Abstract
Since the establishment of the “Belt and Road” initiative, the investment and construction of ports along the 21st Century Maritime Silk Road have received extensive attention from the international community. The evaluation of ports is of great significance to investors’ investments and construction of ports around the world, so it is very necessary to establish a reasonable port evaluation system. At present, there are few studies on defining and evaluating port resilience, and the existing port evaluation index system has defects. Therefore, according to the similarity between cities and ports, this paper introduces the concept of “three-dimensional space” and the “system of systems” theory of cities and divides the resilience of ports along the Maritime Silk Road into three-dimensional spaces of “physical-society-information.” The CRITIC-entropy method and the TOPSIS method constructed a port resilience evaluation model along the Maritime Silk Road and quantitatively evaluated and analyzed the comprehensive resilience and subspatial resilience of 28 ports along the 21st Century Maritime Silk Road. The results show that the route network port degree, the annual throughput of the port container, and the number of fixed broadband subscribers per 100 people are the key indicators that affect the port’s physical space resilience, social space resilience, and information space resilience. Also, coordinated physical, social, and information spatial resilience development plays a catalytic role in improving overall resilience. Therefore, the investment of ports along the Maritime Silk Road should adopt corresponding and more targeted investment plans according to the actual resilience of each port. The research provides new ideas and directions for investors to invest in port construction and has certain practical guiding significance for the increase of investors’ income and the sound development of the national economy.
1. Introduction
China proposed the “Belt and Road” initiative in 2013. The “21st Century Maritime Silk Road” was a new strategic concept for trade cooperation that Chinese president came up with during his visit to Indonesia [1]. This has aroused a lot concern and heated debate in the international community. As for China, building the Maritime Silk Road in the new era is one way to deepen strategic cooperation with ASEAN countries and protect Chinese foreign trade. It is also a way to use critical ports along the route as hubs to build safety and efficiency together. It is a major transportation channel that connects the economic sectors of Southeast Asia, South Asia, North Africa, and Europe so as to achieve win-win cooperation among regions. As of 2021, many Chinese enterprises were actively “going out” by investing in ports along the Maritime Silk Road or directly participating in the operation and management of ports. For example, under the operation of COSCO Shipping, the port of Piraeus has developed from the brink of bankruptcy to a port that helped Athens to be rated as the “eighth largest shipping center.” In this context, how to evaluate the status quo of ports along the Maritime Silk Road that have participated in investment, construction, and operation and how to choose ports for foreign investment and cooperation in the future are important, and this needs to keep pace with the times to define the port’s value measurement.
The outbreak of COVID-19 at the end of 2019 and its raging epidemic have led to chaos in the shipping supply chains around the world, triggering many scholars to think about vulnerability and resilience. Vulnerability first appeared in the field of geography, and now it has been widely used in ecology, public health, and disaster management [2]. In the transportation field, it means how the attack or partial failure of a system affects the transportation function of the whole system [3]. Resilience can also be called elasticity. Klein et al. [4] believed that resilience is applied to the determination of specific system properties, that is, the ability of the system to absorb disturbances, maintain or restore the original state of the system, and self-organize under the condition of external disturbances. In recent years, there has been more and more talk about how vulnerable and strong resilience of maritime supply chains is. Ports are considered an indispensable part of the maritime supply chain because they are important nodes. Because of this, resilience has become an important way to measure the value of a port. In the context of the rapid growth of ports, specific evaluation recommendations can help investors make the right choices.
In order to find out the essence of this field, this paper constructs a port resilience evaluation index system and model along the Maritime Silk Road. First, according to the similarity between cities and ports, the city’s “three-dimensional space” concept and “system of systems” theory are introduced, and the resilience of ports along the Maritime Silk Road is divided into “physical-society-information” three-dimensional spaces. Second, based on the CRITIC-entropy method and the TOPSIS method, a resilience evaluation model of ports along the Maritime Silk Road is constructed. Third, the comprehensive and subspatial resilience of 28 ports in 5 regions along the 21st Century Maritime Silk Road is quantitatively evaluated, and horizontal and vertical analyses are made. Finally, according to the evaluation results of the resilience of ports along the Maritime Silk Road, we provide corresponding reference suggestions for investors.
The main contributions of this research are as follows. First, it effectively solves the problem of lack of political, medical, information, and other factors in previous research, and it also transplants the concept of “three-dimensional space” and the “system of systems” theory of cities; the resilience of ports along the Maritime Silk Road is divided into three perspectives: physical, social, and information, and the Global Peace Index, HAQ Index, and other factors are included in the assessment. Second, comprehensively considering the conditions of each index in the evaluation index system, a port resilience evaluation model based on the CRITIC-entropy method and the TOPSIS method is constructed. In the sample selection stage, factors such as the area to which the port belongs and whether the port has been planned and invested are also considered, and the resilience of multiple ports and regions along the Maritime Silk Road is quantified.
The rest of this paper is organized as follows. Section 2 gives the literature review. Section 3 includes the methodology and data. Section 4 gives the result analysis. Section 5 gives the conclusions.
2. Literature Review
Resilience theory first existed in the field of mechanics and was first put forward by Holling and others in the field of ecology. It is considered to be the ability of a system to restore variables to their original states after absorbing changes in variable states. In the field of sociology, resilience refers to the ability of society to resist and transfer risks under potential dangers so as to maintain its original functions and states. Hudeca et al. [5] believed that resilience was the ability of a system to maintain a constant output level when perturbed. Wu [6] made a systematic discussion of resilience theory, describing resilience as the ability of a system to withstand, resist, and respond actively to external disturbances so as to quickly restore itself to its original state or reestablish a new balance.
