Assessment of Sustainability and Risk Indicators in an Urban Logistics Network Analysis Considering a Business Continuity Plan

Abstract

A business-continuity plan is crucial in providing an organization with the ability to maintain operations against possible risks. Therefore, companies should consider holistic risk management to sustain their activities and enhance their capabilities. Also, sustainability is able to eliminate the number of adverse environmental effects and increase the financial and social performance of a company. The purpose of this paper is to evaluate the sustainability and risk performance pillars for logistics networks, including a business-continuity plan. For this particular aim, this study considers the ten main criteria and sixty-six sub-criteria to evaluate sustainability and risk performances in logistics operations when dealing with a business-continuity plan under uncertainty. A novel and innovative four-phased integrated procedure involving a fuzzy-based AHP method with novel linguistic scales and operators is proposed. The TOPSIS technique, part of the integrated technique, is also presented to rank the alternative cities for an urban logistics network analysis. Moreover, the criteria of transportation and information infrastructures are analyzed for logistics operations. A case study of the thirty metropolitan cities in Türkiye is conducted to determine the best logistics center for a logistics firm. Several scenario analyses are performed, and a comparison study is also carried out from the literature. This study comprehensively analyzes the problem, including sustainability, risks, renewable energy and social aspects. Based on the results from the fuzzy-based AHP method, economic, safety and hazard risk are the top three main criteria. Moreover, Istanbul, Konya and Ankara are the top three alternatives for logistic networks from the results of the TOPSIS technique. Finally, managerial and policy implications are presented for policy-makers who should pay attention to the main criteria and sub-criteria in this paper for successful logistics operations dealing with the business-continuity plan when achieving Sustainable Development Goals.

1. Introduction

Supply chain management comprises various configurations, such as manufacturing-based, service-based, humanitarian, or closed-loop supply chains and it is a complicated process that includes transforming raw materials and parts into final products for the products and services needed by society and managing materials, information and finance in their flow to the final consumer. However, the risks and disruptions faced by the supply chain negatively affect economic efficiency and cause companies to lose their competitiveness and even go bankrupt. A series of disruptions caused by strikes at the Ford’s Kentucky truck factory in 2023 led to production interruptions and loss of profits of around USD 1.7 billion. In addition, the problem has put Tier-2 suppliers at risk of financial risk and bankruptcy due to lack of liquidity, high loan interest rates and difficulties in obtaining loans [1,2]. Furthermore, the deterioration in the economic outlook resulting from the COVID-19 pandemic has led to global supply barriers and increased prices in economies, thus leading to unprecedented global inflation. Central banks globally responded to the deterioration in economies by rising interest rates [3,4]. Also, in 2021, the container ship’s closure of the Suez Canal caused the interruption of trade worth more than USD 10 billion per day. As of 2024, the security risks supply chains face continue, and this situation is expected to increase the difficulties encountered in the supply chain and high inflation pressures [3,4]. Furthermore, risk management in logistics faces many complex obstacles due to centralized distribution, rapidly changing consumer expectations, lack of necessary information, management of information flows, and compliance with various standards [5].

Small and medium enterprises (SMEs) constitute nearly 90% of businesses in economies [6]. Therefore, these businesses constitute the backbone of employment, tax revenue, resource use and growth. Although the definition of SMEs may vary depending on the country, it is a common situation that these businesses face similar potential risks that may arise in every country. It is known that SMEs do not have the resources and experience required to protect and plan their assets against potential risks [7]. A business-continuity plan is of critical importance in increasing the resilience of businesses against risk exposure and ensuring their competitiveness and sustainability in the long term [8,9]. Also, significant services and operations can be kept going using a business-continuity plan during and after the interruption [10]. Unfortunately, most companies underestimate the impact of potential risks and the business’s ability to survive, as they consider the planning activities that business-continuity management refers to as unimportant [11]. Although companies in the supply chain try to take precautions against their own risks and disruptions they are exposed to, it is very important that their partners in the supply chain also have a comprehensive risk-management strategy. Because there is a positive relationship between operational and financial performance, both supply chain and company success depend on this comprehensive risk management [12].

Additionally, logistics companies face a number of restrictions due to increasing environmental awareness at the political level worldwide. For instance, it is known that the transport sector in the EU is responsible for 25% of CO₂ emissions and 60% of noxious emissions [13]. In the USA, the sector is responsible for approximately 29% of total emissions [14]. Freight transport is dependent on fossil fuels and is the only sector where emissions produced are increasing. Hence, it is possible to achieve the goals of reducing or even eliminating global emissions by using vehicles that use environmentally friendly fuel or electric energy instead of fossil fuel vehicles. Logistics companies must both comply with the changes determined by law and keep their operations running as profitably as possible. A sustainable supply chain requires finding a solution by balancing the three main pillars of economic efficiency, environmental stability and social equity [15]. The Paris Climate Agreement, which is an important step in the climate change process, is binding in the fight against climate change. Targets to limit global temperature can contribute to the sector operating in a more sustainable and energy efficient way [16].

The use of environmentally friendly vehicles in the industry initially emerged as a global strategy to lessen emissions and was supported with government incentives. According to recent studies, it has been calculated that environmentally friendly vehicles, especially electric trucks, have a cost advantage exceeding 10%. Additionally, with the development of battery technology, it is estimated that this advantage will reach 50% in 2030 [17]. For instance, global companies such as DHL, UPS, Amazon, FedEx, etc., have already started using environmentally friendly vehicles in their fleets and are planning investments to increase the number of these vehicles in the near future [18].

Sustainable Development Goals, which are accepted as a global call by the United Nations, have been determined to fight poverty, reduce environmental impact and contribute to peace and prosperity [19]. These goals can be achieved with the support of society and efficient planning of information, technology and financial resources. The economic, environmental and social aspects of societies can be contributed by sustainable supply chain networks. Therefore, the main pillars of sustainability and risk management have diverse ties with each other [20]. More specifically, there are 25 targets in the United Nations’ Sustainable Development Goals, 10 of which are related to risk reduction which is also expressed as a basic development strategy [21]. Risk management is a systematic approach that considers the potential exposure of an organization’s assets and contributes to its social, environmental and governance performance [22]. Sustainability enables the business to focus on activities that create value in the long term, such as risk management, and to progress in line with the three main goals mentioned by encouraging continuous improvement. Thus, Sustainable Development Goals and risk management should be managed jointly and holistically from a perspective for urban city planning [23].

The contributions of this paper are five-fold. First, business-continuity plans and sustainability are considered simultaneously to improve logistic network operations and reduce risks and disruptions. Moreover, to the best of our knowledge, business-continuity plans, renewable energy sources, potential risks, transportation infrastructure, information infrastructure and social criteria have not been considered in the literature for an urban logistics network analysis. Second, the ten main criteria and sixty-six sub-criteria are specified to prioritize weights related to the published paper, official reports and the ten decision makers’ opinions. In addition, risk and sustainability criteria are also used to measure logistics network performance. Third, a novel and innovative integrated procedure is presented to evaluate performance pillars and rank the alternatives. A fuzzy-based AHP method is presented using novel linguistic scales and operators. Fourth, the importance of transportation and information infrastructures is examined for logistics operations. Fifth, a case study of the thirty metropolitan cities in Türkiye is provided to select an effective logistic center location with different scenario analyses and compare the previous research in the literature. In addition, Türkiye is selected for the case study due to its strategic geographical importance and the goal of exporting logistics services. The Sustainable Development Goals are also discussed based on the results. Figure 1 presents the flowchart of the study.

Specifically, Türkiye’s unique position as a transcontinental bridge between Europe, Asia and the Middle East enables it to serve as a pivotal logistics hub in global supply chains. Moreover, Türkiye is a key member of the Middle Corridor initiative, a strategic alternative to both the Northern (Russia) and Southern (Suez) trade routes under China’s Belt and Road Initiative. Türkiye’s logistics market is expected to experience significant growth over the coming years, driven by increasing infrastructure investments, expanding trade networks and its strategic role in regional and intercontinental supply chains. Furthermore, Türkiye’s diverse topography and socio-economic distribution among its 30 metropolitan cities offer an ideal setting to evaluate urban logistics sustainability and risk indicators under varying demographic, environmental and infrastructural conditions—making the results more generalizable to similar emerging economies.

The rest of this paper is organized in the following way. The literature review is explained in Section 2, a detailed way to evaluate logistics sectors. Next, the problem statement and developed four-phased integrated method are exhibited in Section 3 when dealing with uncertainty. Then, the application of the introduced method is presented, including the explanation of the main criteria and sub-criteria, the prioritized weights, the ranking of the cities and a comparison study from the literature in Section 4. Further, managerial implications are discussed in Section 5, including the Sustainable Development Goals. Lastly, the conclusion of the study is given in Section 6.

2. Literature Review

The relevant studies in the literature are reviewed in this section. Notably, Ghadge et al. [24] conducted a comprehensive review of climate change risks in supply chains in the literature between 2005 and 2018. The literature survey indicates that risks from extreme weather significantly impact food production, use of natural resources and transport worldwide. The study also states that natural disasters mutually trigger each other through greenhouse gas (GHG) emissions.

