Quantifying Road Transport Resilience to Emergencies: Evidence from China

Abstract

Facing the shock of emergencies, how resilient is the road transport system? In this process, how are the system’s absorption capacity, adaptability, and recovery capacity? These are very important for the development of sustainable road transportation. Taking China’s road transport during the COVID-19 pandemic as the research object, this paper introduces an integrated resilience quantification method, draws a road transport resilience curve, and comprehensively and dynamically analyzes and compares the resilience of China’s road transport system at different stages among different regions and under different epidemic waves. The results show that the resilience of road passenger and freight transport differs in the face of external disturbance. Freight transport resilience is better than that of passenger transport. Compared to passenger transport, freight transport is more robust; the impacted speed is slower, the recovery speed is faster, the recovery capacity is stronger, and the affected period is shorter. There is regional heterogeneity in road transport resilience. This heterogeneity is reflected in the whole change process of system performance with external disturbance, including absorption capacity, adaptation capacity, and recovery capacity. The resilience of road transport under different waves of the epidemic is different. Compared to the first wave of the epidemic, the resilience of road transport indicators at all stages has been dramatically improved in the later rebound wave of the epidemic. This can help in the development of evidence-based road transport sustainability strategies.

1. Introduction

As an essential channel for transporting goods and passengers between modern cities, road transportation plays a vital role in the national transportation system, profoundly affecting the logistics, tourism, and transportation industries. Under the impact of emergencies, the road transport industry is under tremendous pressure. Giving full play to its supporting role, the development of the industry is also significantly affected. For example, the sudden outbreak of SARS in November 2002 had a significant impact on road transport and varying degrees of impact on the industry afterward. The outbreak of the novel coronavirus in December 2019 has severely affected China’s social economy. At the same time, the road transport industry has also been impacted severely and under tremendous pressure. A large number of studies have explored the impact of emergencies on road transportation, including the effects of natural disasters such as floods [1] on road transportation, and the effect on road transport of accidents and disasters [2], public health events [3], and social security incidents [4]. Among the above emergencies, the public health event represented by the COVID-19 pandemic has had a profound impact on society, with a widespread area (global event), a significant impact on society (work and school suspension, etc.), and a longer impact time (from the beginning of 2020 to now), which has far-reaching significance. Therefore, this paper takes the COVID-19 pandemic as an example to explore the impact of such emergencies on road transport.

Unlike an endemic disease (a regional disease that arises under particular circumstances) or an epidemic (an infectious disease that can infect a large number of people, and which can occur only in one region, or it can be a global pandemic), COVID-19 is a pandemic disease (affecting a wide range of areas and crossing provincial, national, and even continental boundaries). Transmission is characterized by outbreaks (the sudden appearance of many patients with similar clinical symptoms within a short time in a population in a local area or collective unit). It has the characteristics of time concentration and space aggregation. The COVID-19 pandemic has boosted many studies to understand the interplay between road transport and pandemics [5,6,7]. Most findings show that the pandemic and road closure have harmed road transport [8,9]. However, a few studies have found that the COVID-19 pandemic has positively impacted the road freight turnover rate, becoming more pronounced as the number of confirmed cases increases [10].

Meanwhile, some studies have found that the impact of COVID-19 on road freight has regional heterogeneity [11,12]. However, the existing studies have mainly focused on the external effect of the pandemic on road transport, such as transport demand [13,14], transport capability [15,16], operating transport cost [17], transport performance [18], carbon emission [19,20,21], and the ability of road transport system to resist this impact and maintain its typical performance, which we define as road transport resilience [22,23], is still unclear. As a part of sustainability, resilience is a sub-objective of sustainability [24], representing the changes in road transport sustainability under external influences. Sustainability focuses on long-term changes, while resilience focuses on short-term changes, and the stronger the resilience, the stronger the sustainability. In the face of the changing external environment, quantifying the resilience of road transportation and analyzing its response to this impact from the system’s perspective, as well as its adaptability and recovery ability, are crucial for the sustainable development of road transportation.

In the face of emergencies, such as the COVID-19 pandemic, this paper tries to comprehensively quantify the resilience of road passenger and freight transport according to its changing process under the shock of the pandemic. Accordingly, the main contributions of this paper are as follows: (1) an integrated framework is proposed to quantify the resilience of the road transport system, which can clearly reflect the whole change process of system performance after external disturbances, including the absorptive capacity, adaptive capacity, and recovery capacity. (2) This paper comprehensively and dynamically analyzes the resilience capacity of road transport systems in different stages among different provinces/cities and under different epidemic waves in China. It reveals the spatio-temporal heterogeneity of resilience of road passenger and freight transport. A literature review is presented in Section 2. The method for quantifying resilience and related data sources is described in Section 3. The results of the resilience quantification are presented in Section 4. Section 5 discusses the findings by combining the results obtained in this study with previous research. Section 6 summarizes the conclusions of the development and change in road transport resilience under emergencies.