About the methodologies and models used in this paper, they are also found in many other studies related to port and maritime transport. Wen and Chen [7] established the evaluation index system based on the DPSIR model and determined the index weight based on entropy weight, which provided a new way for the modernization of construction and shipping management of inland ports. Lu et al. [8] built a combination model of clustering and TOPSIS analysis and used the model to evaluate the soft strength of coastal ports. Kim [9] compared the port competitiveness of China and South Korea through TOPSIS evaluation. Nguyen and Woo [10] used the TOPSIS method and K-mean cluster analysis to evaluate the competitiveness of Southeast Asian container ports. Md Hanafiah et al. [11] used AHP and TOPSIS techniques to control maritime accidents at the Strait of Malacca.
In the context of the raging epidemic and frequent disasters in various regions around the world, research on resilience has mostly focused on security resilience and urban resilience in in recent years. In terms of security resilience, Chen and Cha [12] and others discussed urban design based on the perspective of security resilience, taking the Changsha Convention and Exhibition Area as an example. Researchers at the US Center for Earthquake Research had developed the R4 Resilience Framework, which includes redundancy, responsiveness, alertness, and robustness, to provide concrete ideas for resilience assessment. In terms of urban resilience, Shen et al. [13] and Luo et al. [14] conducted research on the construction of an urban community resilience governance model and the evaluation of urban community economic resilience under health emergencies. Based on the resilience curve, Li et al. [15] proposed a quantitative analysis framework for urban security resilience, established a virtual city model including multiple subsystems, and used the Monte Carlo method to quantitatively study the security resilience of the city.
Resilience-related theories are also involved in the shipping field, most of which are studies on the risk resistance and resilience of shipping networks. Baroud et al. [16] established a metric for the resilience of network infrastructure systems in the industry, which is the ratio of the value of the network after it is destroyed to its value before it is destroyed. Mansouri et al. [17] developed a risk management decision analysis (RMDA) framework on the basis of defining common fundamental elements of port infrastructure system (PIS) resilience.
At present, the related research on resilience theory is more distributed in the fields of ecological resilience, security resilience, and urban resilience, and there is less research related to the resilience of transportation and supply chains, not to mention the resilience of ports along the Maritime Silk Road. Secondly, the current research on port resilience along the Maritime Silk Road mostly focuses on the container shipping network, involving ports, routes, ships, and other aspects. The system resilience is mainly evaluated from the structure and performance of the network, and few studies focus on the resilience of ports along the Maritime Silk Road. At the same time, most of the existing research on the resilience of ports along the Maritime Silk Road is based on a literature review to determine the factors and indicators that affect the shipping network, and there are few quantitative studies. Even though some studies have adopted quantitative methods, they have not achieved a multifaceted and systematic evaluation of ports along the Maritime Silk Road. They simply regard ports as transport infrastructure or places where economic activities take place, and because more factors such as politics, medical treatment, and information technology are not included in the assessment, it is also unable adapt to the various changes and challenges brought about by the general trend of COVID-19, informatization, and the joint development of port and city.
3. Methodology and Data
3.1. Methodology
3.1.1. Principle of Selection of Port Resilience Evaluation Index
This paper draws on the concept of “three-dimensional space” for cities and selects port resilience evaluation indicators from three spatial dimensions of physics, society, and information. The index system that is formed must conform to the theory of “system of systems” and the characteristics of port resilience.
Before learning about the concept of “three-dimensional space” in the city, it is necessary to clarify its specific connotation first. The so-called “three-dimensional space” means that a city is not a simple collection of buildings but contains important components such as human activities and information circulation, so it can be regarded as a “three-dimensional space” coupled by physics, society, and information [18], and each system in urban development and construction can find corresponding projections in these three spaces. Among them, physical space is the most intuitive, followed by infrastructure, natural environment, etc., which can also be classified into this category. Social space is developed on the basis of physical space, and people in the city and human activities such as politics, economy, medical care, culture, and education are important part of it. With the development of society and the progress of science and technology, information has gradually become an indispensable part of the urbanization process, and the information space has been created. This space not only includes face-to-face information communication and exchange among urban residents but also builds bridges between physical space, social space, and information space. Through digital and information technologies such as big data, artificial intelligence, and the Internet, information dissemination and exchange between different spaces and within the same space can be realized. From the content of the three-dimensional space of “physics-society-information,” it can be seen that these three do not exist in isolation but are complex and highly interrelated. The progress or lag of a single space will have a linked impact on other spaces.
As an important urban issue, urban resilience needs to be comprehensively considered from the three-dimensional space of “physics-society-information.” Since the concept of “three-dimensional space” was put forward, many scholars have discussed urban resilience based on it and have also established an urban resilience evaluation index system. On the basis of the concept of “three-dimensional space,” Fang et al. [19] proposed that a city is a “system of systems” composed of multiple subsystems in a three-dimensional space; that is, various services provided by a city can be attributed to the corresponding subsystems, each of which is independent and interrelated and can find projections in three-dimensional space. After that, some scholars built an urban resilience evaluation index system with the concept of “three-dimensional space” as the target layer and resilience characteristics as the criterion layer. Also, some indicators at the level of social space and information space are of reference significance for the construction of the port resilience indicator system, such as the per capita disposable income, the GDP growth rate, the proportion of tertiary industry in GDP, the measurement of medical and social security that reflect the level of economic development in social space, the number of doctors per 10,000 people of development level, and the number of Internet broadband users per 10,000 people, which reflect the speed of information dissemination in the information space.
Existing studies have not clearly defined the characteristics of port resilience; therefore, this paper proposes the characteristics of port resilience by combining the characteristics of resilience and the resilience of international shipping lanes.①Stability: it refers to the ability of the port to withstand external disturbances, resist disturbances, and maintain normal operation of the port.②Resiliency: when external disturbances exceed the port’s tolerance range, causing some functions of the port to be affected and operations to be blocked, the port can quickly identify risks, formulate emergency plans, and implement them, so as to quickly restore and maintain the normal operation of the port.③Redundancy: in order to ensure the solution of the problem, the port usually formulates multiple alternatives.④Adaptability: It is the ability to return to its original structure and function after a shock. But usually, after a port is hit, not only will it try to restore its original capacity, but it will also learn the lessons of this crisis and prepare for similar events that may occur in the future.