Various disruptions encountered by supply chains have a negative consequence on their efficiency and robustness. Particularly, Kleindorfer and Saad [25] proposed a framework based on supply chain coordination models to manage these potential risks. Potential disruptions were considered as natural disasters, strikes, economic problems and terrorist incidents. The proposed method consists of the stages of identifying, evaluating and reducing these risks. Along the same lines, Wu et al. [26] addressed a similar problem in order to identify supplier-focused risk factors and manage inbound supply risks. They grouped potential supply chain risks into internal or external sources and their controllability. They also used the analytic hierarchy process (AHP) method to evaluate risk factors. Next, Torabi et al. [27] proposed a series of frameworks in the context of business-continuity-management systems to assess potential risks in organizations. This framework identifies potential threats to the organization, facilitates the risk assessment and subdivides them into disruption and operational risks. Moreover, an effective best–worst method (BWM) is adopted to assign the weights of sub-factors. Then, Harclerode et al. [28] conducted research in the field of remediation management of complex sites and identified a set of metrics by analyzing the current situation on sustainable risk management. Drivers for risk management are classified under three main headings: physical/environmental and socio-economic. Additionally, the most common factors are technical limitations, length of the process and financial feasibility.

Liu et al. [29] proposed an evaluation system for the smart logistics eco-index consisting of 24 quantitative and qualitative indicators. These indicators were defined as digital integration and visual and intelligent operations. To determine the weights of each indicator, a mathematical model was integrated. In addition, a hybrid decision-making method was proposed for evaluation and the index results of a company operating in the logistics sector in different months were evaluated. Later, Liu et al. [30] studied the risks of the smart logistics ecological chain and divided them into member, technology and external environmental risks. Social network analysis (SNA) was used in the study to analyze existing risks, and the TOPSIS method was employed to evaluate risks. The results of the case study in the logistics sector indicate that technological route change, state management and industry-management risks can easily trigger secondary risks. Next, Kodym et al. [31] proposed a framework that may use blockchain technology against the possible risks faced by logistics 4.0. These risks are defined under five headings: economic, technical/IT, social, environmental and legal/political risks.

Gupta and Singh [32] studied the sustainability of logistics service providers and evaluated the sustainability of service providers in India with the help of the index. The results indicate that management collaboration, business-continuity plans and financial sustainability significantly impact risk reduction. Further, De Oliveira et al. [33] analyzed the company’s reverse logistics risk management of solid waste in the steel industry based on ISO 31000 [34]. As a result, it has been stated that poor separation of waste has been identified as a risk with a significant potential impact. In addition, the risk posed by these pollutants poses a risk to public health and biodiversity, as well as the risk of loss of income from the sale of these wastes. Along the same lines, Majumdar et al. [35] addressed the possible risks and their impacts by considering the green clothing supply chain. Risks are classified into five different categories: supply and demand, process, business environment and financial risks. The authors evaluated the impact of the expressed risks using the fuzzy analytical hierarchy process (FAHP). While the results obtained show that the impact of financial and business environment risks is high, it is stated that the probability of occurrence of supply, demand and process risks is high. In addition, a vulnerability matrix is proposed in the study, where each risk is defined according to its potential impact and probability. Next, Khan et al. [36] identified and prioritized potential risks considering the Halal supply chain. Considering the literature review and expert opinion, they identified 42 risks in seven classes. The main risk categories are planning-, sourcing-, production-, logistics- outsourcing-, market-, information technology- and sustainability-related risks. The authors applied the fuzzy best–worst method (FBWM). The results indicate that the production-related risk is the most critical, while the sustainability-related risk is trivial.

Moktadir et al. [37] attempted to identify SC risk factors specific to the leather industry. Forty-four risk factors belonging to the determined field were identified through a literature review and interviews with field experts. Risk is classified as social, environmental, economic, technical and institutional factors. Pareto analysis and BWM were applied. The study results indicate that water treatment, consumer preference, solid waste disposal, cost volatility, etc., are important risk factors. Moreover, Goswami et al. [38] studied an analytical framework for the product design concept that takes into account technical and commercialization issues, as well as key supply chain risks. The risks taken into account in product design are logistics, operational, planning, market, service, sourcing, customer- and company-related risks. Next, the conditional probabilities of the root nodes and parent nodes were determined employing a Bayesian network (BN). Different risks and design concepts were expressed by a quantitative measure defined as SCRI (supply chain risk index). The results indicate that the risk potential of the design is inversely proportional to its benefit. Furthermore, Sakib et al. [39] studied disasters faced by the oil and gas supply chain (OGSC) and used the BN model to evaluate them based on seven different factors. The identified factors are technical, safety, environmental, economic, political, social and legal. The results indicate that technical factors have the highest impact, while legal and political factors have the lowest impact. Next, Rama et al. [40] employed data mining techniques, namely Classification and Regression Trees (CART) and Random Forest, so as to analyze 32 Spanish cities according to 38 sustainable indicators. According to the analysis results, the authors estimated the urban sustainability of the cities with a-85% accuracy according to indicators. Similarly, Liu et al. [41] adopted exploratory factor analysis and confirmatory factor analysis and applied it to Chinese cities to create a new metric system consisting of 5 subsystems and 18 indicators that evaluate urban sustainability. It has been stated that the environment, infrastructure, policies, health and population are more important in terms of sustainability.

Kusakci et al. [42] developed the Sustainable Cities Index, which consists of economic, social, environmental and institutional dimensions and evaluated 30 metropolitans in Türkiye according to this index. A three-stage method based upon Interval Type-2 Fuzzy AHP and COPRAS was applied for the evaluation. Next, Barmuta et al. [5] studied risk management in the logistics sector. The authors classified the risks in logistics processes related to several external and internal factors affecting the logistics system that companies face. The risks faced by the company are expressed as follows: Organizational, information, supplier-partner, customer-related, market, transportation and force majeure. Then, Yousefi and Tosarkani [43] proposed a sequential dynamic to evaluate risk reduction measures, considering that improving performance and sustainability in logistics processes can be resolved by performing risk analysis. The method comprises the Z-number, Dempster–Shafer evidence theory and Bayesian belief network. The processes in the automotive supply chain are resource and production planning, procurement, loading and unloading, transportation, etc. In addition, the method allows different decision-makers to identify risks in logistics sub-processes and take measures to reduce risks with new technologies.

Gupta et al. [44] addressed the same problem to identify critical factors affecting food supply chains under uncertainty. First, the criteria were determined using the Delphi method and then weighted using the fuzzy AHP. The criteria were evaluated in four classes: durability–flexibility, communication–collaboration, transparency–traceability and regulation–standardization. Research results indicated that R&D investment, transparency–traceability, level of risk-sharing and collaboration among stakeholders are the critical criteria. Further, Gurbuz et al. [45] examined the sales-related disruptions and risks faced by SMEs and their supply chain partners in COVID-19. The results obtained from the case analysis show that developing different collaborations and using multiple resources as a precaution against disasters are more beneficial to the company. To select a Blockchain Technology-IoT platform for the SC network, Tavana et al. [46] applied a hybrid approach. First, the weights were calculated using the interval Weighted Impact Nonlinear Measurement System (WINGS) method. Then, the alternatives were evaluated using the interval VIseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) technique. The main criteria are popularity and support, transition complexity, operational features and technical. Next, Vergara et al. [47] proposed an approach involving fuzzy DEMATEL and ANP to evaluate resilient and sustainable supply chain performance. The proposed hybrid approach has been applied to an automotive supply chain in Mexico. While the results obtained emphasize that the social dimension is critical, it is stated that the environmental dimension is not essential in the chain.

By surveying Chinese manufacturing companies, Song et al. [48] investigated how the digital supply chain affects the disruption risks of the supply chain. The research results not only express the advantages of the digital supply chain but also state that the dynamic capacity of the chain can reduce the risks of supply, production, demand and circulation interruptions. More recently, Saputro et al. [49] studied the well-known, most discussed green supplier-selection problem with a several risk criteria extension. The risks are price changes, production failures, financial and capacity instability, delays, product defect rate and material shortage. To select suppliers, the authors employed a hybrid method involving fuzzy AHP-Combined Compromise Solution and Fuzzy Stepwise Weight Evaluation Ratio Analysis-Failure Mode and Effects Analysis, which are frequently used techniques in the literature.

3. Problem Statement and Proposed Integrated Methodology

This section consists of two subsections as follows. First of all, the problem statement is presented for the assessment of sustainability and risk performance. Next, a novel integrated procedure is proposed to prioritize the weights and rank the alternatives.

3.1. Problem Statement

In the context of logistics networks, a business-continuity plan is a strategic framework designed to ensure the uninterrupted operation of critical functions during and after disruptive events [8,9,10]. Key components of a business-continuity plan include risk and business impact assessment (evaluating how hazards affect core logistics functions), continuity strategies (such as rerouting and infrastructure redundancy), recovery objectives (defining acceptable downtime and data loss thresholds), clearly assigned roles and responsibilities and regular testing and refinement through simulations [8,9,10]. In this study, these elements are operationalized through a multi-criteria framework, where criteria such as transportation and information infrastructures, legal risks and process performance reflect the capacity of urban logistics systems to maintain service resilience and adapt to uncertainty. By integrating business-continuity plan principles with sustainability and risk indicators, our approach supports robust and proactive logistics planning. Also, they should specify Sustainable Development Goals to reduce adverse environmental impacts as part of the business-continuity plan. For these reasons, multi-criteria decision-making (MCDM) techniques are applied to evaluate sustainability and risk indicators when addressing uncertainty [8,23,42]. Also, experts’ opinions, published papers and reports are incorporated into specifying main criteria, including sustainability and risk pillars, along with their sub-criteria. Moreover, experts’ judgments may be vague and may not estimate their preferences with exact numerical data due to their preferences. Therefore, a fuzzy-based AHP technique is utilized as a part of the proposed integrated procedure to prioritize the weights associated with the experts’ judgments.

3.2. Proposed Integrated Procedure Under Uncertainty

This section presents an integrated procedure to evaluate the sustainability and risk pillars under uncertainty in logistics network analysis. The first part of the procedure is related to specifying the main criteria and sub-criteria depended upon the official reports, experts’ opinions and academic papers. The second part of the procedure is to obtain the weights when applying the fuzzy-based AHP. The third part of the procedure ranks the alternatives using the TOPSIS method. Finally, comparison studies are conducted with the different scenario analyses. The steps of the proposed integrated procedure are provided as follows:

Step 1: Specify a group of experts.