2. Literature Review

The word “resilience” is derived from the Latin word “resilio”, meaning “return to the original state” [25], and is usually used to describe the ability of an object or system to return to its original normal state after being affected by a destructive event. The concept of resilience was first applied to physics, which refers to the resilience of a system after external disturbances. In the 1970s, the resilience theory was first applied to the field of systems ecology by Holling [26], a Canadian ecologist. Since then, the discussion of resilience in specific fields has ranged from ecosystems to economic and environmental systems to transportation systems. The resilience of the transport system is related to the shelter and emergency response ability of a city against disasters. As a guarantee for the safety and good operation of a city during disasters, transport resilience has become the focus of researchers of resilient cities. Its general definition can be described as the ability of a transport system to resist the effects of disruptive shocks and maintain its normal functioning [22,23,27,28]. Disruptive shocks here include natural disasters such as earthquakes or heavy rains, accident disasters such as safety incidents, public health events such as the COVID-19 pandemic, and other events that could disrupt the normal functioning of transport systems. Among them, the absorption and recovery ability of the transport system from natural disasters and accidents has been extensively studied [29,30,31], but there is less research on public health events, especially large-scale or even global ones, such as the COVID-19 pandemic. Below, we review these works in terms of the impact of COVID-19 on transport and the resilience assessment framework, respectively.

The COVID-19 pandemic has prompted a large number of studies to understand the impact of COVID-19 on transportation. The existing research has focused on tracking the evolution of the COVID-19 pandemic and its impact on people’s mobility, road, aviation, public transport, cruise ships, and other transportation. People’s mobility and travel modes have been a widely addressed [32,33] because they directly affect transportation. Overall, the impact of COVID-19 on transportation is significant [34,35] and negative [36,37,38], which is agreed generally. Meanwhile, the existing research has also observed that (1) the impact of the pandemic on transport varies by region [39,40]; (2) external intervention in the prevention and control of the pandemic can produce positive temporal and spatial effects [3,6,41]. Among the studies on the impact of the COVID-19 pandemic on transport, the studies in the field of road transport are few, and the studies that have quantified the resilience of road transport to the COVID-19 pandemic are even fewer. The existing research has mostly analyzed the impact of the pandemic on road transport from the perspective of the external system, such as the impact of prevention and control measures on road transport during the pandemic [3,42]. Less attention has been paid to changes in the resilience of road transport systems themselves under the impact of the COVID-19 pandemic. At the same time, most of the research has been qualitative [43], while there has been less quantitative research.

On the other hand, as a guarantee for the safety and operation of the system in response to disasters/disturbances, transportation system resilience is becoming the focus of research. Resilience evaluation is mainly carried out in three stages: a conceptual framework [44], a semi-quantitative index system [45,46], and a quantitative assessment [47]. The existing works can be divided into two categories. The first category measures system resilience by comparing the difference in the functional state of the system before and after a disaster. For example, TANG et al. [48] built a timeline according to a scenario, designed the evolution process of system safety resilience of the whole process before, during, and after the disaster, and measured the safety resilience of urban road traffic systems during waterlogging disasters through system performance changes after and before the disaster. However, this assessment method is qualitative. The key indicators to measure the system’s resilience are obtained by estimation, and the results are not necessarily accurate. In addition, the dynamic change process of system resilience with disaster/disturbance cannot be reflected. In terms of specific quantification methods of system resilience, there are mainly resilience measures of transportation systems based on resilience characteristics [30,46], and resilience measures of transportation systems based on optimal models [49,50]. However, most existing methods of quantifying resilience cannot cover all stages or include all resilience capabilities, such as the absorption capacity to resist disturbances, the recovery capability, etc. Furthermore, some quantitative evaluation methods are not consistent with the concept of resilience, such as considering only the performance loss and recovery of the system, but ignoring the robustness [51]. To overcome this problem, the second category analyzes and quantifies the capabilities that help characterize a system’s resilience, such as its absorption capacity, adaptive capacity, recovery capability, etc., and these capabilities are comprehensively measured to quantify the system’s resilience [52,53]. This is a comprehensive approach to dynamically analyzing system resilience while incorporating different performance measures and characterizing resilience as a system property. For example, Guo et al. [54] used performance-based methods to analyze the airport’s strain capacity and resilience during global public health events and integrated some metrics such as aircraft movements, passenger throughput, and freight throughput in the resilience metrics model to analyze and compare the resilience evaluation under different scenarios. The results show that the resilience index can well reflect the recovery of airports under different scenarios. To take into account the dynamic evolution of the system, time-based indexes [51,55] and evaluation frameworks are often used to measure the resilience of the system.

To sum up, in the study of the impact of emergencies on transportation, few researchers have paid attention to the impact of public health events on road transportation. Second, although the outbreak of COVID-19 has promoted unprecedented research, the existing studies have mostly focused on the overall impact of COVID-19 on the economy and the flow of passengers and goods in transportation without considering how disruption events affect transportation and how the specific capabilities of road transport systems change at different stages of resilience development. Furthermore, they are more based on the research background of developed countries [56], with more qualitative studies [43] and fewer quantitative studies. Third, the frameworks used to assess resilience fail to integrate resilience at different stages. To solve this problem, our study attempted to investigate the resilience of road transport to the COVID-19 pandemic. Compared to the existing research, our work may contribute in two ways. First, unlike previous studies, we used a quantitative approach to analyze how disruption events have affected road transport during the COVID-19 pandemic. Second, differently from previous studies that have only considered the overall transport resilience, we adopted a comprehensive assessment framework to evaluate the resilience of road transport by comprehensively considering the changing process from disruption to recovery, and finally, to a new stable state.

