Strategic Assessment of Port Logistics Information Systems Using Fuzzy FUCOM–Fuzzy RAWEC: A Case from the Black Sea Region

Efficient port operation is a cornerstone of global and regional trade, influencing supply chain performance, trade facilitation, and economic competitiveness (Adabere, 2021; Wang et al., 2022). In this context, ports function not merely as points of cargo transfer but complex logistical ecosystems. Their performance impacts transit times, reliability, and the overall cost structure of international logistics chains. The Black Sea region, which includes Türkiye, Romania, Bulgaria, Ukraine, Russia, and Georgia, is of particular strategic importance due to its geographic location linking Europe and Asia through Eurasian corridors. It serves as a maritime gateway for energy exports, containerized trade, and bulk cargo movements (Popović et al., 2022; Zhelev, 2025). Despite its geopolitical and economic significance, systematic and comparative studies on port efficiency in the Black Sea context remain relatively limited, thus emphasizing a need for more targeted research in the region (Sunitiyoso et al., 2022).

Efficient port performance could critically facilitate the deployment of Logistics Information Systems (LIS), the digital platforms that integrate, process, and distribute logistics-related information to stakeholders across the supply chain. In port settings, LIS support real-time coordination of cargo flows, vessel scheduling, customs processing, inventory tracking, and intermodal logistics, thereby reducing bottlenecks and enhancing operational throughput. Tools such as Port Community Systems (PCS) and Terminal Operating Systems (TOS) have been widely implemented to streamline communication and automate core port processes (World Bank, 2023; Sahraoui et al., 2023). As ports in the Black Sea region endeavor to modernize, LIS represent a key strategic tool to bridge the performance gap with leading global ports, enhance integration with transnational corridors, and meet the increasing demands of digital trade and environmental compliance.

The emergence of the “smart port” paradigm characterized by Internet of Things (IoT) infrastructure, data analytics, automation, and AI-supported decision-making, further illustrates how LIS could propel not only operational efficiency but also sustainability, resilience, and digital transformation in port ecosystems (Sim et al., 2025; Min et al., 2022). These systems are essential to boost the competitiveness of modern ports as they provide visibility across the supply chain, enable predictive maintenance of port equipment, optimize berth allocation, and facilitate environmental monitoring.

Port operation and the implementation of LIS involve numerous, often conflicting criteria ranging from cost, reliability, throughput, interoperability, to sustainability. As a result, Multi-Criteria Decision-Making (MCDM) approaches become indispensable for structured evaluation across multiple dimensions under uncertainty (Wang et al., 2022). Fuzzy-based MCDM methods, which accommodate vagueness in expert judgment, allow more nuanced decision-making.

Along with the emergence of Full Consistency Method (FUCOM) and Rough Analytical Weighted Evaluation Criteria (RAWEC) in the literature of decision science, their direct application in port LIS has yet to be documented in peer review. However, frameworks such as grey FUCOM-strengths, weaknesses, opportunities, and threats (SWOT) have demonstrated the capability to handle uncertainty in logistics decision contexts (Popović, 2022). In a similar vein, hybrid fuzzy MCDM approaches have been successfully applied in the evaluation of container port service performance (Wang et al., 2022; Yılmaz & Kabak, 2020). These precedents suggested that fuzzy FUCOM and fuzzy RAWEC could provide rigorous and uncertainty-aware methodologies for port-related evaluations.

Implementing LIS across Black Sea ports faces several challenges like fragmented digital infrastructure, organizational silos, variable governance frameworks, and limited interoperability among stakeholders. Historical studies of port logistics, such as Port of Santos in Brazil, have identified deficiencies in information flows as a critical barrier to efficiency (Vieira et al., 2015; Pamucar & Görçün, 2022; Oconnor & Vega, 2019), a finding likely to be mirrored in similar regional contexts.

