has a prominent advantage, followed by the helicopter. This suggests a significant role of these aircraft in surveillance and quick intervention in a wider area. Despite these advantages, those means have limited flight autonomy. The technical capability of fuel supply from ships would significantly increase the operational range of the helicopter and its part in sea safety surveillance [45]. Drones have higher flight autonomy but also a smaller operational range and role in navigation safety surveillance, especially in search and rescue after locating the incident at sea in the Republic of Croatia. The results are in line with the analysis of search and rescue operations in the Republic of Croatia, which were carried out between 2018 and 2022 exclusively using SAR vessels even though other alternatives were available in the same period (Figure 5).

The vessels used in the marine SAR system correspond to the types of vessels in other maritime countries and include primary, secondary, and multi-purpose SAR vessels [47,48]. Navigation safety surveillance units include all maritime, air, and land units that carry out navigation safety surveillance managed by the National Centre for Search and Rescue or its subsidiary [49]. Despite the wide range of equipment that could be considered for evaluation, e.g., autonomous marine vehicles [50], the exclusive use of SAR vessels in search and rescue at sea in the previous period in the Republic of Croatia affected the number of considered alternatives, that is, only basic units used for the implementation of the navigation safety monitoring function are included. The stakeholders’ attitudes have not changed since 2018. According to [51], the efficiency of interventions and sea safety surveillance is better when there is a combined air-maritime operation than when only vessels or aircraft are used. There is a need for a broader education regarding new technological achievements, especially in the development of drones [52] and vessels, to include a comprehensive spectrum of available equipment used in navigation safety surveillance in research and to face new trends and experiences with traditional attitudes and demands. Furthermore, when selecting the criteria, it is important to include a broader expert group that would contribute to the selection of better-quality parameters based on individual demands in sea safety surveillance operations. Identification of the key criteria can be carried out through discussion, previous valuation, and filtering, as well as reaching a consensus regarding the selected options. As a result, those criteria would contribute to higher precision and reliability of the research and possibly to the change of attitudes. Due to the safety risks of using drones, it is important to reconsider the selection criteria [53]. Communication with drones in adverse weather conditions is one of the key criteria and challenges of the future development and use of those vehicles in navigation safety surveillance [54]. Technical criteria should always be a part of principled criteria dominated by safety, without which, according to the IMO principle ‘safety first’, there is no navigation [55]. Even though digital platforms, based on multi-criteria analyses and machine learning [56], with the support of radar and satellite systems, are today involved in the affairs of navigation safety and in making optimal decisions in given circumstances, the key decisions are ultimately made by humans. When making a decision using the AHP, it is necessary to understand the technological options and be wise when carrying out weight scoring of criteria and alternatives, as well as when interpreting results.

6. Conclusions

By validating the results of the subject research in the model for decision-making by multi-criteria analysis based on the AHP, it was determined that experts and employees in the field of navigation safety at sea chose reliability in different weather conditions as the optimal criterion for selecting means for interventions and supervision of navigation safety in the Republic of Croatia. This process of comparing criteria was important in establishing a ranking of relative importance. The valuation of alternatives using the AHP method in the selection of funds for interventions and monitoring of navigation safety in the Republic of Croatia resulted in the selection of SAR vessels. The high degree of inconsistency in the results of the AHP analysis indicates the need for a more careful selection of the members of the expert group, as well as for further and continuous education, continuing with technological development and contemporary trends. It is easy to overlook it, while decisions derived from the same method stand out. Locating the weak point in decision-making confirmed the value of the AHP, the contribution of the method to the conclusions, and the importance of correct interpretation of the results. The relevance, quality, and applicability of the research would be increased by including a larger expert group of interested stakeholders by choosing better quality criteria and comprehensive equipment for interventions and supervision of navigation safety. These proposals and highlighted trends towards the combined technologies used with cross-border cooperation could help improve navigation safety surveillance in local conditions.