3. Materials and Methodology

3.1. Study Area and Data

In China, the volume of road transport is large, profoundly impacting society and the economy. The outbreak of COVID-19 in early 2020 spread rapidly in China, seriously affecting road transport. Since the outbreak of COVID-19, China has implemented a series of rapid and effective epidemic prevention and control measures, setting a good example for other countries. Based on the above factors, this paper uses road transportation in China as a case study. It should be emphasized that road transport in this paper refers to inter-city road transport and excludes intra-city road transport, which mainly includes scheduled passenger transport, tourist passenger transport, road logistics, etc. The main statistical indicators were freight volume, freight turnover, passenger volume, and passenger turnover. To present the complete change process of road transportation under the impact of COVID-19 and the differences among regions, this paper analyzes the change process from both the horizontal and vertical dimensions. For the horizontal dimension, Beijing (the national political center), Shanghai (the economic center), Hunan (the national transportation hub city), and Xinjiang (the representative city of Northwest China) were selected, which represent North China, East China, Central China, and Northwest China respectively. For the vertical dimension, the first wave of the epidemic in January 2020 and the subsequent rebound were selected for analysis.

This study selected the following passenger and freight transport indicators to reflect the road transport capacity: freight volume, freight turnover, passenger volume, and passenger turnover. The data in this paper are from the transportation statistical yearbook and the websites of the Ministry of Transport (https://www.mot.gov.cn/tongjishuju/gonglu/ accessed on 29 June 2023), National Bureau of Statistics (https://data.stats.gov.cn/easyquery.htm?cn=A01 accessed on 29 June 2023), Hunan Bureau of Statistics (http://tjj.hunan.gov.cn/hntj/tjsj/sjfb/hnsj/202306/t20230625_29383220.html accessed on 29 June 2023), and so on. All of them are open-access data, including the above four road passenger and freight transport index data of China, Beijing, Shanghai, Hunan, and Xinjiang for each month from November 2017 to December 2020, among which the data of Hunan Province are from a more extended period, from November 2017 to December 2021.

3.2. Methodology

For the quantitative measurement of system resilience, based on the classic conceptual framework of the “resilience triangle” proposed by Bruneau et al., many scholars have extended and proposed several quantitative evaluation schemes and methods for resilience (see Section 2 “Literature Review” for details). These methods measure system resilience through the differences in the system’s functional states before and after a disaster. However, the existing quantitative evaluation of resilience mostly ignores the capability of each stage of resilience development, such as the robustness of the system, the rate of system performance decline and system performance loss after the disturbance of destructive events, and the recovery speed and recovery ability of the system performance. To better measure the overall resilience of road transport to the COVID-19 pandemic and the capacity of each stage of resilience development, we introduced the resilience quantitative assessment method proposed by Nan and Sansavini in 2017 [52], which not only explains the key resilience capacity of each stage but also proposes a quantitative method. At the same time, the resilience capacity of each stage of the system is integrated to develop a comprehensive resilience metric. This is conducive to analyzing the system’s resilience in different stages, and it can better compare the resilience differences among different evaluation objects. Figure 1 provides a conceptual description of the approach framework to better quantify the resilience of the various stages of the system. The y-axis represents the measurement of system performance (P), and the x-axis represents time. The curve (also known as the resilience curve) represents the performance change process of the system after the impact of interference. It is assumed that td is the point at which the system is shocked by interference. After the shock, the system performance declines until the point tr, then begins to recover until the point t_ns, and the system reaches the new stable state. According to these time points, the system development process can be divided into four stages.

The first stage is the initial stability stage (t < t_d), which is the stage before the system is interfered with. The performance level during this stage is often taken as the target value, so the average value can be used as the baseline.

The second stage is the disturbance stage (t_d < t < t_r), in which the system absorbs the disturbance, and the performance declines until the lowest level. In this process, robustness (R) can be used to describe the absorption capacity of the system:

$$\mathrm{R}=\mathrm{m}\mathrm{i}\mathrm{n}\left\{\mathrm{P}\left(\mathrm{t}\right)\right\},{\mathrm{t}}_{\mathrm{d}}<\mathrm{t}<{\mathrm{t}}_{\mathrm{r}}$$

(1)

From this, the maximum impact of disturbance on the system can be calculated as MI = Baseline − R. In addition, the speed of decline (SD) and system performance loss during this stage (PL_DP) can also be used to reflect the system’s absorption capacity. SD can be approximated as:

$$\mathrm{SD}=\frac{\mathrm{P}({\mathrm{t}}_{\mathrm{d}})-\mathrm{P}({\mathrm{t}}_{\mathrm{r}})}{{\mathrm{t}}_{\mathrm{r}}-{\mathrm{t}}_{\mathrm{d}}}$$

(2)

The following formula can calculate the system performance loss at this stage:

$${\mathrm{PL}}_{\mathrm{DP}}={\int }_{{\mathrm{t}}_{\mathrm{d}}}^{{\mathrm{t}}_{\mathrm{r}}}\left(\mathrm{P}\right({\mathrm{t}}_0)-\mathrm{P}\left(\mathrm{t}\right)\mathrm{dt}$$

(3)

The larger the R, the smaller the PL_DP and SD, indicating the more potent the absorption capacity of the system and the higher the resilience at this stage.