Step 2: State the main criteria and sub-criteria depended upon the afore-mentioned stakeholders for the problem.

Step 3: Start the procedure of the fuzzy-based AHP method, and then go to step 4. Note that the fuzzy extension is necessary due to the vagueness and the lack of exact numerical descriptions of the experts’ judgments.

Step 4: Pairwise comparison matrices (PCMs) are generated associated with the linguistic scale, and experts make selections to fill the PCMs with the fuzzy number scales in

Table 1. Linguistic scale and its associated fuzzy number scale.

Linguistic Scale	Fuzzy Number Scale
Definitely less important linguistic term (DLILT)	(0.100, 0.111, 0.125)
Very strongly less important linguistic term (VSLILT)	(0.125, 0.143, 0.166)
Strongly less important linguistic term (SLILT)	(0.166, 0.200, 0.250)
Marginally less important linguistic term (MLILT)	(0.250, 0.333, 0.500)
Equal important linguistic term (EILT)	(1.000, 1.000, 1.000)
Marginally more important linguistic term (MMILT)	(2.000, 3.000, 4.000)
Strongly more important linguistic term (SMILT)	(4.000, 5.000, 6.000)
Very strongly more important linguistic term (VSMILT)	(6.000, 7.000, 8.000)
Definitely more important linguistic term (DMILT)	(8.000, 9.000, 10.000)

. Triangular fuzzy numbers (TFNs) are employed in Table 1, and ${\tilde{x}}_{ijk}=\left(a_{ijk},b_{ijk},c_{ijk}\right)$ denotes the lower, middle, or upper values of the TFN, respectively. In addition, TFNs are selected in this study due to computational efficiency, widespread use in MCDM literature and proven adequacy in representing expert uncertainty with minimal cognitive burden on stakeholders. Although alternative fuzzy numbers such as trapezoidal or Gaussian could be employed, TFNs offer a favorable trade-off between expressiveness and mathematical tractability, especially when multiple expert evaluations and consistency checks are involved.

Step 5: Aggregate the experts’ selections using the fuzzy numbers, ${\tilde{x}}_{ijk}=\left(a_{ijk},b_{ijk},c_{ijk}\right)$, as follows:

$$\begin{array}{c}{\tilde{x}}_{ij}=\left(a_{ij},b_{ij},c_{ij}\right)\left(k=1,2,\dots ,n\right) \mathrm{is} \mathrm{the} \mathrm{aggregated} \mathrm{fuzzy} \mathrm{number}\hfill \\ \mathrm{where} a_{ij}={\left({\displaystyle \prod _{k=1}^n}{a}_{ijk}\right)}^{{\displaystyle \frac{1}{n}}},b_{ij}={\left({\displaystyle \prod _{k=1}^n}{b}_{ijk}\right)}^{{\displaystyle \frac{1}{n}}},c_{ij}={\left({\displaystyle \prod _{k=1}^n}{c}_{ijk}\right)}^{{\displaystyle \frac{1}{n}}}\hfill \end{array}$$

(1)

Step 6: Apply the defuzzification of the aggregated fuzzy number as follows:

$$Deff\left({\tilde{x}}_{ij}\right)={\left(a_{ij}\times b_{ij}\times c_{ij}\right)}^{{\displaystyle \frac{1}{3}}}$$

(2)

where $Deff\left({\tilde{x}}_{ij}\right)$ denotes the defuzzification function.

Step 7: Apply the normalization of defuzzifed aggregated pairwise comparison matrices for the following formula.

$${d_{ij}}^{\prime }={\displaystyle \frac{Deff\left({\tilde{x}}_{ij}\right)}{{\displaystyle \sum _{i=1}^N}Deff\left({\tilde{x}}_{ij}\right)}}\left(i,j=1,2,\dots ,N\right)$$

(3)

Step 8: Obtain the weights for the specified criteria as follows:

$$n{w}_j=\left({\displaystyle \frac{1}{N}}\right)\sum _{i=1}^N{d_{ij}}^{\prime }\left(i,j=1,2,\dots ,N\right)$$

(4)

where $n{w}_j$ denotes the jth weight.

Step 9: Calculate a consistency ratio (CR) to check the credibility of the preferences using the following formula.

$$\begin{array}{c}CR={\displaystyle \frac{CI}{RCI}}\hfill \\ \mathrm{where}CI={\displaystyle \frac{\left({\lambda }_{max}-N\right)}{\left(N-1\right)}}, {\lambda }_{max} \mathrm{is} \mathrm{the} \mathrm{principal} \mathrm{eigenvalue},\hfill \\ \mathrm{and}RI \mathrm{is} \mathrm{the} \mathrm{random} \mathrm{consistency} \mathrm{index} \left(\mathrm{RCI}\right).\hfill \end{array}$$

(5)

Step 10: The threshold of CR is 0.10. If the threshold is greater than 0.10, the preferences of the experts should be revised until they obtain a CR value that is smaller than 0.10.

Step 11: Repeat Steps 3–10 for the main and all sub-criteria. Then, go to Step 12

Step 12: Start the procedure of the TOPSIS method, and then go to Step 13.

Step 13: The normalized multi-attribute decision-making matrix is found using the following formula:

$$\begin{array}{c}fno{r}_{ij}=f_{ij}/\sqrt{{\displaystyle \sum _{i=1}^n}{f}_{ij}^2}\hfill \end{array}$$

(6)

where $fno{r}_{ij}$ the normalized performance value.

Step 14: The ideal solutions are identified based on the normalized performance value using the following formulas.

$$\begin{array}{c}PI{S}_{TOPSIS}=\left[fno{r}_1^{+},fno{r}_2^{+},\dots ,fno{r}_m^{+}\right]\hfill \\ NI{S}_{TOPSIS}=\left[fno{r}_1^{-},fno{r}_2^{-},\dots ,fno{r}_m^{-}\right]\hfill \\ \mathrm{where} fno{r}_j^{+}=\left\{\begin{array}{c}\max fno{r}_{ij}\mathrm{for} \mathrm{benefit} \mathrm{criteria}\hfill \\ \min fno{r}_{ij}\mathrm{for} \mathrm{cost} \mathrm{criteria}\hfill \end{array}\right.\hfill \\ fno{r}_j^{-}=\left\{\begin{array}{c}\min fno{r}_{ij}\mathrm{for} \mathrm{benefit} \mathrm{criteria}\hfill \\ \max fno{r}_{ij}\mathrm{for} \mathrm{cost} \mathrm{criteria}\hfill \end{array}\right.\hfill \end{array}$$

(7)

Note that $PI{S}_{TOPSIS}$ and $NI{S}_{TOPSIS}$ are the positive and negative ideal solutions, respectively.

Step 15: The weighted Euclidean distances are obtained from the positive and negative ideal solutions of each alternative as follows:

$$\begin{array}{c}E{D}_i^{+}=\sqrt{{\displaystyle \sum _{j=1}^n}n{w}_j{\left(f_{ij}-fno{r}_j^{+}\right)}^2}\hfill \\ E{D}_i^{-}=\sqrt{{\displaystyle \sum _{j=1}^n}n{w}_j{\left(f_{ij}-fno{r}_j^{-}\right)}^2}\hfill \end{array}$$

(8)

where $E{D}_i^{+}$ and $E{D}_i^{-}$ denote the weighted Euclidean distances based on the positive and ideal solutions, respectively. Note that, as indicated in Step 11, all $n{w}_j$ weights are obtained separately for the main criteria and sub-criteria. The global weights are calculated by multiplying the main criteria weights with the sub-criteria local weights. Also, the local sub-criteria weights are obtained using Equation (4).

Step 16: The overall performance score (OPS) is found for each alternative using the following equation.

$$OP{S}_i={\displaystyle \frac{E{D}_i^{-}}{E{D}_i^{-}+E{D}_i^{+}}}$$

(9)

where $OP{S}_i$ represent the ith overall performance score for the ith alternative.

Step 17: Rank the alternatives based on the overall performance score. Next, go to step 18 for the comparison study.

Step 18: Conclude the results for the MCDM problem with the different scenario analyses, including the traditional counterpart and provide managerial implications.

4. Application of the Proposed Methodology Under Uncertainty

This section consists of four subsections as follows: criteria and alternatives, prioritizing the criteria, ranking cities and comparison of the cities. The sub-section of criteria and alternatives explains the main criteria and sub-criteria in Section 4.1 when presenting the alternatives for the problem. Also,

Table 2. Brief explanations of the criteria.