Conversely, opportunities for advancement exist. The global momentum toward smart port infrastructure emphasizes digital standardization, real-time data sharing, and coordinated decision-making (Yılmaz & Kabak, 2020; Said et al., 2014; Ferreira et al., 2025). Regional cooperation frameworks, including EU-supported trade corridors and pan-Black Sea initiatives, offer further institutional pathways for harmonizing digital logistics systems. Moreover, the imperatives of sustainability provide additional incentives; smart and green port models exemplify that digitalization can simultaneously promote environmental objectives and operational effectiveness (Su et al., 2024; Alzate et al., 2024; Elhussieny, 2025).

Despite the recognized importance of LIS for port efficiency and the potential of fuzzy MCDM approaches, there remains a research gap. There is no extant study that has systematically applied fuzzy FUCOM and fuzzy RAWEC to evaluate the implementation of LIS across Black Sea ports. Existing literature often relied on conventional MCDM models, or qualitative assessments, without incorporating the advanced uncertainty management offered by fuzzy methods.

The present study addressed this gap by:

1. Identifying key performance criteria for the effectiveness of LIS in port operation through expert elicitation and literature synthesis;

2. Applying fuzzy FUCOM to determine consistent criteria weights under uncertainty;

3. Utilizing fuzzy RAWEC to rank alternative LIS configurations or levels of implementation across port environments in the Black Sea region; and

4. Deriving implementable insights for port authorities and policymakers to enhance LIS-driven operational efficiency, competitiveness, and regional interoperability.

The novelty of this research lied in the methodological integration of two fuzzy MCDM frameworks, which were srategically applied in a vital and underexamined maritime region. The findings will contribute both to scholarly discourse on decision analysis in port logistics and to practical strategies for the development of regional digital infrastructure.

Multi‑criteria decision-making (MCDM) techniques have been widely applied to evaluate complex logistics and port-related decisions, especially under uncertainty.

· Fuzzy AHP and Quality of Port Services
Nguyen et  al. (2022) employed an integrated Fuzzy Analytic Hierarchy Process (F‑AHP) with Importance‑Performance Analysis (IPA) to assess the quality of port services in Vietnam. They revealed that “empathy” and “tangibles” were the most highly weighted dimensions in improving port competitiveness.

· Fuzzy ANP and Port Capacity Upgrades
In a Taiwanese context, Wang and Chao (2022) proposed a Fuzzy Analytic Network Process (F‑ANP) combined with Weighted Aggregated Sum Product Assessment (WASPAS) to support decision-making in port capacity upgrade projects. Their model demonstrated enhanced decision support under fuzzy environments.

· Fuzzy TOPSIS for Intangible Resources
Pak et al. (2015) utilized a Fuzzy Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) approach to evaluate intangible port resources influencing service quality, identifying and ranking key intangibles that affect performance.

· Interval‑Valued Fuzzy Hybrid (QFD‑TOPSIS)
Wang et  al. (2022) combined Quality Function Deployment (QFD) with Fuzzy TOPSIS under an interval-valued fuzzy environment (IVFE) to evaluate service performance of international container ports. Their hybrid IVFE method proved effective in handling dependent evaluation criteria and dynamic challenges, such as the disruption of COVID‑19.

These studies collectively established that fuzzy-based MCDM methodologies were appropriate for evaluating port logistics by enabling nuanced treatment of both qualitative and quantitative criteria.

Recent advances in MCDM highlighted the use and versatility of FUCOM and its fuzzy variants, particularly in risk-sensitive environments.

· F‑FUCOM for Logistics Platform Location
Ayadi et  al. (2021) developed a logistics platform location selection framework that employed F‑FUCOM to calculate the weights of criteria, followed by Fuzzy MultiAtributive Ideal-Real Comparative Analysis (F‑MAIRCA) and Fuzzy Preference Ranking Organization Method for Enrichment Evaluation (F‑PROMETHEE) to rank alternatives with sustainability considerations. Their findings underscored the predominance of economic sustainability in location decision-making.

· Spherical Fuzzy‑FUCOM-ARTASI in Maritime Ergonomic Risks
Tatar et  al. (2025) applied a spherical fuzzy‑FUCOM‑Alernative Ranking Technique Based on Adaptative Standardized Intervals (ARTASI) hybrid approach to assess ergonomic risks among port workers. This methodology enabled comprehensive risk modeling across multidimensional ergonomic criteria.