The third stage is the recovery stage (t_r < t < t_ns), in which the system performance rises until it reaches a new stable state. In this process, the system’s adaptive capability and recovery capability can be expressed by the speed of recovery (SR) and the system performance loss (PL_RP). They can be approximately described as:

$$\mathrm{SR}=\frac{\mathrm{P}({\mathrm{t}}_{\mathrm{ns}})-\mathrm{P}({\mathrm{t}}_{\mathrm{r}})}{{\mathrm{t}}_{\mathrm{ns}}-{\mathrm{t}}_{\mathrm{r}}}$$

(4)

$${\mathrm{PL}}_{\mathrm{RP}}={\int }_{{\mathrm{t}}_{\mathrm{r}}}^{{\mathrm{t}}_{\mathrm{n}\mathrm{s}}}\left(\mathrm{P}\right({\mathrm{t}}_0)-\mathrm{P}\left(\mathrm{t}\right)\mathrm{dt}$$

(5)

Combined with the previous stage, the total performance loss of the system in the whole process is:

$${{\mathrm{TPL}=\mathrm{PL}}_{\mathrm{DP}}+\mathrm{PL}}_{\mathrm{RP}}={\int }_{{\mathrm{t}}_{\mathrm{d}}}^{{\mathrm{t}}_{\mathrm{n}\mathrm{s}}\mathrm{s}}\left(\mathrm{P}\right({\mathrm{t}}_0)-\mathrm{P}\left(\mathrm{t}\right)\mathrm{dt}$$

(6)

To reflect the adaptive ability of the system, the index of time average performance loss (TAPL) is introduced, which removes the influence of time and reflects the system’s adaptive ability and recovery ability better:

$$\mathrm{T}\mathrm{A}\mathrm{P}\mathrm{L}=\frac{\mathrm{T}\mathrm{P}\mathrm{L}}{{\mathrm{t}}_{\mathrm{ns}}-{\mathrm{t}}_{\mathrm{d}}}=\frac{{\int }_{{\mathrm{t}}_{\mathrm{d}}}^{{\mathrm{t}}_{\mathrm{n}\mathrm{s}}}\left(\mathrm{P}\right({\mathrm{t}}_0)-\mathrm{P}\left(\mathrm{t}\right)\mathrm{dt}}{{\mathrm{t}}_{\mathrm{ns}}-{\mathrm{t}}_{\mathrm{d}}}$$

(7)

The smaller the TAPL, the larger the SR, indicating that the stronger the system’s adaptive ability and recovery ability, the greater the system resilience at this stage.

The fourth stage is the new stable stage (t > t_ns), in which the system performance reaches a new stable state and is maintained at this level. It is worth noting that the new stable level may be equal to the initial stable level or higher or lower than the initial stable level. To reflect this change in the system, the recovery ability (RA) can be expressed as:

$$\mathrm{R}\mathrm{A}=\left|\frac{\mathrm{P}({\mathrm{t}}_{\mathrm{ns}})-\mathrm{P}({\mathrm{t}}_{\mathrm{r}})}{\mathrm{P}({\mathrm{t}}_0)-\mathrm{P}({\mathrm{t}}_{\mathrm{r}})}\right|$$

(8)

The larger the RA is, the stronger the system recovery ability is.

To assess system resilience comprehensively, an integrated metric (GR) was proposed by Nan and Sansavini (2017) [52], which integrates the resilience capability of the above stages:

$$\mathrm{G}\mathrm{R}=\int (\mathrm{R},\mathrm{S}\mathrm{D},\mathrm{S}\mathrm{R},\mathrm{T}\mathrm{A}\mathrm{P}\mathrm{L},\mathrm{R}\mathrm{A})=\mathrm{R}\times \frac{\mathrm{S}\mathrm{R}}{\mathrm{S}\mathrm{D}}\times {\mathrm{T}\mathrm{A}\mathrm{P}\mathrm{L}}^{-1}\times \mathrm{R}\mathrm{A}$$

(9)

The integrated measure reflects the whole change process of the system performance. In this paper, the system performance corresponds to the four road passenger and freight transport indexes proposed in Section 3.1. Each index’s change process forms a resilience curve to reflect the resilience of road passengers and freight transport. To compare the resilience among indicators with different dimensions, each index was normalized by comparing each indicator with its base value (the mean value of each index before the impact of the epidemic). Then the resilience curve of each index was drawn. Taking into account the cyclical nature of road passenger and freight transport indicators and the fact that the first outbreak of COVID-19 happened in early 2020, we used the average in 2019 of each indicator as the baseline. In this paper, 23 January 2020, is taken as the time of the first epidemic outbreak, which is the earliest time that China initiated a level 1 response to a major public health emergency. Based on the above, resilience curves of the road transport indicators under the impact of COVID-19 were drawn, and the resilience capacity values of each stage were calculated.