C1	Economic	Explanation
C1.1	Economic effect of CO2 emission	It measures the economic effect of carbon dioxide released.
C1.2	Energy consumption	It measures the amount of energy consumed.
C1.3	Domestic material consumption	It measures the amount of materials used in production.
C1.4	Investment	It measures the amount of gross fixed capital formation.
C1.5	Volatility of price and cost	It expresses how widely prices and costs fluctuate.
C1.6	Disposal costs of hazardous wastes	The amount of money accrued for the removal or disposal of waste or materials remaining at the end of a production process.
C1.7	Maintenance costs	It refers to the total costs encountered by a business while operating.
C1.8	Fiscal changes	It refers to the use of expenditures and tax policies by the government to direct the total demand for the sector, employment, inflation and economic growth.
C1.9	Number of Enterprises	It measures the number of enterprises operating in the sector.
C2	Safety
C2.1	Inadequate personal protective equipment	It counts the number of uses of inadequate protective equipment during control.
C2.2	Dangerous working condition	It refers to the number of cases of unsafe working conditions encountered in the workplace during controls.
C2.3	Mortality caused by traffic injury	It refers to the number of deaths resulting from traffic injuries.
C2.4	Explosions, fires, chemical accidents	It refers to the number of chemical accidents that occur in the workplace.
C2.5	Machine, equipment or facility failure	It refers to the number of machine, equipment or facility failures.
C3	Hazards risk
C3.1	Earthquake	The term describes the exposure to earthquakes in both absolute and relative terms.
C3.2	Flood	The term describes the exposure to floods in both absolute and relative terms.
C3.3	Epidemic	The term describes the exposure to epidemic in both absolute and relative terms.
C3.4	Landslide	The term describes the exposure to landslides in both absolute and relative terms.
C3.5	Forest fire	The term describes the exposure to forest fires in both absolute and relative terms.
C4	Legal risk
C4.1	Judicial Independence	It assesses how independent the judiciary system is from government, individuals, or companies.
C4.2	Legal Framework Efficiency	It refers to the effectiveness of legal and judicial systems in resolving disputes.
C4.3	The Protection of the Intellectual Property	It indicates the level of protection afforded to intellectual property rights.
C4.4	Land Administration Quality	It describes the quality of land management.
C4.5	Uprooted Refugees	It measures the impact of uprooted refugees on a city.
C5	Energy
C5.1	Renewable energy consumption	The indicator refers to energy consumption from all renewable sources.
C5.2	Renewable energy production	The indicator refers to energy production from all renewable sources.
C5.3	Primary energy supply	It measures the primary energy supply.
C5.4	Electricity generation	It measures in gigawatt hours and as a percentage of total energy production.
C6	Environmetal impact and utilization
C6.1	Air pollution (micrograms per cubic meter)	It measures the average annual concentration of fine particles suspended in the air (PM2.5).
C6.2	Air pollution, population exposure	It measures the percentage of the population exposed to PM2.5 concentrations
C6.3	Total GHG emissions	It measures total greenhouse gas emissions
C6.4	Biodiversity covered by protected areas	It defines areas designated for the long-term protection and maintenance of biological diversity.
C7	Transportation Infrastructure
C7.1	Air transport, freight tonne-kilometres	It is the sum obtained by multiplying the weight of the load carried on each flight by its distance.
C7.2	Road network	It measures the total length of the road network.
C7.3	Container port traffic	It measures container flow in a standard size container in port container traffic.
C7.4	Quality of air transport infrastructure	It refers to the quality of air transportation infrastructure.
C7.5	Quality of port infrastructure	It refers to the quality of port infrastructure.
C7.6	Quality of railroad infrastructure	It refers to the quality of railroad infrastructure.
C7.7	Green number of vehicles	It refers to the number of green/electrical vehicles.
C7.8	Number of energy/charge stations	It refers to the number of energy/charge stations.
C8	Information Technology Infrastructure
C8.1	Secure Servers	It refers to the number of secure socket layer protocol certificates.
C8.2	Information and communication technology	It defines the level of development of information and communication.
C8.3	Broadband subscriptions	It measures broadband subscriptions by the total number of subscribers.
C8.4	System Failure	It measures recorded outages that occur in the system or network.
C8.5	Network security	It defines its protection against breaches, intrusions and other threats occurring in the network.
C8.6	Data corruption	It describes errors encountered that cause undesirable changes in the original data.
C8.7	IoT	It refers to the platform’s integration with new technologies such as sensors, RFID and NFC and the internet.
C8.8	Software	It refers to the software malfunctions or failures.
C9	Process/operation
C9.1	Lack of tech skills	It measures the technical expertise of labor force.
C9.2	Machinery breakdowns	It measures machine malfunctions encountered in a certain period.
C9.3	Shipping damages	It refers to the damage of the products during transportation and transportation.
C9.4	Poor process output	It refers to the efficiency of processes.
C9.5	Changing consumer preferences	It describes its flexibility in responding to the changing consumer.
C9.6	Supplier failure	It measures the total or partial failure of suppliers or service providers or disruption in the supply of products or the provision of a particular service.
C9.7	Raw materials shortage	It refers to the possible raw material shortage.
C9.8	Poor supplier selection	It refers to the disruption caused by the supplier’s failure to fulfill its obligations.
C10	Social
C10.1	Rural Access Index	It considers the proportion of the rural population who live within 2 km of an all-season road.
C10.2	Convenient access to public transport	It takes into account people’s access to public or private transportation.
C10.3	Passenger car registrations	It measures the number of new passenger cars or vehicles registered.
C10.4	Population	It measures the population of the metropolitan area.
C10.5	Proportion of informal employment	It measures the rate of informal employment.
C10.6	Manager–employee relationships	It defines the relationship between manager and employee in the workplace.
C10.7	Workplace culture	It refers to the institutionality of the organization where the workforce works.
C10.8	Successful partnership among partners	It refers to a mutually beneficial partnership or collaboration.
C10.9	Business ethics	It refers to moral principles, policies and values that may harm internal or external stakeholders.
C10.10	Vendor or Supplier Breaches	It refers to vendor or supplier breaches in a certain period.

denotes brief explanations of the criteria. Then, the importance of the main criteria and sub-criteria weights is discussed in Section 4.2. Next, cities are ranked, including the different scenario analysis in Section 4.3. Lastly, the presented results in this study are compared to the results of the previous research in the literature in Section 4.4.

4.1. Criteria and Alternatives

Being a local and/or global stakeholder in the supply chain network can affect all stages. In addition, determining the location correctly can affect the success and investment life of all stages, from the design of the facilities to be invested to the delivery process. Therefore, the analysis of cities is important in the first stage of determining the potential for the logistics centers and analyzing the current situations. The ten main criteria and sixty-six sub-criteria are determined for this particular purpose, and details are provided below. In addition, the evaluations performed in this section were compiled from expert opinions, academic studies and ministry official statistical reports [50,51,52]. Please see Figure 2 for alternatives.

C1. Economic. While supply chain in its traditional definition means transmitting a product or service from the manufacturer to the end customer, in modern terms, it is important to offer these products at an affordable cost by improving the living standards of society. The economic dimension is vital in the supply chain. The existence of increased competition, possible unpredictable financial changes and volatility in prices and costs may affect the investment environment and the time value of money and lead to uncertainty. This may cause costs to increase and control to become difficult.

C2. Safety. The report published by the International Labor Organization (ILO) estimates that more than 2.7 million workers die every year from work accidents and occupational diseases. Additionally, the report states that more than 160 million employees lost their health due to workplace risks [53]. It is stated that the economic dimension of this social problem is more than 4% of the global GDP. Reducing and eliminating occupational diseases can be achieved by reducing the hazards in the workplace and by knowing and preventing the hazards that may arise from work processes, operations and equipment. Additionally, road traffic accidents kill approximately 1.2 million people every year, with more than 90% of deaths occurring in low- and middle-income countries [54].

C3. Hazards risk. As the global supply chain expands and becomes more connected, disrupting a particular part of the network could lead to significant shortages that disrupt the supply network. With the concern of reducing costs, the network spread over a wide area has become interdependent, and unforeseen delays can cause severe disruptions and fluctuations in economies [55]. From this perspective, identifying potential critical risks and mitigating risks in order to reduce exposure to assets is essential for business continuity [56]. Risks such as earthquakes, floods, pandemics, landslides, forest fires, etc., are considered minor but increasing and frequently encountered risks in this logistics network.

C4. Legal risk. As a part of the global supply chain network, companies need to meet the obligations imposed by the countries in which they operate and international collaborations. Legal risk is considered with indicators related to judicial independence, resolution of disputes that may arise, intellectual property, land-administration quality and refugees.

C5. Energy. Sustainable energy is one of the critical issues for supply chain management. The importance of sustainable energy is emphasized with Goal 7 of the UN’s 2030 sustainable development agenda, which is related to providing affordable, reliable, sustainable and modern energy to society regarding its economic, environmental and social impacts [19]. In this way, environmental impact will be reduced by reducing carbon emissions, and dependence on conventional fuels will be reduced.

C6. Environmental impact and utilization. Supply chain management has become an essential part of the business world today. Increasing awareness about sustainable supply chains is forcing the business world to take steps to protect people and the environment along the entire chain. Environmental impacts of supply chain activities include hazardous air emissions, biodiversity loss, deforestation, toxic waste, water security and long-term damage.

C7. Transportation infrastructure. Supply chain infrastructure is a critical component of the physical and digital systems and networks that transport, store and manage products from origin to destination. It is essential that this infrastructure is developed, which increases efficiency, reduces costs and strengthens competitiveness [57,58]. Ports, highways, railways, airports, warehouses and distribution centers where products are transported are important components of this network structure.

C8. Information technology infrastructure. Information technologies play an essential role in more effective supply chain management, faster response to customer needs and providing resilient, sustainable transportation and supply ecosystems for safer, cleaner and more inclusive movement of products [59].

Figure 2. Map of metropolitan municipalities in Türkiye [60]. The English meaning of “MARMARA DENİZİ” is “SEA OF MARMARA”. The English meaning of “EGE DENİZİ” is “AEGEAN SEA”. Moreover, the colored areas in the Türkiye map represent the metropolitan municipalities in Türkiye.

Figure 2. Map of metropolitan municipalities in Türkiye [60]. The English meaning of “MARMARA DENİZİ” is “SEA OF MARMARA”. The English meaning of “EGE DENİZİ” is “AEGEAN SEA”. Moreover, the colored areas in the Türkiye map represent the metropolitan municipalities in Türkiye.