3.1.2. Construction of Port Resilience Evaluation Index System
Comprehensively considering the influencing factors of port resilience along the Maritime Silk Road, based on the three-dimensional concept of “physics-society-information” and the characteristics of port resilience as the selection principle, and referring to relevant literature [20], a resilience evaluation of ports along the Maritime Silk Road was constructed. The indicator system (Table 1) includes 3 target layers of physical space, social space, and information space and consists of 4 criteria and 23 indicators of stability, resiliency, redundancy, and adaptability. The selection of port resilience indicators must not only accurately reflect the resilience of ports but also have certain economic significance. Investors can then carry out corresponding construction and investment activities through the relevant content in the resilience indicator system. Establishing and improving a scientific, economical, and reasonable port resilience index system can promote the construction of ports’ resilience and can also provide support and assistance for investors in their construction and investment. Investors participating in the construction of improving the resilience of regional ports can improve the resilience of regional ports at the macrolevel and further promote the sustainable development of the entire city. At the microlevel, investors can also obtain certain policy dividends and low-interest support by making corresponding investments and construction activities in the construction of port resilience according to the evaluation index system.
The indicators in the port resilience evaluation system correspond to the influencing factors of port resilience, and necessary explanations are made for each indicator.①Port infrastructure x1 (maximum depth of berth): the maximum water depth of a port berth, which is used to measure the water depth resources of the port. x2 (maximum berthing capacity): it refers to the maximum deadweight tonnage class of a fully loaded ship that can berth and carry out normal operation activities such as cargo loading, unloading, and carrying passengers when the berthing is at the local designed low tide [21]. x3 (number of berths): Berths are the basic units for loading and unloading operations in ports. The number of berths refers to the actual number of berths at the end of the statistical report. x4 (container port performance index): based on the time weighting of different container volumes when five types of ships arrive at the port, it can objectively evaluate the size of ships and container cargo volume at each stop at global ports and rank the sample ports. x7 (infrastructure development index): It is a comprehensive index to measure the development environment, demand, heat, and cost of the infrastructure industry in countries along the Belt and Road. It is generally positively correlated with the country’s infrastructure industry prospects and investment attractiveness. x8 (infrastructure development demand index): It is the sum of relative demand and absolute demand in a country’s infrastructure construction field. Among them, relative demand represents the country’s infrastructure investment to ensure production and consumption under the current per capita income level, while absolute demand refers to the country’s demand for basic construction to achieve the best social services.②Shipping network status x5 (average distance connected to other ports): it refers to the minimum number of segments that two ports need to pass to realize mutual connection [22]. x6 (route network port degree): it refers to the number of routes that ports are connected to in the route network.③Port political environment x9 (Global Peace Index (GPI)): it takes 99.7% of the world’s population as a sample and uses 23 qualitative and quantitative indicators to measure the degree of peace and stability of life in 121 countries from the perspectives of social security, domestic and international conflicts, and the degree of national militarization.④Port and hinterland economy x11 (per capita GDP): The domestic GDP obtained by one country during the accounting period is proportional to the country’s household registration population, which can reflect the living standards of the people in the country. Per capita GDP is based on the current price of US dollars. x12 (annual throughput of the port container): the number of containers from the waterway entrance, exit area, and loading and unloading, with 10,000 TEU (20-foot standard container) as the unit. x15 (GDP growth rate): the annual growth rate of domestic GDP calculated at comparable prices. x16 (annual growth rate of port container throughput): the annual container throughput ratio at ports this year versus last year. x17 (proportion of added value of tertiary industry in GDP): the growth value of the tertiary industry compared to the previous year’s GDP.⑤Medical facilities and services x10 (HAQ Index): it quantifies the medical level of major countries and regions around the world. x13 (number of doctors per 1,000 people): The number of doctors who can provide services per 1,000 people where the port is located. Among them, doctors include general doctors and specialists. x14 (number of hospital beds per 1,000 people): the number of hospital beds per 1,000 people in the country or region where the port is located. x18 (proportion of recurrent health expenditure in GDP): the proportion of government public medical and health expenditures to GDP.⑥Information dissemination and technology x19 (proportion of Internet users): the degree of popularity of the country and region where the port is located. x20 (number of fixed broadband subscribers per 100 people): it includes fixed WIMAX and any other residential and organization subscriptions under any fixed wireless technology. x21 (number of cell phone service subscriptions per 100 people): it refers to subscriptions to public mobile phone services.⑦Technology and education input x22 (proportion of R&D expenditure in GDP): it can measure the importance of a country or region for scientific and technological research and development. x23 (proportion of education expenditure in total expenditure): it shows a country’s financial support and emphasis on the education industry.
3.1.3. Selection of Port Resilience Evaluation Methods
This paper selects the objective assignment method to determine the weights of each evaluation index of port resilience. Also, we choose the CRITIC-entropy method; this method is a combination of the entropy and CRITIC methods, which can highlight the characteristics of each indicator data and also take into account the dispersion of the data and reduce the extreme values.
Then, we select the TOPSIS method as a comprehensive evaluation approach; this method has no special requirements on the sample data and is suitable for the comparison of different evaluation objects, so it is very suitable for the evaluation of the resilience of ports along the Maritime Silk Road. The TOPSIS method has no special requirements for sample information and is suitable for comparison among different evaluation objects, so it fits well with the needs of port resilience evaluation along the Maritime Silk Road.
In general, the CRITIC-entropy method is used to objectively assign weights to the evaluation index system, and on this basis, the TOPSIS model is used to comprehensively evaluate the resilience of ports along the Maritime Silk Road. Finally, the resilience evaluation model of ports along the 21st Century Maritime Silk Road based on the CRITIC-entropy method and the TOPSIS method is constructed.