· FUCOM in Cybersecurity Risk Analysis
Kayişoğlu et al. (2024) adopted the fuzzy‑FUCOM method to rank cyber risks affecting Electronic Chart Display and Information Systems (ECDIS) aboard vessels. The study identified “malware infection via internet/intranet” as the primary threat and utilized a bow-tie framework for systemic risk visualization.

· FUCOM in the Prioritization of Green Logistics Solutions
Derse (2024) integrated Fuzzy Decision Making Trial and Evaluation Laboratory (DEMATEL), FUCOM, and Step-Wise Weight Assessment Ratio Analysis (SWARA) to evaluate problem-solving strategies for green and reverse logistics barriers. The comparative analysis across methods produced consistent priority rankings and reinforced the reliability of FUCOM in strategic decision-making.

These studies demonstrated both the methodolAogical maturity of fuzzy-FUCOM variants and their practical applicability in decision-making domains rife with uncertainty, such as ergonomics, cybersecurity, and sustainable logistics.

Beyond pure decision‑analysis techniques, the literature acknowledged broader trends in port digitalization and system integration.

· “Port of the Future” and Standardization of Digital Services
Pagano et al. (2021) proposed the C‑Ports framework, a standardized package of digital services for “Port of the Future”. Their conceptual vector included vessels & navigation, e‑freight and intermodal logistics, passenger transport, and environmental sustainability, with a case study based on the Port of Livorno.

· Business Process Management Systems in Port Processes
Martin‑Navarro et al. (2020) performed a systematic literature review on the use of Business Process Management Systems (BPMS) in ports. They concluded that applications of BPMS were notably scarce, resulting in a significant gap and area for future exploration.

· Digital Twins in Ports from Cross‑Domain Insights
Klar et al. (2023) reviewed digital twin frameworks across smart cities and supply chains and identified their potential application in ports. They emphasized three core requirements for port digital twins: situational awareness, data analytics for decision-making, and interfaces supporting multi-stakeholder governance.

These studies established the context of ongoing digital transformation and operational modernization in port environments, which was directly relevant to the adoption and evaluation of LIS.

The reviewed literature demonstrated that:

1. Fuzzy MCDM methods (e.g., F‑AHP, F‑ANP, fuzzy TOPSIS, IVFE hybrids) were aptly applied in the assessments of port performance, service quality, capacity upgrades, and intangible resources;

2. F‑FUCOM and its variants have been successfully employed in logistics platform location selection, ergonomic risk, and cyber risk assessments under uncertainty;

3. Conceptual and process modernization frameworks for ports (e.g., digital service standardization, BPMS, digital twins) are emerging but still underutilized empirically; and

4. No study to date has combined or applied fuzzy‑FUCOM with fuzzy RAWEC to systematically evaluate LIS in port operation, particularly in the Black Sea region.

These lacunae suggested obvious methodological and contextual gaps, namely: (a) the absence of dual fuzzy‑MCDM techniques (i.e., FUCOM and RAWEC) in the evaluation of port logistics systems; and (b) the lack of regionally focused empirical research on the effectiveness of LIS in Black Sea ports.

An overview of the thematic synthesis of fuzzy MCDM and port digitalization studies is presented in Table 1.

The preceding literature review underscored the relevance of fuzzy-based MCDM techniques for evaluating complex logistical systems under uncertainty. It did not only highlight significant progress in applying such methods to port operation, but also revealed notable gaps. Indeed, there was a lack of integrated application of F‑FUCOM and fuzzy RAWEC, and a striking deficiency in regional studies on the performance of LIS in the Black Sea context. Addressing these gaps, the present study pioneered the application of dual fuzzy MCDM methods, FUCOM and RAWEC, to evaluate the effectiveness of LIS in enhancing port operational efficiency across the Black Sea region.