C9. Process/operation. In supply chain management, it is essential for operations/processes at different levels and in various disciplines to coordinate to meet customers’ needs and reduce operational efficiencies and expenses in line with company goals. The process, which starts with analyzing customers’ needs, covers a range of activities, including sourcing, warehousing, inventory management, planning, distribution, retail and recycling.

C10 social. The social sustainability of supply chains is critically important due to its moral, ecological, legal and economic dimensions. The United Nations guides businesses, companies and states with adaptations that respect human rights and protect victims against abuses. Social sustainability considers the goals that provide a suitable working environment for employees, contribute to fair trade and equality standards and increase the quality of life.

4.2. Prioritizing the Main Criteria and Sub-Criteria

This sub-section is associated with the second part of the introduced integrated procedure to obtain weights of the main criteria and sub-criteria. For this particular aim, the developed fuzzy-based AHP, including steps (3)–(11) in the presented procedure, is applied to prioritize the criteria, such as Economic, safety, hazards risk, legal risk, energy, environmental impact and utilization, transportation infrastructure, information technology infrastructure, process/operation and social before assessing the alternatives. Also, the ten decision-makers (DMs) are experts in information systems, risk management and logistics network analysis. Moreover, all DMs have bachelor of science (BS) degrees, and the four DMs also have doctor of philosophy (PhD) degrees. In addition, the ten DMs have at least five years of experience in their work. The ten DMs were selected based on their educational background, work experience and knowledge about urban logistics. Further, the ten DMs were asked to generate the PCMs as denoted in the fourth step of the presented integrated procedure for the main criteria and sub-criteria when using the linguistic scale and its associated fuzzy number scale. Each DM independently evaluated the main criteria and sub-criteria. The credibility of the preferences was checked using the CR. Thus, reliability is provided for the judgments of aggregated PCMs.

Table A1. Aggregated PCM for the main criteria based on the experts’ selection when applying the defuzzification.

	C1	C2	C3	C4	C5	C6	C7	C8	C9	C10
C1	1.0000	1.0000	1.0000	2.5946	1.0000	2.8845	2.5946	1.8882	1.7007	2.5946
C2	1.0000	1.0000	1.0000	1.1117	1.0000	3.0435	2.5946	1.8882	2.0992	2.8845
C3	1.0000	1.0000	1.0000	1.0000	2.3338	2.8845	2.3338	2.0992	1.0000	2.5946
C4	0.3854	0.8995	1.0000	1.0000	1.9922	1.3741	2.5946	0.1648	1.0000	2.8845
C5	1.0000	1.0000	0.4285	0.5019	1.0000	1.0000	1.0000	2.8845	2.8845	2.8845
C6	0.3467	0.3286	0.3467	0.7277	1.0000	1.0000	0.3466	0.3466	2.8845	1.6984
C7	0.3854	0.3854	0.4285	0.3854	1.0000	2.8854	1.0000	1.0000	1.0000	2.8845
C8	0.5296	0.5296	0.4764	6.0672	0.3467	2.8854	1.0000	1.0000	1.0000	2.8845
C9	0.5880	0.4764	1.0000	1.0000	0.3467	0.3467	1.0000	1.0000	1.0000	1.0000
C10	0.3854	0.3467	0.3854	0.3467	0.3467	0.5888	0.3467	0.3467	1.0000	1.0000

,

Table A2. Aggregated PCM for the sub-criteria of economics based on the experts’ selection when applying the defuzzification.

	C1.1	C1.2	C1.3	C1.4	C1.5	C1.6	C1.7	C1.8	C1.9
C1.1	1.0000	1.3741	1.3741	0.2379	0.2649	0.2795	0.2950	0.2510	0.2379
C1.2	0.7277	1.0000	1.0000	0.3466	0.2025	0.3466	0.3284	0.3284	0.3112
C1.3	0.7277	1.0000	1.0000	0.2025	1.2360	0.3466	0.3466	0.3466	0.2136
C1.4	4.2035	2.8855	4.9390	1.0000	1.0000	1.8882	2.6680	2.3338	1.1117
C1.5	3.7751	4.9390	0.8091	1.0000	1.0000	2.3338	2.3338	1.1117	2.8845
C1.6	3.5775	2.8855	2.8855	0.5296	0.4285	1.0000	1.0000	0.2025	0.3466
C1.7	3.3903	3.0448	2.8855	0.3748	0.4285	1.0000	1.0000	1.0000	0.3466
C1.8	3.9836	3.0448	2.8855	0.4285	0.8995	4.9390	1.0000	1.0000	2.8845
C1.9	4.2035	3.2129	4.6806	0.8995	0.3467	2.8855	2.8855	0.3467	1.0000

,

Table A3. Aggregated PCM for the sub-criteria of safety based on the experts’ selection when applying the defuzzification.

	C2.1	C2.2	C2.3	C2.4	C2.5
C2.1	1.0000	0.7277	0.4763	0.8090	1.0000
C2.2	1.3742	1.0000	0.4763	1.0000	0.5287
C2.3	2.0997	2.0997	1.0000	1.3740	1.3741
C2.4	1.2361	1.0000	0.7278	1.0000	0.8995
C2.5	1.0000	1.8914	0.7277	1.1118	1.0000

,

Table A4. Aggregated PCM for the sub-criteria of hazards risk based on the experts’ selection when applying the defuzzification.

	C3.1	C3.2	C3.3	C3.4	C3.5
C3.1	1.0000	1.8907	0.5423	3.2156	3.2156
C3.2	0.5289	1.0000	0.4500	1.5277	2.2149
C3.3	1.8438	2.2224	1.0000	1.9922	1.9922
C3.4	0.3110	0.6546	0.5019	1.0000	0.2791
C3.5	0.3110	0.4515	0.5019	3.5830	1.0000

,

Table A5. Aggregated PCM for the sub-criteria of legal risk based on the experts’ selection when applying the defuzzification.

	C4.1	C4.2	C4.3	C4.4	C4.5
C4.1	1.0000	1.6984	1.1117	1.5277	3.0476
C4.2	0.5888	1.0000	1.0000	0.2950	1.5277
C4.3	0.8995	1.0000	1.0000	0.2950	1.8882
C4.4	0.6546	3.3903	3.3903	1.0000	1.0000
C4.5	0.3281	0.6546	0.5296	1.0000	1.0000

,

Table A6. Aggregated PCM for the sub-criteria of energy based on the experts’ selection when applying the defuzzification.

	C5.1	C5.2	C5.3	C5.4
C5.1	1.0000	1.1117	1.1117	0.3988
C5.2	0.8995	1.0000	1.0000	1.3760
C5.3	0.8995	1.0000	1.0000	1.0000
C5.4	2.5077	0.7268	1.0000	1.0000

,

Table A7. Aggregated PCM for the sub-criteria of environmetal impact and utilization based on the experts’ selection when applying the defuzzification.

	C6.1	C6.2	C6.3	C6.4
C6.1	1.0000	1.0000	2.3338	4.4975
C6.2	1.0000	1.0000	2.5946	3.8289
C6.3	0.4285	0.3854	1.0000	2.8845
C6.4	0.2223	0.2612	0.3467	1.0000

,

Table A8. Aggregated PCM for the sub-criteria of transportation infrastructure based on the experts’ selection when applying the defuzzification.

	C7.1	C7.2	C7.3	C7.4	C7.5	C7.6	C7.7	C7.8
C7.1	1.0000	1.0000	0.8524	1.0000	0.2795	0.3284	0.2025	0.3466
C7.2	1.0000	1.0000	1.0000	0.3466	0.2649	0.2379	0.2795	0.2950
C7.3	1.1732	1.0000	1.0000	0.3466	0.3466	0.3112	0.2950	0.4500
C7.4	1.0000	2.8855	2.8855	1.0000	0.2025	0.3284	0.3466	0.8524
C7.5	3.5775	3.7751	2.8855	4.9390	1.0000	0.3466	0.3466	0.3466
C7.6	3.0448	4.2035	3.2129	3.0448	2.8855	1.0000	1.0000	1.0000
C7.7	4.9390	3.5775	3.3903	2.8855	2.8855	1.0000	1.0000	1.0000
C7.8	2.8855	3.3903	2.2224	1.1732	2.8855	1.0000	1.0000	1.0000

,

Table A9. Aggregated PCM for the sub-criteria of Information technology infrastructure based on the experts’ selection when applying the defuzzification.

	C8.1	C8.2	C8.3	C8.4	C8.5	C8.6	C8.7	C8.8
C8.1	1.0000	1.0000	1.0000	1.0000	1.0000	1.8882	2.3338	1.0000
C8.2	1.0000	1.0000	1.0000	1.0000	1.0000	1.0000	1.0000	1.0000
C8.3	1.0000	1.0000	1.0000	1.0000	1.0000	1.0000	1.0000	1.5277
C8.4	1.0000	1.0000	1.0000	1.0000	0.5579	1.0000	1.0000	1.6095
C8.5	1.0000	1.0000	1.0000	1.7925	1.0000	1.0000	1.0000	1.0000
C8.6	0.5296	1.0000	1.0000	1.0000	1.0000	1.0000	2.8845	2.3338
C8.7	0.4285	1.0000	1.0000	1.0000	1.0000	0.3467	1.0000	1.0000
C8.8	1.0000	1.0000	0.6546	0.6213	1.0000	0.4285	1.0000	1.0000

,

Table A10. Aggregated PCM for the sub-criteria of process/operation based on the experts’ selection when applying the defuzzification.

