Research on the Regional Transport Development Index and Its Application in Decision Making and Sustainable Development of Transport Services: A Case Study in Yunnan Province, China

Abstract

Transport and macroeconomic development are closely linked. Under the comprehensive management system for China’s rail, road, water, and air transport, there is an urgent need for a comprehensive, timely, and accurate index that reflects the relationship between transport and the macroeconomy, to help the government analyze the current situation of transport, judge the macroeconomic and sustainable transport trends, and make scientific decisions, so as to achieve the carbon emission peak and carbon neutrality goals. This paper innovatively proposed the Transport Development Index (TDI), a new evaluation system with 38 indicators, covering the infrastructure, development scale, efficiency, sustainability, and safety and profitability of the four modes of transport: road, rail, water, and air. The analytic hierarchy process and the entropy weight method were used to determine the weights for the indicators. Yunnan was selected as a case study to calculate the TDI values for 2016 to 2021 and analyze the transport service performance for each year. The study results show that the TDI can be used for analyzing the regional transport services and economic operation status, proving the positive effect of transport development on the macroeconomy or warning of possible risks, and facilitating scientific decision making for sustainable transport development.

1. Introduction

1.1. Background

As a fundamental service industry, the transport industry is closely linked to macroeconomic development [1]. Transport contributes to many processes and phenomena while participating directly or indirectly in them. In the new era of economic development, the physical transport volume is increasingly important in the judgment of the macroeconomic situation. For achieving the carbon emission peak and carbon neutrality goals, the scientific characterization of the linkage between sustainable transport and the overall transport economy has forward-looking significance in facilitating eco-friendly and low-carbon transport modes as well as sustainable development of the entire industry. Since 2013, China has been reforming its transport industry management agencies, and has established a comprehensive management system for rail, road, water, and air transport [2]. Under the new management system, there is an urgent need for a comprehensive, timely, and accurate index that reflects the relationship between transport and macroeconomy, to help the government analyze the current situation of transport, judge the macroeconomic and sustainable transport trends, and make scientific decisions.

The rapid development of the Internet makes it easier for the government to manage industries and companies to provide travel services. As Internet-based transport management and travel services are more and more popular, a large amount of transport data is collected and accumulated by relevant industry management agencies and companies, resulting in a massive amount of industry big data. For example, according to the Ministry of Transport of the People’s Republic of China [3], there was a total of 286 licensed online car-hailing companies nationwide, which had 4,816,000 licensed online car-hailing drivers and 1,987,000 vehicles, and received 554 million orders in September. However, a massive amount of big data is scattered across different information systems. The data in each system can reflect a certain aspect of transport services at a single level, providing limited support for government decision making, but cannot be used to comprehensively, timely, or accurately evaluate the transport industry.

Such data can be mined and used to construct an industry index for comprehensive, timely, and accurate evaluation of the performance of transport services.

1.2. Brief Literature Review

Currently, research on the construction of industry indexes and evaluation of transport services falls into two categories.

1.2.1. Government Research on the Relationship between Transport and Macroeconomy

The Bureau of Transportation Statistics of the US developed the Transportation Output Index (TOI) and officially released it on the New York Stock Exchange in January 2004. The TOI includes key indicators of the monthly operating volume and revenue data for the for-hire transport industry, covering six sub-sectors: air, rail, water, truck, transit, and pipeline. The TOI, along with separate indexes for freight and passenger, was constructed by aggregating component series. Each series, representing a transport sub-sector, was converted into an index number, seasonally adjusted, and nondimensionalized. The relationship between the index and key macroeconomic indicators, changes in the index, and economic growth cycles were studied to reveal the linkage between the transport sector and the economy [4].

The Ministry of Transport of the People’s Republic of China officially released the China Transportation Services Index (CTSI) in 2018. The CTSI covers four transport sub-sectors: rail, road, water, and air. Five freight indexes and four passenger indexes are aggregated with weights to reflect the relationship between the transport economy and macroeconomy [2].

Both the TOI and CTSI are limited regarding their applicability. They provide effective measures of the national transport performance but may not work for regional transport services such as Yunnan Province, China. In addition, the TOI and CTSI focus on freight and passenger transport, without analyzing the impact of different transport modes or dimensions on transport performance.