The section of methodology consists of two parts, with F-FUCOM and its calculation steps stated in the first part, and fuzzy RAWEC explained in the second.

Full Consistency Method (FUCOM) proposed by Pamucar et al. (2018) considers the weight coefficients related to criteria in terms of specified hierarchy and takes consistency into account with respect to comparison. FUCOM applies simple algorithm including smaller number of binary comparisons in order to obtain the criteria weights (Demir et al., 2022). Consider that there are $n$ evaluation criteria represented as $w_j, j=1,2, \ldots, n$ and their weight coefficients need to be computed for an MCDM problem. According to the subjective models, decision makers need to identify the degree of impact of criterion $i$ over criterion $j$ which is stated as the value of comparison $\left(a_{i j}\right)$ with respect to acquire the weights of criteria based on binary comparison. The obtained $a_{i j}$ values are not solely based on precise measurements. The fuzzy numbers will be applied to address the uncertainties occurred in terms of subjective estimates. For this purpose, a fuzzy linguistic scale consisting of triangular fuzzy numbers, in Table 2, is taken into account for presenting the preferences of decision makers in terms of F-FUCOM (Demir et al., 2022; Pamucar & Ecer, 2020).

The algorithm related to F-FUCOM can be summarized in the following four steps (Demir et al., 2022; Pamucar & Ecer, 2020):

Step 1. Determining the criteria: Consider that there are $n(j=1,2, \ldots, n)$ evaluation criteria that are shown by $C_j=\left\{C_1, C_2, \ldots, C_n\right\}$.

Step 2. Ranking the criteria: Criteria are ranked according to the preferences of decision makers. Decision makers assign the first rank to a criterion that is expected to have the highest value related to the weight coefficient. Accordingly, the last place are assigned by decision makers to a criterion that is expected to have the lowest value in terms of weight coefficient. Thus the criteria ranking is acquired according to Eq. (1).

where, the rank related to the evaluated criterion shown as $k$. If two or more criteria have the same rankings, the ${ }^{\prime \prime}={ }^{\prime \prime}$ sign is placed between the criteria instead of ${ }^{\prime \prime}>{ }^{\prime \prime}$.

Step 3. Making comparisons related to the criteria via triangular fuzzy numbers: Comparisons related to the criteria are made by using Table 1. The comparison is made in terms of the first-ranked (most important) criterion. Thus, the significance related to fuzzy criterion $\left(\widetilde{\omega}_{c_{j(k)}}\right)$ is acquired for all the criteria that are ranked in Step 2. Since the first-ranked criterion is compared to itself (its importance $\widetilde{\omega}_{C_{j(1)}}:$ equally important), the $n-1$ comparisons related to the remaning criteria need to be made. According to the identified significance with respect to criteria, fuzzy comparative significance $\left(\tilde{\varphi}_{k /(k+1)}\right)$ is obtained via Eq. (2).

Thus, a fuzzy vector related to the comparative significance for the evaluation criteria is acquired via Eq. (3).

where, $\tilde{\varphi}_{k /(k+1)}$ shows the significance in terms of the rank of criterion $C_{j(k)}$ with respect to criterion $C_{j(k+1)}$.

Step 4. Computing the optimum fuzzy weights: The final values related to the fuzzy weight coefficients for the criteria $\left(\widetilde{w}_1, \widetilde{w}_2, \ldots, \widetilde{w}_n\right)^T$ are computed. The final values related to the weight coefficients need to satisfy the conditions stated as Eqs. (4) and (5).