	C9.1	C9.2	C9.3	C9.4	C9.5	C9.6	C9.7	C9.8
C9.1	1.0000	1.4498	2.7376	2.5946	1.6984	1.6984	2.2498	1.3741
C9.2	0.6897	1.0000	2.3338	2.3338	1.1117	2.5946	1.3741	1.1318
C9.3	0.3653	0.4285	1.0000	1.0000	0.3859	2.2498	1.1117	0.4214
C9.4	0.3854	0.4285	1.0000	1.0000	0.1745	1.0000	1.0000	2.0992
C9.5	0.5888	0.8995	2.5913	5.7318	1.0000	1.0000	1.0000	1.0000
C9.6	0.5888	0.3854	0.4445	1.0000	1.0000	1.0000	2.8845	2.8845
C9.7	0.4445	0.7277	0.8995	1.0000	1.0000	0.3467	1.0000	1.0000
C9.8	0.7277	0.8836	2.3728	0.4764	1.0000	0.3467	1.0000	1.0000

and

Table A11. Aggregated PCM for the sub-criteria of social based on the experts’ selection when applying the defuzzification.

	C10.1	C10.2	C10.3	C10.4	C10.5	C10.6	C10.7	C10.8	C10.9	C10.10
C10.1	1.0000	0.2506	0.3853	0.3103	0.3853	0.3466	0.4763	0.4284	0.47627	0.4284
C10.2	3.9896	1.0000	1.0000	0.5295	0.2787	0.8995	0.8090	0.6545	1.0000	0.8994
C10.3	2.5953	1.0000	1.0000	0.8995	1.1117	1.0000	1.0000	0.8995	0.65451	0.7277
C10.4	3.2228	1.8886	1.1118	1.0000	1.2360	1.0000	1.0000	1.0000	0.52951	1.0000
C10.5	2.5953	3.5885	0.8995	0.8091	1.0000	0.3466	0.3466	1.0000	1.0000	1.2360
C10.6	2.8855	1.1118	1.0000	1.0000	2.8855	1.0000	0.2025	0.2025	1.0000	0.7277
C10.7	2.0997	1.2361	1.0000	1.0000	2.8855	4.9390	1.0000	0.3853	0.42838	1.0000
C10.8	2.3344	1.5279	1.1118	1.0000	1.0000	4.9390	2.5953	1.0000	0.38531	1.0000
C10.9	2.0997	1.0000	1.5279	1.8886	1.0000	1.0000	2.3344	2.5953	1.0000	0.5295
C10.10	2.3344	1.1118	1.3742	1.0000	0.8091	1.3742	1.0000	1.0000	1.88855	1.0000

in the Appendix A denote the aggregated PCMs for the main criteria and sub-criteria associated with the experts’ selection when applying the defuzzification method in the sixth step of the procedure.

The calculated weights are denoted in

Table 3. Obtained weights from the introduced fuzzy-based AHP technique.

Main Criterion	Calculated Weights for Each Criterion	Sub-Criterion	Calculated Local Weights	Calculated Global Weights
C1	0.1542
		C1.1.	0.0410	0.0063
		C1.2.	0.0411	0.0063
		C1.3.	0.0460	0.0071
		C1.4.	0.1967	0.0303
		C1.5.	0.1743	0.0269
		C1.6.	0.0864	0.0133
		C1.7.	0.0992	0.0153
		C1.8.	0.1683	0.0260
		C1.9.	0.1470	0.0227
C2	0.1470
		C2.1.	0.1503	0.0221
		C2.2.	0.1568	0.0231
		C2.3.	0.2961	0.0435
		C2.4.	0.1858	0.0273
		C2.5.	0.2110	0.0310
C3	0.1447
		C3.1.	0.2885	0.0417
		C3.2.	0.1723	0.0249
		C3.3.	0.3143	0.0455
		C3.4.	0.0883	0.0128
		C3.5.	0.1366	0.0198
C4	0.0942
		C4.1.	0.2899	0.0273
		C4.2.	0.1440	0.0136
		C4.3.	0.1635	0.0154
		C4.4.	0.2811	0.0265
		C4.5.	0.1215	0.0114
C5	0.1093
		C5.1.	0.2080	0.0227
		C5.2.	0.2618	0.0287
		C5.3.	0.2417	0.0264
		C5.4.	0.2884	0.0315
C6	0.0617
		C6.1.	0.3763	0.0232
		C6.2.	0.3712	0.0229
		C6.3.	0.1737	0.0107
		C6.4.	0.0788	0.0049
C7	0.0791
		C7.1.	0.0545	0.0043
		C7.2.	0.0474	0.0037
		C7.3.	0.0549	0.0043
		C7.4.	0.0833	0.0066
		C7.5.	0.1355	0.0107
		C7.6.	0.2182	0.0173
		C7.7.	0.2273	0.0180
		C7.8.	0.1789	0.0142
C8	0.1010
		C8.1.	0.1489	0.0150
		C8.2.	0.1237	0.0125
		C8.3.	0.1304	0.0132
		C8.4.	0.1220	0.0123
		C8.5.	0.1331	0.0134
		C8.6.	0.1450	0.0146
		C8.7.	0.0975	0.0099
		C8.8.	0.0994	0.0101
C9	0.0660
		C9.1.	0.2080	0.0137
		C9.2.	0.1682	0.0111
		C9.3.	0.0838	0.0055
		C9.4.	0.0832	0.0055
		C9.5.	0.1530	0.0101
		C9.6.	0.1156	0.0076
		C9.7.	0.0887	0.0059
		C9.8.	0.0995	0.0066
C10	0.0428
		C10.1.	0.0401	0.0017
		C10.2.	0.0833	0.0036
		C10.3.	0.0967	0.0041
		C10.4.	0.1110	0.0048
		C10.5.	0.0957	0.0041
		C10.6.	0.0841	0.0036
		C10.7.	0.1146	0.0049
		C10.8.	0.1288	0.0055
		C10.9.	0.1287	0.0055
		C10.10.	0.1170	0.0050

for the main criteria and sub-criteria when applying steps 3–11. The economic criterion is the maximum weight among the ten main criteria. Inflation and higher costs threaten the operations of the logistics network, so the economic criterion is significant, as previously reported in the literature [4]. The safety criterion is the second highest weight in Table 3. The importance of the safety criterion was presented in the published reports by [53,54]. So, the experts’ opinions also confirm its importance. Next, hazards risk is the third highest weight, and its weight is 0.1447. The analysis of the hazards risk criterion is crucial to logistics networks. Thus, the finding also verifies the previous studies [55,61]. Then, the weight of energy is 0.1093, and energy is the fourth most important weight. Further, the fifth important weight in Table 3 is information technology infrastructure. Indeed, the information technology infrastructure is crucial to obtaining the high efficiency of logistics networks. After, the legal risk is the sixth highest weight, and its weight is 0.0942. Next, the seventh significance criterion is the transportation infrastructure based on the fuzzy-based AHP results. Although the transportation infrastructure has the seventh highest weight, the importance of weight is not low. In addition, it is essential to logistics operations. The eighth-highest weight is the process/operations criterion, and its weight is 0.0660. Then, environmental impact and utilization and social criteria are the ninth and tenth importance weights, respectively.

Table 3 denotes the calculated local and global weight of each sub-criterion. Investment is the most significant sub-criterion of the economic criterion. Then, the mortality caused by traffic injury is the highest weight among the sub-criteria of safety. It could be expected that traffic safety is an important sub-criterion of safety. Next, the epidemic is the most critical sub-criterion of hazards risk. In addition, the earthquake is also another important sub-criterion, and it should not be ignored due to the possible severe effects of logistics networks. Further, judicial independence is the highest sub-criterion of legal risk. In addition, land-administration quality is notable for legal risk as well. Then, each sub-criterion weight of the energy is obtained almost uniformly. Also, electricity generation is the most significant sub-criteria of the energy criterion. Air pollution is an important concern due to environmental awareness. So, air population is the highest sub-criterion weight of environmental impact and utilization. Further, the green number of vehicles and the quality of railroad infrastructure are the top two sub-criteria of transportation infrastructure. Next, secure servers, data corruption and network security are the top three sub-criteria of information technology infrastructure. Afterward, the lack of tech skills is the highest weight for process/operation. Lastly, a successful partnership among partners and business ethics are the top two sub-criteria of the social criterion.

4.3. Ranking Cities

The 30 metropolitan cities in Türkiye are listed in this section by applying the method stated in the previous section.

Table 4. Ranking cities regarding various scenarios.