3.1.4. Determination of the Weights of Each Index of the Port Resilience Evaluation System
Before applying the CRITIC-entropy value method to determine the weights of each indicator of the evaluation index system, the negative indicators should first be transformed into positive indicators and subsequently transformed into pure values through data standardization. The port resilience evaluation index system shown in Table 1 has positive and negative indicators, and this paper selects Min-Max standardization (extreme difference standardization method) to preprocess the data.
Suppose m ports are to be evaluated and n evaluation indicators are given, which can form a data matrix , and find the minimum and maximum values of each indicator, which are set as and . A raw value x is mapped to the interval [0, 1] by extreme difference analysis, and the standardized value of each index is , so that . The specific steps are as follows. In the case of positive indicators, it is calculated as follows: In the case of negative indicators, it is calculated as follows: The definition is standardized with the following equation: where is the jth indicator value for the ith port after normalization and represents the jth raw data for the ith port (i = 1, 2, 3, …, m; j = 1, 2, 3, …, n). If new data are added, and may change, so recalculations are usually required.
The CRITIC weighting method is based on two concepts: contrast and conflict. Contrast refers to the standard deviation of each evaluation object of the same index, which is positively correlated with the contrast degree. Conflict occurs between different evaluation indicators, which is influenced by the correlation between indicators and has a positive relationship with the indicator weights. The specific steps of the CRITIC method are as follows.
Assuming that the contrast of the jth indicator is , the formula is as follows:
Assuming that the conflicting quantifiers of the jth indicator and the remaining indicators are , the formula is as follows:where is the Pearson correlation coefficient between the ith indicator and the jth indicator, with the formula
The amount of information contained in the jth indicator can be calculated from σj and ƒj by the following formula:
Assuming that the weight of the jth indicator is , then the weight of indicator j under the CRITIC method iswhere σj represents the contrast of the jth indicator; is the value of the jth indicator of the ith port after unstandardized treatment; is the mean of the jth indicator; is the conflict between indicators; and denote the mean of the sample data of indicators and j; and is the weight of indicator j under the CRITIC method.
The entropy method can be used to measure the degree of disorder of evaluation indicators. To determine the weight of each index using the entropy method, it is necessary to calculate the information entropy and the information utility value. Information entropy is the average amount of information after excluding redundant information, which can be set as ej. The information utility value is the difference between 1 and the information entropy value, which is positively correlated with the amount of information contained in the index. The specific steps of the entropy value method to determine the weights are as follows. The information entropy value of the jth indicator is calculated by the following formula: The information utility value of the jth indicator is derived from the calculation of the information entropy value with the following formula: Assume that the weight of the jth indicator is , and the formula is as follows: where ej is the information entropy of the jth indicator; is the information utility value of the jth indicator; is the jth indicator value of the ith port after dimensionless processing; and is the weight of the jth indicator under the entropy method.
In the case of combined CRITIC-entropy value method assignment, assuming that the weight of the final indicator j is , the calculation formula iswhere is the weight of indicator j under the CRITIC method and is the weight of the jth indicator under the entropy method.
3.1.5. Port Resilience Evaluation Model Construction Based on CRITIC-Entropy Value Method and TOPSIS Method
The determination of the weights of the indicators under the CRITIC-entropy value method marks the completion of the construction of the port resilience evaluation index system. On this basis, the TOPSIS method can be used to establish a port resilience evaluation model to realize the comprehensive evaluation and spatial evaluation of port resilience along the Maritime Silk Road. In general, the TOPSIS method requires firstly positive and standardized data processing, which is already done in the process of determining the weights using the CRITIC-entropy method above. Therefore, the positive and negative ideal solutions can be determined directly, and the distance between each evaluation object and the optimal solution can be calculated as follows.
The standardized port resilience evaluation matrix and the standardized weighted evaluation matrix are constructed:
In the matrix Z, determine the positive ideal solution and negative ideal solution and solve the distance between the ith port to be evaluated and the positive and negative ideal solutions, respectively. The Euclidean distance between the ith port to be evaluated and the positive ideal solution is
The Euclidean distance between the ith port to be evaluated and the negative ideal solution Z is
Assuming that the closeness of the ith port to be evaluated to the optimal solution is , the calculation formula iswhere represents the positive ideal solution, represents the negative ideal solution, which is the normalized value, is the distance between the ith port to be evaluated and the positive ideal solution , and is the distance between the ith port to be evaluated and the negative ideal solution .
The comprehensive resilience and subspace resilience of each port to be evaluated along the Maritime Silk Road can be reflected by its proximity to the optimal solution , thereby establishing the evaluation criteria for the resilience of ports along the Maritime Silk Road as shown in Table 2.
When the number of ports to be evaluated is m ≥ 1, this criterion can not only visually assess the combined resilience of a single port and the resilience of each space but also compare the resilience of different ports.
3.2. Data
3.2.1. Selection of Sample Ports
Considering the availability of data and the feasibility of cross-sectional comparison, 28 ports in 20 countries along the 21st Century Maritime Silk Road in China, Southeast Asia, South Asia, the Middle East and Africa, and Europe were selected (Figure 1), covering almost all the major ports along the route except the South Pacific region (see Table 3).
3.2.2. Data Source
The indicator data for each sample port in this paper are obtained from IHS Markit’s Sea-Web Port section, the Review of Marine Transport 2021 by the United Nations Conference on Trade and Development (UNCTAD), the World Bank database and the publication of The Container Port Performance Index 2020,” “The Lancet” journal article, Lloyd’s List, the official website of the China National Bureau of Statistics, the China Statistical Yearbook, China Ports Network, “Infrastructure Development Index for Belt and Road Countries (2021)” and “21st Century Maritime Silk Road Port Development Report” jointly published by the China Chamber of Commerce for Foreign Contracted Projects and China Export and Credit Insurance Corporation (edited by Zeng Qingcheng), etc. Some of the missing data were obtained from the official websites and news reports of the ports, and data preprocessing was performed when they were still unavailable, i.e., the missing values were filled with mean values.