In order to satisfy the aforementioned conditions, the values related to the weight coefficients $\left(\widetilde{w}_1, \widetilde{w}_2, \ldots, \widetilde{w}_n\right)^T$ need to meet the states of $\left|\frac{\tilde{w}_k}{\tilde{w}_{k+1}}-\tilde{\varphi}_{k /(k+1)}\right| \leq \chi$ and $\left|\frac{\tilde{w}_k}{\tilde{w}_{k+2}}-\tilde{\varphi}_{k /(k+1)} \otimes \tilde{\varphi}_{(k+1) /(k+2)}\right| \leq \chi$ with minimizing the value of $\chi$. The maximum consistency requirement is executed by this way. According to the identified settings, the final nonlinear model for determining the optimal fuzzy values related to the weight coefficients of the evaluation criteria $\left(\widetilde{w}_1, \widetilde{w}_2, \ldots, \widetilde{w}_n\right)^T$ can be shown as Eq. (6).

where, $\widetilde{w}_j=\left(w_j^l, w_j^m, w_j^u\right)$ and $\tilde{\varphi}_{k /(k+1)}=\left(\mid \varphi_{k /(k+1)}^l, \varphi_{k /(k+1)}^m, \varphi_{k /(k+1)}^u\right)$.

The model given in Eq. (6) can be converted into a fuzzy linear model shown as Eq. (7).

where, $\widetilde{w}_j=\left(w_j^l, w_j^m, w_j^u\right)$ and $\tilde{\varphi}_{k /(k+1)}=\left(\varphi_{k /(k+1)}^l, \varphi_{k /(k+1)}^m, \varphi_{k /(k+1)}^u\right)$.

Rough Analytical Weighted Evaluation Criteria (RAWEC) proposed by Puska et al. (2024a) aims to rank the alternatives according to the weighted deviations. RAWEC method has simple and less computational procedure and considers double normalization approach by acquiring benefit and cost criteria separately in order to provide balanced data treatment. Besides, RAWEC can be adaptable to dynamic data, maintain rank consistency by preventing rank reversal problems and proves robust in sensitivity analysis despite changes in criteria weights. Qualitative and quantitative criteria types are suuported by RAWEC method and the obtained results vary between -1 and 1 (Demir and Chatterjee, 2025). In order to handle imprecise, vague and uncertain information related to decision makers’ judgements and address real-world decision-making problems, fuzzy extension of RAWEC was developed by Nedeljkovic et al. (2024). The steps of fuzzy RAWEC can be stated as follows (Demir & Chatterjee, 2025; Nedeljkovic et al., 2024; Katrancı et al., 2025; Puska et al., 2024b):

Step 1: Alternatives are assessed by decision makers according to the linguistic values given in Table 3.

Following to that evaluation matrices are formed by transforming the linguistic values into the corresponding fuzzy numbers in terms of each decision maker’s judgement. Then the average fuzzy decision matrix $(\tilde{X})$ is acquired by averaging the values in evaluation matrices and can be seen in Eq. (8).

$\tilde{x}_{i j}=\left(x_{i j}^l, x_{i j}^m, x_{i j}^u\right)$ shows the average fuzzy value for jth criterion related to ith alternative.

Step 2: The two types of normalization consisting of maximum and minimum are applied for normalizing the average fuzzy decision matrix $(\tilde{X})$. Maximum normalization is performed for benefit and cost criteria via Eqs. (9) and (10) respectively.

Accordingly, minimum normalization is carried out for benefit and cost criteria by using Eqs. (11) and (12).

where, $x_{j \min }$ represents the minimum value related to a given criterion, $x_{j \max }$ denotes the maximum value for a given criterion.

Step 3: By using the weights acquired from the fuzzy FUCOM, the deviation value from these weights is computed via Eqs. (13) and (14). This process consists of summing the deviations related to all alternatives.

Step 4: The fuzzy numbers are transformed into crisp numbers via Eqs. (15) and (16) in terms of the defuzzification of the deviation from the criterion weight.

Step 5: Final value related to the RAWEC method $\left(Q_i\right)$ which takes value between -1 and 1 is acquired according to Eq. (17).

While the best alternative has a value closest to 1, the worst alternative takes a value closest to -1. According to these values, alternatives are ranked in descending order.

Figure 1 illustrates the integrated fuzzy FUCOM–RAWEC framework used in this study. FUCOM determines the relative importance of evaluation criteria, while RAWEC processes expert assessments to rank alternatives under uncertainty. Validation is proceeded with sensitivity, comparison, and rank-reversal analyses.