	All Criteria		C1		C2		C3		C4		C5		C6		C7		C8		C9		C10
Cities	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank	OPS	Rank
Adana	0.276	13	0.432	23	0.867	21	0.871	23	0.145	26	0.475	3	0.804	1	0.318	4	0.094	12	0.819	14	0.203	7
Ankara	0.342	3	0.705	1	0.659	29	0.632	29	0.307	2	0.298	7	0.565	27	0.485	2	0.276	3	0.840	9	0.444	2
Antalya	0.274	20	0.467	3	0.798	25	0.821	25	0.221	3	0.183	16	0.798	2	0.246	6	0.117	10	0.553	23	0.215	6
Aydın	0.273	21	0.441	8	0.939	10	0.948	8	0.186	11	0.233	10	0.718	8	0.109	20	0.066	27	0.842	8	0.156	14
Balıkesir	0.278	11	0.433	22	0.908	18	0.934	13	0.187	6	0.459	4	0.756	5	0.163	11	0.063	29	0.828	10	0.202	9
Bursa	0.310	5	0.433	21	0.748	27	0.752	27	0.171	21	0.249	8	0.605	25	0.165	10	0.231	4	0.845	4	0.224	5
Denizli	0.269	27	0.439	10	0.924	14	0.931	14	0.186	12	0.152	19	0.695	14	0.108	23	0.087	16	0.843	7	0.156	13
Diyarbakır	0.270	24	0.439	9	0.937	11	0.918	17	0.182	16	0.150	20	0.710	10	0.100	24	0.071	26	0.560	22	0.089	26
Erzurum	0.274	19	0.435	16	0.977	2	0.984	2	0.185	14	0.049	26	0.571	26	0.110	18	0.074	24	0.374	27	0.147	21
Eskişehir	0.269	28	0.448	5	0.941	9	0.939	11	0.185	13	0.050	25	0.771	4	0.236	7	0.064	28	0.845	5	0.153	17
Gaziantep	0.306	6	0.437	12	0.916	16	0.915	18	0.116	30	0.117	21	0.671	21	0.167	9	0.180	7	0.808	15	0.126	23
Hatay	0.270	25	0.415	28	0.924	13	0.927	15	0.119	29	0.323	6	0.516	29	0.074	26	0.061	30	0.376	26	0.074	27
Istanbul	0.602	1	0.624	2	0.000	30	0.000	30	0.805	1	0.526	2	0.388	30	0.746	1	0.657	1	0.639	17	0.964	1
Izmir	0.315	4	0.467	4	0.661	28	0.667	28	0.198	4	0.637	1	0.674	20	0.430	3	0.226	6	0.725	16	0.306	3
Kahramanmaraş	0.271	23	0.431	24	0.944	8	0.946	9	0.164	25	0.246	9	0.679	18	0.096	25	0.072	25	0.365	28	0.033	29
Kayseri	0.268	29	0.437	11	0.890	19	0.877	20	0.170	22	0.189	15	0.635	24	0.139	15	0.078	22	0.821	12	0.162	10
Kocaeli	0.306	7	0.375	29	0.775	26	0.770	26	0.178	19	0.206	13	0.683	16	0.146	14	0.231	5	0.851	2	0.203	8
Konya	0.411	2	0.444	7	0.816	24	0.846	24	0.165	24	0.170	17	0.534	28	0.306	5	0.374	2	0.828	11	0.224	4
Malatya	0.278	10	0.434	20	0.973	5	0.982	3	0.179	18	0.022	30	0.707	13	0.109	22	0.083	20	0.597	21	0.149	20
Manisa	0.274	18	0.446	6	0.865	22	0.875	21	0.186	10	0.383	5	0.663	23	0.159	12	0.102	11	0.843	6	0.161	11
Mardin	0.277	12	0.364	30	0.997	1	0.995	1	0.169	23	0.079	24	0.714	9	0.069	27	0.087	15	0.360	30	0.059	28
Mersin	0.294	9	0.430	25	0.857	23	0.872	22	0.138	27	0.110	23	0.789	3	0.110	19	0.167	8	0.821	13	0.125	24
Muğla	0.275	15	0.436	15	0.914	17	0.939	12	0.186	9	0.218	11	0.707	12	0.153	13	0.090	13	0.604	19	0.149	18
Ordu	0.272	22	0.434	18	0.973	6	0.962	6	0.179	17	0.024	29	0.693	15	0.038	30	0.083	21	0.392	24	0.146	22
Sakarya	0.299	8	0.425	26	0.945	7	0.941	10	0.173	20	0.161	18	0.682	17	0.124	16	0.164	9	0.850	3	0.153	16
Samsun	0.270	26	0.435	17	0.921	15	0.926	16	0.186	8	0.217	12	0.710	11	0.110	17	0.075	23	0.619	18	0.158	12
Şanlıurfa	0.274	17	0.436	14	0.926	12	0.953	7	0.137	28	0.201	14	0.676	19	0.109	21	0.083	19	0.362	29	0.113	25
Tekirdağ	0.261	30	0.424	27	0.889	20	0.889	19	0.185	15	0.111	22	0.747	6	0.173	8	0.088	14	0.851	1	0.154	15
Trabzon	0.275	14	0.434	19	0.973	4	0.979	4	0.187	7	0.032	27	0.723	7	0.054	29	0.085	17	0.392	25	0.149	19
Van	0.275	16	0.436	13	0.974	3	0.971	5	0.197	5	0.027	28	0.667	22	0.065	28	0.084	18	0.601	20	0.018	30

presents the application results of the proposed integrated method. While the first column corresponds to the metropolitan cities in the country, the first row presents the results of the proposed method according to all criteria and each criterion separately. In other words, the second and third columns present the results of all indicators; the rest columns correspond to the results of each criterion separately.

With an overall OPS of 0.602 out of a maximum 1, Istanbul is one of the most favorable environments for logistics centers based on all criteria. Istanbul ranks highest on legal risk, transportation infrastructure, information technology infrastructure and social pillars. On the other hand, it performed the lowest performance on the safety, hazards risk and environmental impact and utilization main pillars. It is followed by Konya (score of 0.411) and Ankara (score of 0.342). On the contrary, Tekirdağ ranks lowest with a score of 0.261. Kayseri (score of 0.268) and Eskişehir (score of 0.269) are computed as the least conducive environments.

Considering the economic criterion, Ankara ranks first (score of 0.705), followed by Istanbul (score of 0.624) and Antalya (score of 0.467), respectively. The main reason for the fruitful economic environment of these cities is that the sub-criteria of domestic material consumption, investment volume, maintenance costs and number of businesses have better values. Moreover, Mardin, Hatay and Kocaeli rank lowest performance in terms of economic criterion due to higher energy consumption, higher disposal costs of hazardous wastes and lower number of enterprises.

Regarding the safety criterion, Mardin stands out as one of the safer environments, with an overall OPS score of 0.997. It is followed by Erzurum (score of 0.977) and Van (score of 0.974). At the other end of the spectrum Istanbul (score of 0.0), Ankara (score of 0.659) and Izmir (score of 0.661) are among the least safe locations in terms of occupational safety and accidents.

Hazards risk is the lowest in Mardin (0.995), Erzurum (0.984) and Malatya (0.971) and highest in Istanbul (0.0), Ankara (0.632) and Izmir (0.667). While Mardin shows the best performance in all sub-categories due to its location and characteristics, Istanbul has the lowest performance in almost all sub-criteria.

The Legal Risk is the lowest in Istanbul (0.805), Ankara (0.307) and Antalya (0.221) and highest in Gaziantep (0.116), Hatay (0.119) and Şanlıurfa (0.138). Istanbul performs well in almost all sub-criteria.

With an overall score of 0.637 regarding the Energy criterion, Izmir ranks first out of 30 metropolitan cities. It is followed by Istanbul (0.526) and Adana (0.475). On the other hand, Malatya ranks lowest with an OPS score of 0.022. Van (0.027) and Ordu (0.024) are followed as the least efficient energy cities. The reasons for the low performance of cities are the low share of renewable energy and electricity production.

Regarding the environmental impact and utilization criterion, Adana ranks first with an overall score of 0.804. It is followed by Antalya (0.798) and Mersin (0.789). Low air pollution and less population exposure to pollution have increased Adana’s environmental scores. At the other end of the spectrum, Istanbul (0.388), Hatay (0.516) and Konya (0.534) are among the least environmentally friendly cities. Istanbul scores the metropolitan with the highest total GHG emissions.

With a score of 0.746, considering the transportation infrastructure criterion, Istanbul ranks first out of 30 cities. Ankara and Izmir followed with scores of 0.485 and 0.430, respectively. Istanbul has the highest scores on air quality of transport infrastructure, green number of vehicles and number of energy/charge stations. On the other hand, Ordu (0.038), Trabzon (0.054) and Van (0.065) are among the least developed locations in terms of transport infrastructure quality.

Regarding 30 metropolitan cities, Istanbul tops the charts with a score of 0.657. Konya and Ankara tailed with scores of 0.374 and 0.276, respectively. Istanbul performs the best scores on secure servers, broadband subscriptions and network security sub-criteria. With a score of 0.061, Hatay stands out last among 30 metropolitan cities in terms of IT infrastructure, tailed by Eskişehir (0.064) and Balıkesir (0.063).

Regarding the 30 metropolitan cities with a process/operation capabilities score of 0.851, Kocaeli and Tekirdağ rank first. Considering the industrial investments and companies in these cities, the technical skills of the existing workforce, the output of existing processes and rapid response to changing consumer preferences are the biggest factors in these cities getting high scores. It is closely followed by Sakarya (0.850) and Bursa (0.845). Mardin (0.360) ranks last among the 30 metropolitan cities, tailed by Şanlıurfa (0.362) and Kahramanmaraş (0.365).

Regarding the social criterion, Istanbul ranks first with a score of 0.964. It is followed by Ankara (0.444) and Izmir (0.306). Istanbul ranks first thanks to its high score in the following sub-criteria: rural access index, convenient access to public transport, car registrations, population manager–employee relationships, successful partnership and vendor/supplier breaches. With a score of 0.018, Van ranks last among the 30 metropolitan cities, followed by Kahramanmaraş (0.033) and Mardin (0.059).

4.4. Comparison of the Cities

This section compares the results of Kusakci et al. [42]’s Sustainable Cities Index with our proposed method. As stated earlier, their index consists of economic, social, environmental and institutional dimensions and evaluated 30 metropolitans in Türkiye.

Table 5. Comparison of the rank of 30 metropolitans.