1.2.2. Indexes Constructed by Researchers for Evaluating Transport Services, Such as Indexes for Service Efficiency, Sustainability, Safety, Economic Efficiency, and Social Impact

There is increasing interest in the subject of the relationship between transport and the macroeconomy among researchers. Indexes are constructed by researchers for evaluating transport services in different dimensions. A massive amount of transport industry big data is helpful for researchers to carry out these studies. However, in some circumstances, dimensions are hard to describe with data, the data are incomplete, or there is a high level of uncertainty involved in the data. So, a combination of qualitative and quantitative modeling methods is commonly used to evaluate the transport service, such as the Entropy weight TOPSIS model, Data envelopment analysis model [5,6,7,8,9,10], etc. The dimension division mainly includes Service Efficiency, Sustainability, Safety, Economic Efficiency, and Social Impact in the available literature. Some representative research results are shown in

Table 1. Research review in the field of evaluating transport services and their impact.

Field	Author	Main Research Findings
Efficiency	[11,12,13,14,15,16]	The efficiency of airports takes into account three dimensions simultaneously: energy consumption, airplane traffic, and passenger movement. The consideration of equity in addition to efficiency when evaluating transport effectiveness. West African ports have bigger port sizes and higher container throughput TEUs compared to East African ports. These ports are in general less efficient than East African ports. Travel time (speed) and out-of-pocket costs paid for travel are the most important criteria influencing the choice of transport mode.
Sustainability	[17,18,19,20,21,22,23]	The specification of the factors in transportation, environment, society, and energy relationship as well as the interactions. The shock from the construction sector lowered the energy consumption of road transport in the long run. The shock in production in the industry and processing sectors reduced energy intensity in the short term but increased in the long term. Both technological change and technical efficiency change can promote comprehensive transportation green efficiency improvement. There is a noticeable difference in the relationship between economic growth and transport-related energy consumption after the economic crisis.
Safety	[24,25,26,27]	Under the premise of improving the statistical methods of relevant data, the death rate and traffic mileage the can be used to evaluate the level of road transport safety level. During the COVID-19 pandemic, there was a notable drop in the number of passengers in most of the countries affected by the spread of the virus.
Economic efficiency	[28,29,30,31,32,33]	For urban public transport in China, the impact of economic benefit on their coordinated development level was the largest, infrastructure was lower, and social benefit was insignificant. The UK has been more successful than most EU countries in decoupling the environmental impacts of road freight transport from GDP, because of the declining value of manufactured goods relative to GDP.
Social impact	[34]	The effectiveness of domain, content, and sentiment classifiers proposed in the comprehensive evaluation index framework is verified respectively.

.

1.2.3. Limitations of Current Research

These studies are limited in terms of their data sources and scope of application.

(1)

Range of indicators: Few studies have used a wide range of transport indicators to comprehensively evaluate the common transport modes (road, rail, water, and air). The aforementioned TOI and CTSI only include several basic indicators, which are not sufficient to fully evaluate transport services.

(2)

Scope of the research: Most existing studies focus on a single aspect of the transport industry, such as transport efficiency, sustainability, or safety, rather than evaluating the entire transport system from multiple dimensions in a macroscopic and comprehensive manner.

(3)

Research methodology: System dynamics have long been used as a basis for analyzing complex systems. However, few research methods are available for modeling and analyzing transport systems, and there are no mature algorithms yet.

(4)

Application of research findings: The results of current studies only provide limited support to government departments, especially the transport industry authorities in China, for scientific decision making and industry management.

1.3. Problem Statement

To address the aforementioned research limitations, this paper proposed the Transport Development Index (TDI), a new road, rail, water, and air transport evaluation system with 38 indicators, covering the five dimensions of highest concern for government decision making: infrastructure, development scale, efficiency, sustainability, and safety and profitability. This evaluation system can be used for analyzing transport service characteristics and evaluating the regional transport performance.

The infrastructure dimension characterizes the changes in transport corridors, transport network nodes, and means of transport. The development scale dimension characterizes the volume of passenger and freight transport. The transport efficiency dimension characterizes the utilization of transport means. The transport sustainability dimension characterizes the use of low-carbon, energy-efficient means of transport and public transport. The safety and profitability dimension characterizes the transport safety and the profitability of transport service companies.