4. Results

To explore the situation of road transport in the context of the COVID-19 pandemic in the country as a whole and the sub-regions, the following analysis was carried out from two perspectives: at the national level and in provinces/cities in different zones.

4.1. Road Transport Resilience at the National Level

How has the performance of road transport systems changed since the COVID-19 outbreak? Did other factors influence these changes during the period? To exclude the interference of other factors during the pandemic outbreak, the following is a comparative analysis of road transport at the time of the COVID-19 pandemic and at the time in previous years.

4.1.1. Analysis of the Outbreak Time of COVID-19

The resilience curves of road transport to COVID-19 at the national level are shown in Figure 2. Different transportation indicators are represented in different colors, and the baseline values of each indicator are shown in the legend. Before the outbreak of COVID-19, all transportation indicators fluctuated around the baseline. However, with the outbreak of COVID-19, the values of each transportation indicator dropped rapidly and reached the lowest point in February, and then gradually recovered after the outbreak was brought under control. Around April, the volume of freight (Vof) and turnover volume of freight (Tvof) had recovered to normal levels, but the volume of passenger (Vop) and turnover volume of passenger (Tvop) had not recovered to normal levels, reaching a new equilibrium state around August. Under the impact of the COVID-19 epidemic, the four indicators generally reflect the same results, showing an overall change pattern of decline first and then recovery, but they show different resilience levels at different stages.

To compare these differences further, according to Formulas (1)–(8), this paper extracts the resilience ability of each indicator at different stages. The specific calculation process is shown in Appendix A, and the results are shown in

Table 1. Resilience capabilities of road transport to COVID-19 in different phases.

	SD	R (%)	SR	RA	TI	TAPL
Vof	73.87	28.69	36.27	1.00	3.19	33.08
Tvof	73.11	28.52	37.58	1.00	3.08	33.53
Vop	62.99	12.72	10.62	0.55	6.97	53.61
Tvop	64.83	14.04	11.03	0.55	6.87	54.10

Note: Vof, Tvof, Vop, and Tvop denote the volume of freight, the turnover volume of freight, the volume of passengers, and the turnover volume of passengers, respectively.

. It was found that the volume of freight and turnover volume of freight had a strong ability to absorb disturbance (Vof: 28.69, Tvof: 28.52). The impact duration was short (Vof: 3.19, Tvof: 3.08), the speed of recovery was fast, and the recovery ability was strong (Vof: 1.00, Tvof: 1.00). The time average performance loss was small in both the disturbance stage and the recovery stage. Meanwhile, the volume of passengers and turnover volume of passengers showed poor resilience in both the disturbance stage and the recovery stage.

According to the resilience metric Formula (9), the comprehensive resilience index of road transport is shown in Figure 3. The four dimensions in the figure respectively represent the resilience values of the four road transport indicators. Among them, the turnover volume of freight has the strongest resilience, followed by the volume of freight, and the volume of passengers has the weakest resilience. The resilience index of each indicator is as follows: turnover volume of freight > volume of freight > turnover volume of passengers > volume of passengers.

4.1.2. Analysis before and after COVID-19

The outbreak of COVID-19 happened to be during the Spring Festival travel rush, which is close to the Chinese New Year. Therefore, to exclude the impact of the Spring Festival travel rush on road transport, the passenger and freight transport during the same period in 2018 and 2019 were analyzed, as shown in Figure 4.

As can be seen in Figure 4, during the Spring Festival travel rush both in 2018 and 2019, the road volume of freight and turnover volume of freight decreased significantly and reached the lowest value in February, while passenger transport, on the contrary, increased considerably during the Spring Festival travel rush and reached the highest value in February, presenting a periodic change. In combination with the changes in passenger and freight transportation after the outbreak of COVID-19 in Section 4.1.1, it was found that the outbreak of COVID-19 had a heavy impact on road passenger transportation, and the degree of impact is self-evident. However, in terms of freight transportation, was the impact of COVID-19 positive or negative? For further analysis, resilience curves of each indicator were drawn based on the normalized processing of each indicator (to ensure the comparability of the indicators, the mean value of the previous year was taken as the baseline), as shown in Figure 5. According to Formulas (1)–(8), the resilience capabilities of the indicators at different stages in the same period were extracted, as shown in

Table 2. Resilience of road transport at different stages during the Spring Festivals of 2018 and 2019.

		SD	R (%)	SR	RA	TI	TAPL
2018	Vof	22.79	54.41	44.21	1.00	2.29	20.64
	Tvof	23.42	53.17	48.77	1.00	2.22	21.37
2019	Vof	24.63	50.75	50.28	1.00	2.24	22.70
	Tvof	24.99	50.03	52.90	1.00	2.20	22.66
2020	Vof	73.87	28.69	36.27	1.00	3.19	33.08
	Tvof	73.11	28.52	37.58	1.00	3.08	33.53

.