3.2.3. Raw Data Preprocessing
After collecting the data for 23 resilience evaluation indicators from 28 sample ports separately, each piece of raw data needs to be preprocessed. First, due to the shortcomings of the statistical scope and uniform standards of some information, some data are missing, so SPSS data processing software is applied to fill the missing parts with the mean value of the existing data for this index. Second, Table 1 shows that the three indicators of container port performance index, average distance connected with other ports, and Global Peace Index are all negative indicators. Third, in order to reduce the influence caused by the different units of each index, the data of each index are dimensionless in SPSS.
3.2.4. Determination of the Weights of Each Index under the CRITIC Weighting Method and the CRITIC-Entropy Value Method
Using Stata software, the preprocessed data results are further processed according to the specific steps of the CRITIC method above, and the contrast of each indicator, the conflicting nature of the indicators, and the information content (information carrying capacity) of the indicators are calculated separately, and the weights of each evaluation indicator of the port resilience evaluation system under the CRITIC method are finally derived, as shown in Table 4.
As shown in Table 4, the maximum and minimum values of the contrast are in the GDP growth rate and the number of hospital beds per 1,000 people, which intuitively reflect the economic development and medical facilities of each port and its hinterland under the COVID-19 outbreak. As a result of COVID-19, the economic development of many countries and regions around the world suffered a strong blow, with more than 64.29% of the sample ports having a negative GDP growth trend. At the same time, COVID-19 has placed demands on the medical facilities of the ports and their hinterlands. According to IHS Markit’s Sea-Web Port segment, most sample ports are equipped with basic medical facilities. However, according to the latest data from the Lancet article, the number of beds per 1,000 people in port hinterlands varies widely, from 6.5 beds per 1,000 people in France to 0.7 beds per 1,000 people in India. In terms of the conflict, there is little correlation among the indicators due to the fact that resilience involves multiple dimensions of the ports and their hinterlands resulting in a strong conflict. However, among the weights of the indicators under the CRITIC method, the GDP growth rate has the largest weight.
Using Stata software, the weights of the indicators under the entropy method are calculated according to the calculation steps of the entropy method described in the previous section, which are combined with the weights of the indicators calculated by the CRITIC method, and the combination weights of the indicators of the port resilience evaluation system are finally obtained as shown in Table 5.
By analyzing the weights of each indicator under the CRITIC-entropy value method, the scientificity and reasonableness of the evaluation model of port resilience along the 21st Century Maritime Silk Road constructed in this paper can be initially judged. Comparing the weights of each indicator within the three-degree space is helpful in finding the key indicators in each space.
(1) Physical Space (x1–x8). Under the CRITIC method, the maximum berthing capacity (x2) of the port has the highest weight, while the two indicators of the status of the shipping network (x5: average distance connected to other ports and x6: route network port degree) are tied for second place. In the entropy method, the three indicators with the highest weights from the largest to the smallest are the route network port degree (x6), number of berths (x3), and maximum berthing capacity (x2), which are still the highest after the combination. By comparing the specific values, it can be found that the route network port degree (x6) is the key index affecting the physical space resilience of the port, which is also in line with the requirements of building a three-dimensional interconnected transportation network for the shipping industry. With the development of the shipping industry, except for the South Pacific region, most of the port infrastructure construction has tended to be perfect, including the maximum berthing capacity of the port and the number of berths, although these can still be increased through technology. In this case, the scientific layout of the port network will not only help to improve the operation of key ports and rationalize the allocation of funds and resources but also improve the resilience of ports, as evidenced by COVID-19 and the congestion in ports caused by the Ever Given.
(2) Social Space (x9–x18). GDP growth rate (x15), proportion of added value of tertiary industry in GDP (x17), and annual growth rate of port container throughput (x16) occupy the top three values of each weight under the CRITIC method, while the annual throughput of the port container (x12), GDP per capita (x11), and number of doctors per 1,000 people (x13) are the three indicators with higher entropy. The annual throughput of the port container (x12), GDP per capita (x11), and number of doctors per 1,000 people (x13) are the three indicators with greater weight under the entropy method. The annual throughput of the port container (x12), GDP per capita (x11), and GDP growth rate (x15) have the highest weights under the combined CRITIC-entropy assignment method, where the annual throughput of the port container (x12) is a key indicator of port socio-spatial resilience. In order to more comprehensively and accurately measure the resilience of ports along the Maritime Silk Road, this paper sets indicators related to regional political and military situation, port and hinterland economy, and medical care in the social space of the port resilience evaluation index system (Table 1). But no matter how the functions and services of ports are expanded, port production activities, mainly loading and unloading operations and transshipment, are always the focus of port development, and factors such as medical care and political and military situation still serve the port production activities.
(3) Information Space (x19–x23). Under the CRITIC method, the number of fixed broadband subscribers per 100 people (x20), the number of cell phone service subscriptions per 100 people (x21), and the proportion of Internet users (x19) have relatively large weights. Among the weights of each indicator under the entropy method, the proportion of R&D expenditure in GDP (x22) takes the place of the indicator of the proportion of Internet users (x19), and the weight even surpasses that of the number of cell phone service subscriptions per 100 people (x21). Finally, the number of fixed broadband subscribers per 100 people (x20), the number of cell phone service subscriptions per 100 people (x21), and the proportion of R&D expenditure in GDP (x22) are at the top three positions in the weighting of each indicator of port information space resilience, and the number of fixed broadband subscribers per 100 people (x20) becomes a key indicator of port information space resilience. Whether before or during a disaster, rapid and widespread information dissemination is particularly important, not only to enable ports to be informed of the disaster or risk situation in a timely manner so that they can respond quickly (resilience), but also to help people strengthen communication and cooperation to discuss solutions to the disaster.