	Kusakci et al. [42]		The Proposed Method
Cities	Score	Rank	OPS	Rank
Adana	0.535	25	0.276	13
Ankara	0.721	4	0.342	3
Antalya	1.000	1	0.274	20
Aydın	0.695	5	0.273	21
Balıkesir	0.639	13	0.278	11
Bursa	0.615	18	0.310	5
Denizli	0.668	9	0.269	27
Diyarbakır	0.517	28	0.270	24
Erzurum	0.640	12	0.274	19
Eskişehir	0.802	3	0.269	28
Gaziantep	0.610	19	0.306	6
Hatay	0.580	22	0.270	25
Istanbul	0.590	20	0.602	1
Izmir	0.642	11	0.315	4
Kahramanmaraş	0.558	24	0.271	23
Kayseri	0.620	17	0.268	29
Kocaeli	0.690	6	0.306	7
Konya	0.687	7	0.411	2
Malatya	0.627	16	0.278	10
Manisa	0.634	14	0.274	18
Mardin	0.501	29	0.277	12
Mersin	0.587	21	0.294	9
Muğla	0.921	2	0.275	15
Ordu	0.534	26	0.272	22
Sakarya	0.647	10	0.299	8
Samsun	0.563	23	0.270	26
Şanlıurfa	0.527	27	0.274	17
Tekirdağ	0.686	8	0.261	30
Trabzon	0.631	15	0.275	14
Van	0.480	30	0.275	16

presents the ranking of 30 major cities obtained by the methods used in both studies. Although Kusakci et al. [42]’s work includes sustainability-related dimensions, it is important to understand the deficiencies in the following main criteria before analyzing the order of the cities. Firstly, the environmental indicators used in their Sustainable Cities index, one of the most critical dimensions of sustainability, are statistics about the water and sewage system. Additionally, important city statistics regarding the negative impact on the environment, such as greenhouse emissions, have been ignored. Secondly, there is no indicator regarding the use and production of renewable energy resources in terms of reducing environmental impact. Thirdly, there is no dimension regarding accidents that can interrupt business continuity in the city and business life. Fifthly, most importantly, there are no risks/threads to which cities are exposed or potential dangers. Sixthly, a dimension related to the information technology infrastructure of cities and the transportation infrastructure, which Türkiye emphasizes due to its geostrategic location, has been ignored again. Seventhly, while the indicators expressed on the social side in our study reflect the workplace environment of institutions and employees, they are not mentioned in Kusakci et al. [42]’s study. Finally, operational capacity in cities and institutions has not been taken into account at the technical level.

Kusakci et al. [42] found the Sustainable Cities Index for the metropolitan cities of Türkiye and reported that Antalya, Muğla and Eskişehir are the most sustainable cities. The authors explained that the first two sustainable cities are important tourism destinations and the number of tourists coming to the cities. The UN’s Sustainable Development Goals are misguided by implying that unsustainable cities should invest and pay attention to tourism activities in order to achieve a better score. On the other hand, the authors found that Van, Mardin, Diyarbakır and Şanlıurfa less sustainable cities due to the terrorist incidents and migrations. The reasons stated in the study for the cities with low sustainability scores are not included in the proposed index, and the reports that are cited to be compatible were written more than ten years ago.

With an overall the proposed method score of 0.274, Antalya ranks 20th out of 30 metropolitan cities. Its three crucial weaknesses are safety, hazard risk and process/operation criteria. On the other hand, economic, legal risk and environmental impact and utilization are the three most vital factors. In addition, the proposed method ranks Muğla and Eskişehir as 15th and 28th, respectively. At the other end of the spectrum, Van, with a score of 0.75, ranks 16th. Energy, transportation infrastructure and social performance are its three critical weaknesses. Conversely, safety, hazard risk and legal risk are the three most vital components. Furthermore, the proposed method ranks Diyarbakır and Mardin as 24th and 12th, respectively.

5. Managerial and Policy Implications

Supply chain management is a complex operation for the production and service systems in order to operate the flow of materials, information and finance to the final customer. On the other hand, risks and disruptions negatively impact the supply chain operations. So, the potential risks should be considered in logistics operations. Along the same lines, a business-continuity plan is crucial to enhance the business’ resilience against possible risk and provide a sustainability assessment. Further, sustainable logistics operations are inevitable because of environmental awareness, several restrictions and decreasing global emissions by using environmentally friendly vehicles. For all reasons, sustainability and risk assessments are addressed to provide efficient logistics operations when Sustainable Development Goals are successfully achieved.

Economic, safety and hazards risk criteria are the top three main criteria based on the introduced fuzzy AHP, which is part of the integrated method. When designing a logistics network, the volatility of price costs, mortality caused by traffic injury and epidemics should be comprehensively investigated due to the efficiency of logistics operations. Otherwise, the risks and disruptions may affect the companies and cause them to lose their market shares. Renewable energy is also vital for reducing emissions and providing requirements for environmental concerns. Notably, the information technology infrastructure is essential to manage information flows for logistics networks. Based on the fuzzy-based AHP results, the green number of vehicles is a significant sub-criterion. So, the number of environmentally friendly vehicles should be increased to operate sustainable logistics networks. Also, this situation may lead to positive environmental effects, such as reducing air pollution and emissions.

Policy-makers may consider evaluating sustainable and risk performance pillars in order to achieve successful logistics operations. Therefore, the potential risks and disruptions may be eliminated for the producer, investors and consumers. Logistics centers in cities should minimize their exposure to hazards and risks to benefit logistics network operations. Further, based on the case study results, Istanbul is the most appropriate location for logistics centers. However, policy-makers should focus on the safety, hazards risk and environmental impact and utilization main pillars for Istanbul in order to increase OPS and eliminate negative effects.

Sustainable Development Goals (SDGs) [19] are investigated based on the results of this study. SDGs 7.3, 9.1, 9.4, 11.2 and 11.6 may be achieved for logistics network analyses. The energy efficiency enhancement rate is provided due to sustainable logistics operations. The weight of the energy criterion is obtained as 0.1093. Also, SDG 7.3 is relevant to energy efficinency [19]. So, SDG 7.3 is provided. Next, sustainable and resilient infrastructure is important based on the fuzzy-based AHP results in Table 3. In fact, the weight of the transportation infrastructure criterion is found to be 0.0791. Also, SDG 9.1 is related to the development of quality, sustainable and resilient infrastructure [19]. Hence, SDG 9.1 is reachable. Then, clean and environmentally based technologies, such as green vehicles, are recommended for logistics operations. In Table 3, the weight of the environmental impact and utilization criterion is 0.0617 and the global weight of the green number of vehicles (C7.7) is 0.0180. Moreover, SDG 9.4 is associated with clean and environmentally based technologies [19]. Therefore, SDG 9.4 is attainable. Further, road-accessible and sustainable transportation systems are evaluated in this paper. colorred SDG 11.2 is related to sustainable transport systems [19]. So, SDG 11.2 is another achievable goal. From the case study of the metropolitan cities in Türkiye, analyzing severe environmental effects is important for designing a sustainable logistics network. Also, SDG 11.6 denotes the reduction of the adverse environmental effect of cities [19]. Thus, SDG 11.6 is achieved.

6. Conclusions

Urban logistics network ensures that the supply and demand mechanisms of the market and the economy operate smoothly when providing efficient and reliable transportation. On the other hand, tangible and intangible assets in the urban logistics network face various potential threats. Disruptions that may occur in this critical logistics network can have significant impacts. Based on this awareness, a business-continuity plan presents a prevention and recovery system for dealing with potential risks and disruptions, such as natural disasters and system failures, as denoted in Table 2. So, potential risks should be considered when designing complicated sustainable logistics operations. For this particular purpose, the ten main criteria and sixty-six sub-criteria are considered based on the published papers and reports, as well as experts’ opinions for evaluating logistics network alternatives under uncertainty. Moreover, this paper comprehensively evaluates the sustainability and risk pillars, including transportation infrastructure, information technology infrastructure, process/operations and social pillars. A novel and innovative integrated procedure under uncertainty is proposed, including fuzzy-based AHP with the new linguistic scale and its operators and TOPSIS techniques.

The three highest weights of the main criteria are economic (C1), safety (C2) and hazards risk (C3), respectively, based on the results of the introduced fuzzy-based AHP method. Also, epidemics (C3.3), mortality caused by traffic injury (C2.3), earthquake (C3.1), electricity generation (C5.4), machine, equipment, or facility failure (C2.5), investment (C1.4), renewable energy production (C5.2), explosions, fires, chemical accidents (C2.4), judicial independence (C4.1) and volatility of price and cost (C1.5) are the top ten sub-criteria and the sum of their global weights is 0.2507. The thirty metropolitan cities in Türkiye are ranked using the TOPSIS method, a part of the introduced methodology when considering different scenarios. Istanbul (OPS = 0.602), Konya (OPS = 0.411) and Ankara (OPS = 0.342) are the top three choices for designing and planning logistics centers regarding sustainability and risk assessment. Comparison analyses were performed from the literature. The existing research works did not consider business-continuity plans, renewable energy sources, potential risks, information infrastructure and social criteria. Therefore, this study provides a comprehensive analysis of the ten main criteria, including sustainability and risk assessment.

Managerial and policy implications are summarized for policy-makers as follows. Evaluating the sustainable and risk performance pillars is crucial to prevent and recover potential risks and disruptions for successful logistics networks when dealing with business-continuity plans. Then, hazards and risks should be minimized to take advantage of logistics operations. In addition, the number of environmentally friendly vehicles should be increased in urban logistics networks. So, reductions in emissions and air pollution are achieved. Further, SDGs 7.3, 9.1, 9.4, 11.2. and 11.6 are able to be achieved based on the results of this study.

There could be a few limitations to this study. TFNs are selected to perform fuzzy arithmetic in the proposed procedure. In addition, linguistic scales and their associated fuzzy number scales are based on TFNs. Although TOPSIS is used as a powerful and appropriate technique to rank the alternatives, other techniques can be utilized. Therefore, there could be possible extensions of this study. The unusual fuzzy sets and operators could be applied to the AHP technique to obtain the weights of the main criteria and sub-criteria under uncertainty. Further, the alternatives could be ranked by other MCDM techniques to choose the center of urban logistics operations. Next, the proposed methodology, including main criteria and sub-criteria, could be applied to other countries to assess sustainability and risk indicators in urban logistics network analyses.