As can be seen in Figure 5 and Table 2, before the impact of COVID-19, freight transport saw a sharp decline in February each year and then quickly recovered. Before the outbreak of COVID-19, although there was a decline in freight transportation in previous years, the speed of damage and the maximum impact was low, the recovery speed was fast, the recovery ability was strong, and the affected time was short. Therefore, the outbreak of COVID-19 harmed freight transportation and, combined with the Spring Festival travel rush, impacted freight transportation.

4.2. Road Transport Resilience of Provinces/Cities in Different Zones

China is a vast territory, and there are significant differences among its districts. Was road transport affected differently by the COVID-19 epidemic in different regions? What were the differences? In addition, how has the resilience of road transport changed in the face of the rebound from the COVID-19 pandemic, which has lasted a long time in China, compared to the first shock? To this end, the resilience of road transport under the impact of COVID-19 from both the horizontal and vertical perspectives is analyzed in this section.

4.2.1. Horizontal Analysis

This section selects Beijing, Shanghai, Hunan, and Xinjiang for horizontal comparison and analysis. There are two reasons. First, they represent different regions (North China, East China, Central China, and West China). Second, they were all obviously affected by the epidemic. The baseline values of the road transport indicators of the provinces/cities are shown in Figure 6a–d. In terms of freight volume and freight turnover, Hunan Province had the highest, followed by Xinjiang. Hunan Province also had the highest passenger volume and turnover, followed by Beijing. Analysis of variance (ANOVA) was used to test the significance of city differences. The results are shown in Figure 6e, where “1” represents Beijing, “2” represents Shanghai, “3” represents Hunan, and “4” represents Xinjiang. The differences in all indicators among any cities are significant except the volume of freight traffic between Beijing and Shanghai, the turnover volume of freight traffic between Beijing and Shanghai, and the turnover volume of passenger traffic between Beijing and Shanghai, between Beijing and Xinjiang, and between Shanghai and Xinjiang.

The resilience curve of road transport to COVID-19 in provinces/cities is shown in Figure 7. Significant differences among the provinces/towns regarding road transport indicate high regional heterogeneity in road transport resilience, which is reflected in the different indicators of road transport and different performances in response to COVID-19. For example, Hunan suffered a larger damage speed in terms of freight volume. Still, its speed of recovery was fast, its impact time was short, its robustness was strong, and its recovery ability was strong. Meanwhile, Shanghai suffered a lower damage speed, but its speed of recovery was slow, and the impact time was long. Hunan suffered relatively small damage speed and strong robustness in terms of passenger transport, especially passenger turnover. Still, its recovery speed was slow, and its recovery ability was weak. Meanwhile, Shanghai suffered large damage speed and weak robustness. Still, its recovery speed was fast, and its recovery ability was strong. A detailed comparison can be seen in

Table 3. Resilience ability of road transport indicators at different stages in provinces/cities.

		SD	R (%)	SR	RA	TI	TAPL	GR
Vof	Beijing	30.85	39.02	33.99	1.00	3.05	35.52	1.21
	Shanghai	29.79	52.11	7.20	0.98	7.76	14.00	0.88
	Hunan	53.17	33.10	38.99	1.00	2.97	27.10	0.90
	Xinjiang	31.30	14.69	25.44	0.78	3.86	31.27	0.30
Tvof	Beijing	25.85	44.48	34.10	1.00	2.89	34.70	1.69
	Shanghai	42.34	25.32	11.62	1.00	7.69	29.48	0.24
	Hunan	51.28	35.48	38.16	1.00	2.95	25.72	1.03
	Xinjiang	32.33	15.97	21.64	0.77	4.25	27.18	0.30
Vop	Beijing	34.10	18.59	10.55	0.32	4.62	57.60	0.03
	Shanghai	65.66	0.76	9.24	0.72	7.67	49.58	0.00
	Hunan	64.15	15.28	9.38	0.39	4.62	53.08	0.02
	Xinjiang	63.46	1.22	11.04	0.47	4.62	49.27	0.00
Tvop	Beijing	57.58	13.62	5.95	0.28	4.62	55.35	0.00
	Shanghai	63.23	0.66	10.35	0.74	6.92	46.23	0.00
	Hunan	51.02	21.87	7.27	0.38	4.62	42.02	0.03
	Xinjiang	59.08	2.00	10.21	0.46	4.62	45.09	0.00

.

According to the detailed resilience indicators, the comprehensive resilience index (GR) of road transport in the provinces/cities is shown in Figure 8. Significant differences are shown both in freight transport and passenger transport. Regarding freight transport, Beijing showed the highest resilience (Vof: 1.21, Tvof: 1.69), followed by Hunan (Vof: 0.9, Tvof: 1.03), and Xinjiang and Shanghai showed low resilience. In terms of passenger transport, the resilience of passenger volume in Beijing was the highest (0.032) and the resilience of passenger turnover in Hunan was highest (0.028); the resilience in Shanghai and Xinjiang was low. As can be seen from the comparison in Figure 8a,b, the resilience of passenger transport under the impact of the epidemic was much lower than that of freight transport. That is, passenger transport was more vulnerable to the epidemic’s impact.