In summary, the route network port degree (x6), the annual throughput of the port container (x12), and the number of fixed broadband subscribers per 100 people (x20) are the key indicators affecting the resilience of port physical space, social space, and information space, respectively, under the CRITIC-entropy method, where the route network port degree (x6) has the greatest weight, the annual throughput of the port container (x12) is the second, and the number of fixed broadband subscribers per 100 people (x20) is the third.
4. Result Analysis
4.1. Empirical Results of Port Resilience Evaluation
Based on the objective weighting method of the CRITIC-entropy value to determine the weights of the indicators of the port resilience evaluation system along the 21st Century Maritime Silk Road, the comprehensive resilience of 28 sample ports was evaluated by the TOPSIS model using Stata software, and the evaluation results are shown in Table 6.
According to the comprehensive resilience of the ports along the Maritime Silk Road shown in Table 6, Figure 2 is drawn.
Figure 2 shows that the overall resilience of the sample ports is mainly concentrated in the “C” region (0.4–0.6). The differences in the resilience of the ports in China and Europe are not significant, mainly concentrated in “C+” and “C−,” respectively. In the Middle East and Africa, although there are a few “C−,” “D+” is still the majority of the port resilience. In contrast to the Middle East and Africa region, there are two “D+” grades for port resilience in South Asia, but the number of “C−” ports is relatively high. The overall resilience of ports in Southeast Asia varies significantly, with Tanjung Priok and Singapore ports having “C+,” while the ports of Klang, Laem Chabang, and Hai Phong have “C−,” and Manila port is in “D+.”
4.2. Horizontal Analysis of Port Resilience Evaluation
Based on the proximity of each port to the optimal solution for resilience CN shown in Table 6, the ports are ranked for overall resilience as follows.
Table 7 shows that the comprehensive resilience of ports along the Maritime Silk Road has a more obvious regional heterogeneity: China > Southeast Asia > Europe > Middle East and Africa > South Asia. Among them, the comprehensive resilience rating of Chinese ports is generally high, and the comprehensive resilience rating ranks high among the 28 sample ports, such as the comprehensive resilience intensity of Ningbo Zhoushan port, which is the highest among these ports. For example, Ningbo Zhoushan port has the highest ranking among these ports, which shows that the comprehensive resilience of regional ports is relatively good. However, Manila port, Laem Chabang port, and Hai Phong port in Southeast Asia, which have lower resilience, need to get attention and promote the coordinated development of inter-regional ports to form a better shipping network and further improve the resilience of ports. The Middle East and Africa region has more ports in the “D+” resilience level, and the sample ports except Said port and the port of Aqaba are ranked low among the 28 selected ports, so the comprehensive resilience of regional ports needs to be improved. The overall regional comprehensive resilience level of South Asian ports is low, and the ports of Nehru and Colombo are ranked lower among these ports. After grasping the layout of the comprehensive resilience of major ports along the Maritime Silk Road, specific analyses can be carried out for different regions to provide a reference for investors to select ports for investment.
The comprehensive resilience and spatial resilience of ports in China are shown in Table 8. The comprehensive resilience of Ningbo Zhoushan port, Qingdao port, and Shanghai port is higher, and the results of each spatial resilience show that the physical spatial resilience of Ningbo Zhoushan port is higher, and the advantages of maximum berthing depth and maximum berthing capacity are obvious. While the three degrees of spatial resilience of Qingdao port are not the largest among the sample ports of China, it has coordinated the development of each spatial resilience. The port of Shanghai ranks third in overall resilience. Relatively speaking, Lianyungang port, Hong Kong port, and Shenzhen port have lower comprehensive resilience. By analyzing the spatial resilience, although Lianyungang port’s social space resilience and information space resilience are relatively low, its physical space resilience is better and it has great development potential as an important node at the eastern end of the New Eurasia and the New Silk Road. Hong Kong’s information space resilience is good, but the physical space and social space resilience need to be improved; especially with the continuous improvement of other Chinese port infrastructure, automation, and intelligent processes, Hong Kong’s unique natural conditions in the physical space brought about by these advantages are gradually reduced.
The overall resilience and spatial resilience of ports in Southeast Asia are shown in Table 9. From the results of spatial resilience, we can see that the comprehensive resilience of the ports of Singapore, Tanjung Priok, and Klang is higher, and the Port of Singapore is not as high as the Port of Hai Phong in terms of socio-spatial resilience, but it is relatively higher in terms of physical spatial resilience and informational spatial resilience, which makes it rank first among the sample ports in Southeast Asia and the second among these ports. Tanjung Priok port has a good spatial resilience compared to other sample ports in the region, but the difference in physical and informational spatial resilience with Singapore port is large, resulting in a big difference in the overall resilience of the port. The port of Klang ranks third among the sample ports in Southeast Asia due to its outstanding information spatial resilience among the major ports in the region. The overall resilience of Hai Phong port, Laem Chabang port, and Manila port is relatively low. Among them, Hai Phong port and Laem Chabang port have advantages in socio-spatial resilience and information spatial resilience but outstanding shortcomings in other spatial resilience. The port of Manila has a low level of spatial resilience, and its overall resilience rating is “D+.”
The overall resilience and spatial resilience of ports in South Asia are shown in Table 10. The overall resilience of Karachi port, Qasim port, and Chittagong port is high, and the results of spatial resilience show that the socio-spatial resilience of Karachi port and Qasim port is not outstanding compared with Chittagong port and the informational spatial resilience compared with Colombo port, while the physical spatial resilience is relatively high in the region, but there is still a big gap with Ningbo Zhoushan port in China and Singapore port in Southeast Asia. The social-spatial resilience of Chittagong is relatively good in the region, but the physical spatial resilience and information spatial resilience need to be improved (CN is about 0.3). The overall resilience of Colombo port and Nehru port is “D+” (CN < 0.4). Among them, Colombo port has the best informational spatial resilience and fair social spatial resilience but relatively low physical spatial resilience (physical space CN < 0.3, which is higher than Chittagong only). The overall resilience of Nehru port is relatively low among the sample ports in South Asia, and it is especially necessary to strengthen the resilience of physical space and social space.