4.2.2. Analysis in Vertical

In this section, Hunan, a central city connecting the north and south of China, is used for the study’s vertical analysis for two reasons. Firstly, Hunan Province has an extensive transportation volume, which was affected by the epidemic. Secondly, Hunan, close to Hubei Province, was one of the first provinces/cities to initiate a level 1 response to the major public health emergency. The resilience analysis of road transport to COVID-19 is representative and can provide a reference for other cities.

Since the first COVID-19 patient was confirmed on 21 January 2020, Hunan has experienced multiple waves of COVID-19. Here, the first wave of COVID-19 in early 2020 and the rebound on 28 July 2021, were chosen for analysis. The resilience curves of the road transport indicators over the period are shown in Figure 9. The COVID-19 rebound led to a similar pattern of changes in road transport as in the first wave of the outbreak, but the extent and duration of the changes varied. A comparison of the indicators shows that the resilience of road traffic at different epidemic waves is different, although they show a similar overall pattern of change.

To reveal these differences further, according to Formulas (1)–(8), detailed resilience indicators extracted at different stages are shown in

Table 4. Resilience ability of road transport indicators at different stages in Hunan Province.

	First Wave of COVID-19 (23 January 2020)				Rebound of COVID-19 (28 July 2021)
	Vof	Tvof	Vop	Tvop	Vof	Tvof	Vop	Tvop
SD	53.17	51.28	64.15	51.02	4.64	4.27	34.81	44.40
R (%)	33.10	35.48	15.28	21.87	94.91	95.31	61.83	51.30
SR	38.99	38.16	9.38	7.27	6.22	6.00	16.60	21.28
RA	1.00	1.00	0.39	0.38	1.00	1.00	0.87	0.87
TI	2.97	2.95	4.62	4.62	1.92	1.88	3.10	3.10
TAPL	27.10	25.72	53.08	42.02	2.68	2.46	18.37	24.11
GR	0.90	1.03	0.02	0.03	47.53	54.26	1.40	0.89

. In both the first wave of COVID-19 and the later rebound, we observed that passenger transport exhibited high disruption rates and low recovery rates, with weak robustness and longer periods of disruption, while freight transport exhibited the opposite. Notably, passenger traffic did not return to its previous level after the first wave of the pandemic, and it also did not on 28 July 2021, when the pandemic rebounded. However, compared to the impact of the first wave, the resilience of the road transport indicators at all stages was significantly improved when the epidemic rebounded, indicating that the resilience of road transport has increased with the development of COVID-19.

5. Discussion

From the above, we found that road transport showed significant differences between passenger and freight transport, among provinces/cities, and between epidemic waves under the impact of COVID-19. In this section, we further uncover the mechanisms behind these differences by discussing the potential influencing factors that may have affected the resilience of road transport, providing some implications for sustainable and resilient urban management in the future.

5.1. Potential Influencing Factors of Road Transport Resilience

In general, the potential factors that may affect the resilience of road transport to the pandemic can be divided into two aspects: the spread characteristics, prevention, and control measures of the pandemic, and people’s travel willingness and travel demand. Below, we discuss these aspects by comparing the situation at the national and provincial/city level and under two epidemic waves in Hunan province.

The COVID-19 pandemic broke out nationwide in early 2020, spreading rapidly and widely. To prevent and control the spread of the epidemic, many places across the country have implemented lockdown measures [57], such as stopping work, stopping business, suspending classes, restricting people gathering, and advocating for work and learning at home, which greatly restricts the travel of residents. Coupled with a fear of the unknown, residents’ willingness to travel has been reduced to a minimum. Therefore, during the Spring Festival travel rush (10 January to 18 February 2020), when passenger transportation should have been increasing as usual, it fell instead of rising, showing its lowest value. However, freight transportation was different. Due to the traditional Chinese Spring Festival, workers stop work and logistics stop operation. Affected by this, freight transportation has a periodic decline during the Spring Festival travel rush, and then it gradually recovers to the usual level with the end of the holiday. The decline was especially high during the Spring Festival of 2020 (25 January 2020), reaching its lowest value. Therefore, the impact on freight transportation was not very big because the outbreak of COVID-19 happened just before the Chinese New Year. On the contrary, the travel restrictions for residents have increased the demand for logistics transportation and have even led to a rapid increase in freight transportation as the epidemic continues. This result is consistent with the findings of other scholars [10].

The finding that there is regional heterogeneity [11,12] in road transport resilience under the same external disturbance is consistent with the results of previous research. This result may be related to the differences in epidemic prevention and control measures, travel willingness, and travel demand of residents, which have been analyzed in existing studies [58,59,60]. Specifically, the resilience of road transport in Beijing and Hunan is higher than that in Shanghai and Xinjiang. We can explain this result with the following factors. First, Beijing and Hunan adopted rapid, accurate, active, and strict prevention and control measures first after the epidemic outbreak [61], while Shanghai and Xinjiang quickly entered the state of prevention and control but adopted “city closure” measures. Second, the impact brought by the reduction in travel demand for work and business during the pandemic in Shanghai was greater than in Beijing and Hunan, as Shanghai likely has more businesses that have been heavily affected by the pandemic, such as services, tourism, and hospitality [62]. Third, the people’s willingness to travel in Beijing and Hunan is greater than that in Shanghai and Xinjiang because Beijing and Hunan may undertake more road transportation tasks due to the political center and the geographical location of traffic.