The overall resilience and spatial resilience of ports in the Middle East and Africa region are shown in Table 11. The overall resilience of Said port and Aqaba port is relatively high (rated as “C−”), while the resilience ratings of Djibouti port, Jeddah port, Dubai port, and Abu Dhabi port are all “D+.” The results of every measure of spatial resilience show that Said port ranks highest in the regional sample in terms of physical and social spatial resilience and lower in terms of information spatial resilience but is still above 0.5. Aqaba port has lower resilience in physical space but is moderately resilient in social and information space. The rest of the ports have better information space resilience (information space CN > 0.55) but have shortcomings in physical space and social space resilience and lower overall resilience.
The overall resilience and spatial resilience of ports in Europe are shown in Table 12. The overall resilience of the sample ports in this region is generally at a high level (“C±”), and the port of Marseille is better with a “C+.” From the results of spatial resilience, it can be seen that the physical and social spatial resilience of Marseille port is relatively good (CN > 0.5), especially the social spatial resilience, which is close to the lower limit of “B−,” so that the comprehensive resilience of Marseille port not only ranks the highest among the sample ports in this region but also ranks second only to Ningbo Zhoushan port and Singapore port among the 28 sample ports. Other sample ports in Europe have a similar level of resilience, which fluctuates between 0.40 and 0.44. Although the informational spatial resilience of these ports is relatively good, especially the port of Piraeus, which is close to the “B−” level, the physical spatial resilience and social spatial resilience are at the “D+” and “C−” levels, respectively, and there is a big gap with Marseille port, Ningbo Zhoushan port, Singapore port, and Tanjung Priok port.
By comparing the sample ports in terms of all sample ports and the regions in which they are located, it can be found that Ningbo Zhoushan port, Singapore port, Karachi port, Said port, and Marseille port are the ports with the best overall resilience among the sample ports in China, Southeast Asia, South Asia, the Middle East and Africa, and Europe. Ningbo Zhoushan port, Singapore port, and Marseille port are also ranked in the top three of the 28 sample ports in terms of comprehensive resilience.
4.3. Vertical Analysis of Port Resilience Evaluation
We take four relatively high resilience ports, namely, Ningbo Zhoushan port, Singapore port, Karachi port, and Said port, as research objects, combining three key indicators that affect the resilience of physical space, social space, and information space (route network port degree (x6), annual throughput of the port container (x12), and number of fixed broadband subscribers per 100 people (x20)) to explore the development potential of these ports. For the route network port degree (x6), which is difficult to obtain data for and relatively insignificant to change, only static values are compared. According to the Report on Port Development along the 21st Century Maritime Silk Road, it shows that Singapore port > Ningbo Zhoushan port > Said port > Karachi port. For the annual throughput of the port container (x12) and the number of fixed broadband subscribers per 100 people (x20), these two indicators with sufficient data and significant change development can be analyzed vertically.
Based on the annual port container throughput of Ningbo Zhoushan port, Singapore port, Karachi port, and Said port from 2019 to 2021 in the Global Port Development Report 2021 published by the Shanghai International Shipping Research Center, Figure 3 is plotted.
From the figure, we can see that there is a stable upward trend in general from 2019 to 2021. On the background of the COVID-19 outbreak and its persistence to date, the annual container throughputs of Ningbo Zhoushan port and Said port have been increasing for two consecutive years, while the annual container throughputs of Singapore port and Karachi port show an upward trend in 2021 after a slight decline in 2020 and even exceed the annual container throughputs in 2019. Ningbo Zhoushan port’s annual container throughput has continued to rise for two consecutive years, with an annual growth rate of 12.9%, reaching 31.08 million TEU in 2021. The port of Said first increases its container throughput from 3.66 million TEU in 2019 to 4.01 million TEU in 2020 and then reaches 4.07 million TEU in 2021. The annual port container throughput of Singapore port fluctuates slightly in 2020, falling from 37.2 million TEU in 2019 to 36.87 million TEU but rebounding in 2021 and even surpassing the previous level at 37.47 million TEU. The annual port container throughput of Karachi port in 2020 (2.08 million TEU) is also lower than that of 2019 (2.1 million TEU), but in 2021, the throughput rises to 2.53 million TEU, with a growth rate of over 20% compared to 2020. In summary, the socio-spatial resilience of Ningbo Zhoushan port, Singapore port, Karachi port, and Said port is better, which is consistent with the assessed “C±” (CN > 0.4).
Based on the information from the World Bank database, Figure 4 shows the following.
As shown in Figure 4, the number of fixed broadband subscribers per 100 people is at a high level in both Ningbo Zhoushan port and Singapore port, while Karachi port and Said port do not exceed 10 subscribers until 2020. Among them, port of Singapore grew rapidly until 2010 and then stabilized at around 25 households. The port of Ningbo Zhoushan and its hinterland are relatively late in the development of fixed broadband compared to the port of Singapore, but they are growing relatively fast and continue to increase after being on par with the port of Singapore in 2017 and will have nearly 35 households by 2020. Similar to Ningbo Zhoushan port, Said port has been growing in the number of fixed broadband subscribers per 100 people since 2006, but the increase has been relatively small. It now needs to strengthen its fixed broadband construction to improve its informational spatial resilience. The port of Karachi in Pakistan has always been at a low level, but there is a lot of room for improvement. By developing information technology and strengthening the construction of fixed broadband, the information’s spatial resilience can be improved to a large extent and the comprehensive resilience of the port can be promoted.
In general, Ningbo Zhoushan port and Singapore port perform better in terms of route network construction, annual container port throughput, and the number of fixed broadband users per 100 people, while Karachi port and Said port have more room for improvement, which should especially strengthen the investment and construction of indicators related to physical space and information space and form a good interaction with social space to improve the overall resilience of the port.