In two waves of the epidemic in Hunan, due to different influencing factors, road transport resilience showed different heterogeneity patterns, which are consistent with previous research [60]. We can explain this result with the following factors. First, to prevent the rapid spread of the epidemic, Hunan Province implemented relatively strict epidemic prevention and control measures on a large scale in the early days of the epidemic, such as suspending production and schools and reducing the flow of people. Second, in the face of huge unknowns, people’s panic and vigilance are strong, and their compliance with prevention and control measures is higher. Under the comprehensive influence of the above factors, the resilience of road passenger and freight transport shows a low value, including transport volume and turnover. Third, in the subsequent rebound epidemic, as it has lasted for a long time, people’s initial panic and vigilance have been relieved, and they are more eager for “freedom”. At the same time, the prevention and control measures become more precise and relaxed because of the accumulation of anti-epidemic experience and the pressure to develop social and economic. More precise and differentiated pandemic prevention and control measures, such as close contact tracing based on big data, rapid nucleic acid testing, and targeted containment management, have been implemented [63]. Social vitality has been enhanced, and the resilience of road transport has been significantly enhanced. This result is consistent with the results of other studies [3].

5.2. Implications for Sustainable and Resilient Urban Management

Different from existing studies that have focused on the changes in road transport in the initial stage of the epidemic only, not only the change process of passenger and freight transport in all stages focused on in this paper, but also the resilience of road transport under the impact of COVID-19 was quantified, showing a complete change process and revealing the heterogeneity. These findings may provide valuable implications for sustainable and resilient urban management to deal with similar crises in the future.

First of all, it can inspire further improvement of epidemic control measures in the future. For persistent pandemics like COVID-19, balancing pandemic prevention and control and maintaining normal socio-economic development is a key concern. In the two waves of the epidemic, the improvement of the resilience of road transportation in Hunan Province has brought enlightenment. In particular, precise and differentiated epidemic prevention and control measures are recommended, which can be confirmed and supported by other relevant studies [59]. On the one hand, scientific tracking, rapid traceability, and directional closed management are needed [64]. On the other hand, it is necessary to increase safety education and publicity, eliminate social panic, and ensure normal and safe travel.

Secondly, some implications can be drawn for alleviating road passenger and freight transport imbalance during the pandemic. During the spread of the COVID-19 pandemic, the development trend of road passenger and freight transportation was almost opposite, which is not a benign development. How to alleviate this imbalance is a problem that we should pay attention to. The heterogeneity of road transport resilience in different provinces/cities during the epidemic has brought enlightenment. Specifically, it is recommended to accurately “close the city” [59] and fully mobilize social forces [65]. On the one hand, it is necessary to seal and control accurately to avoid unnecessary isolation and blockade. On the other hand, it is necessary to mobilize social forces, enhance health awareness, and correctly guide residents to travel, improving the flexibility and adaptability of road transport.

6. Conclusions

Based on an integrated resilience measurement model, taking China as an example, we quantified the resilience of road transport under the impact of COVID-19 and presented a whole change process of road transport resilience in representative provinces/cities from horizontal and vertical dimensions. The main conclusions are summarized as follows:

• Under the impact of COVID-19, the resilience shown by road passenger and freight transportation was different. The resilience of freight transport was stronger than that of passenger transport. Compared to passenger transport, freight transport was more robust, and its impact speed was slower, but its recovery speed was faster, recovery ability was stronger, and affected period was shorter.
• There was regional heterogeneity in road transport resilience. This heterogeneity is reflected in the whole change process of system performance with external disturbance, including the absorption capacity, adaptation capacity, and recovery capacity.
• The resilience of road transport under different waves of the epidemic was different. Compared to the first wave of the epidemic, the resilience of road transport indicators at all stages were greatly improved in the later rebound wave of the epidemic.

Overall, our study analyzed the resilience of road transport under the impact of the COVID-19 pandemic, which can help to develop evidence-based road transport sustainable development strategies. There are some limitations to the study. First, due to data collection limitations, we used the monthly data of road passenger and freight transport, which may be inaccurate. In the future, the changes in road transport should be measured more accurately by collecting daily data on passenger and freight transport. Second, only four provinces/cities (Beijing, Shanghai, Hunan, and Xinjiang) were selected for analysis, and the sample size for the horizontal comparison was small. The analysis sample should be increased in the future to reveal better the mechanism behind the heterogeneity and differences in road transport resilience. Third, only two waves of the epidemic, the first wave in January 2020 and the rebound wave in July 2021, were chosen for a vertical analysis. This is not enough, and it should be improved in the future.