Given the above, from the investors’ point of view, the comprehensive resilience of the port and its individual spatial resilience can be compared, which makes it possible to make an accurate judgment on the future construction direction of the port and then make an in-depth assessment of investment and construction. Investors should make suitable investment plans according to different projects so as to reduce the risk of investment as much as possible. For example, Singapore port, Tanjung Priok port, and Klang port, where the socio-spatial resilience and comprehensive resilience are unbalanced, will inevitably strengthen the socio-spatial resilience of the port in the future. The government will increase the role of public investment to lead investors and optimize the port’s services in logistics, commerce, and industry to make it better, providing a good environment for investors to make decisions to reduce the risk of their investment and construction. Chittagong port and Marseille port’s information space resilience needs to be strengthened. Information technology plays an important role in the development of the port now, and a port with unbalanced information space resilience and comprehensive resilience will focus on creating an intelligent information environment for the port. Information space resilience as the basis for the long-term development of the port requires the concerted efforts of enterprises and regional governments. Improving the information space resilience should be in line with the local information development policy, and the regional government will give more dividend policy and other support. The socio-spatial resilience and information spatial resilience of Hai Phong port and Abu Dhabi port are high, but the low physical spatial resilience makes their comprehensive resilience rating “C.” Since physical space resilience is the key to improving the comprehensive resilience of ports, those with low physical space resilience will need to strengthen investment and construction in the future. As we know, when ports are attacked by natural disasters, the infrastructure is the first to suffer; therefore, physical space resilience is an important indicator to assess the resilience of ports to disasters. The regional governments of ports with low physical space resilience are likely to pay more attention to the construction of new infrastructure, which will also provide new ideas and directions for investors to find new investment priorities. Investors can cooperate with regional governments in many ways to further expand the depth and breadth of investments, which will not only improve the physical spatial resilience of regional ports but also increase the income of investors.
5. Conclusion
This paper introduces the concept of “three-dimensional space” and the theory of “system of systems” into the field of ports, divides the resilience of ports along the Maritime Silk Road into “physical-social-information,” and, according to the characteristics of port resilience and the influencing factors of port resilience along the Maritime Silk Road, constructs an evaluation index system covering the whole process of port resilience. Also, an evaluation model of port resilience along the 21st Century Maritime Silk Road is built based on CRITIC-entropy value method and TOPSIS method. Under consideration of the comprehensive resilience and spatial resilience of the ports, the quantitative evaluation was carried out with 28 ports in 5 regions along the 21st Century Maritime Silk Road. Based on the horizontal analysis between regions in the spatial dimension, the ports with the highest resilience or the best resilience in the region are selected for vertical analysis. The main findings of the study are as follows. Firstly, in the port resilience evaluation index system constructed by the objective combination of the CRITIC-entropy value assignment method, the route network port degree (x6), the annual throughput of the port container (x12), and the number of fixed broadband subscribers per 100 people (x20) are the key indicators affecting the physical spatial resilience, social spatial resilience, and informational spatial resilience of ports, respectively. Secondly, the comprehensive resilience of ports along the Maritime Silk Road is mainly concentrated in the “C” region (0.4–0.6), but with a more obvious regional heterogeneity, which shows China > Southeast Asia > Europe > Middle East and Africa > South Asia. Thirdly, there is heterogeneity in the comprehensive resilience of ports within each region along the Maritime Silk Road. Fourthly, the coordinated development of physical, social, and informational spatial resilience has a catalytic effect on enhancing comprehensive resilience.
In order to promote the development of resilient ports, investors must choose to invest in each region, and this study makes the following recommendations for investors. First, for ports with low physical space resilience, investors should focus on the investment and construction of infrastructure such as terminals and berths in each port. Second, for ports with low social space resilience, investors should pay more attention to the regulations issued by each regional government under the influence of COVID-19 and the various active fiscal policies implemented by the local central bank in the economic downturn and invest in port projects supported by local government. Third, for ports with low information space resilience, investors should actively participate in the construction of 5G networks in ports, and even supporting industries such as computer software and hardware is also a focus of investment [23].
This paper has the following contributions. First of all, it greatly enriches the study of ports in terms of construction and investment, which finds new methods for the study of port resilience. Also, it avoids to a great extent the conflicts of interest between investors’ decisions and national and regional governments’ port construction and promotes the reasonable and optimal allocation of various resources, thus reducing the various risks that may arise when investors make investments. Moreover, it enriches and improves resilience research in the field of transportation. In addition, it has combined the CRITIC weight method, the entropy method, and the TOPSIS model to apply to the analysis of port resilience, which broadens the application field of the above methods and models. Last but not least, our paper is also the first to include political, medical, information technology, and other factors in port resilience assessment, which makes for a more comprehensive quantitative study of ports along the Maritime Silk Road.
There are some limitations in this paper that need further research and improvement. First, in terms of its research, more detailed investment suggestions are what investors need when providing them with a clear investment direction. Also, how to do it at the lowest cost will be the most worthy of deep investigation in the field of port construction and investment in the future. Second, the number of empirical research samples is small and not enough. Based on the availability of data and the limitation of processing time, only 28 ports in 5 geographical locations along the Maritime Silk Road were selected for empirical analysis in this study, and ports in the South Pacific were not involved, so the sample coverage was not enough. Third, the longitudinal analysis of ports with good resilience is relatively simple, and all indicators of the port resilience evaluation system are not analyzed. Due to the difficulty of data acquisition, only key indicators affecting the resilience of physical, social, and information space were selected in the longitudinal analysis part of this study, instead of covering the entire evaluation index system, and the time series analysis method was not selected in the research method. The above problems can be improved in the future.
Data Availability
The data used to support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This research was funded by the Major Project of the National Social Science Fund of China (grant no. 20&ZD070).