The analytic hierarchy process and entropy weight method were used to assign weights to the indicators. Yunnan was selected as a case study. Taking 2018 as the base year, this study used data from information management systems of the government and transport companies for the period of 2016 to 2021 for standardization and calculation to obtain the total, road, rail, water, and air TDI values for 2016 to 2021 and analyze the province’s transport services in each year from the dimensions of infrastructure, development scale, efficiency, sustainability, and safety, and profitability.

The research has a special cognitive value and contributes to the enrichment of the theory of transport economics and the management. The research may be used by an economic practice or economic policy creator at the stage of policy programming.

1.4. Contents of This Paper

This paper contains six chapters. Chapter 1 introduces the background of the study and presents the research problem. Chapter 2 describes the research object, Yunnan Province, China, with information about its geography, economy, and transport development. Chapter 3 is about the research methodology, data sources, selection of transport development indicators, and model construction. Chapter 4 presents the results of the study—the total, road, rail, water, and air TDI values as well as the values of the indexes for the five dimensions of transport in Yunnan from 2016 to 2021. Chapter 5 analyzes the transport service characteristics based on the TDI values for each year and the relationship between changes in the TDI values and economic growth, which proves that the results are reliable. Chapter 6 summarizes the results, innovations, and contributions of this study, recommends applications of the results, and provides an outlook for further research.

2. Research Object

The object of the research is Yunnan, a province in Southwest China (21°08′–29°15′ N, 97°31′–106°11′ E), bordering Myanmar, Vietnam, and Laos. The province has an area of 394,100 km², 4.1% of the nation’s total, and a population of 47,209,000, 3.34% of the nation’s total. It is situated in a mountainous area, with a subtropical plateau monsoon climate [35,36,37,38].

In 2021, Yunnan’s GDP was CNY 2.71 trillion, an increase of 7.3% over 2020, ranking 18th in the country. Its primary, secondary, and tertiary industries contributed 14.3%, 35.3%, and 50.4%, respectively [39]. In recent years, Yunnan has made historic breakthroughs in the construction of transport infrastructure, and has built a comprehensive transport system. As of 2021, Yunnan has built over 10,000 km of expressways, 260,000 km of rural highways, 4741 km of railways, 5108 km of inland river navigation lines, and 15 civil airports. In 2021, Yunnan’s total freight transport volume was 1.439 billion tons, up 10.4% from 2020, freight turnover was 182.494 billion ton-kilometers, up 17.7% from 2020, passenger transport volume was 227 million, down 12.7% from 2020, and passenger turnover was 39.91 billion person-kilometers, up 6.1% from the year of 2020 [40].

3. Data and Methodology

3.1. Index Construction

The TDI was constructed taking into consideration the existing research results and the development of Yunnan’s transport industry, involving the following steps: establishing an evaluation system, determining indicators, processing data, setting weights and the base year, and aggregating the indicators into indexes. Figure 1 shows the research approaches and procedure of this paper.

After Yunnan’s transport reform in 2018, the transport authority is now primarily responsible for rail, road, water, and air transport, so the TDI system mainly focuses on these four modes of transport. The system evaluates these transport modes from five dimensions: infrastructure, development scale, transport efficiency, sustainability, and safety and profitability, with 38 indicators, including 12 for road transport, 11 for rail transport, 7 for water transport, and 8 for air transport, as well as data sourced from Yunnan Statistical Yearbooks and the information management system of the Department of Transport of Yunnan Province. According to the latest industry classification standard defined by the National Bureau of Statistics of China and the statistical yearbooks [39], the indicators cover up to 79% of the statistical indicators for the industry, so they can reflect the development of the transport industry to some extent.

Table 2. TDI indicators.

Mode of Transport	Dimension	Indicator	Calculation
Road	Infrastructure	Length of highways	N/A
		Tier-3 and above highway passenger stations	N/A
		Passengers and trucks	N/A
	Development scale	Road passenger volume	N/A
		Road freight volume	N/A
	Transport efficiency	Passenger efficiency	Road passenger turnover/Seating capacity of for-hire vehicles
		Freight efficiency	Road freight turnover/For-hire truck tonnage
	Sustainability	Bus lines	N/A
		New-energy buses	N/A
		The utilization rate of ETC on expressways	N/A
	Safety and profitability	Profits of key service companies	N/A
		Production safety	Number of bus accident claims/Road passenger turnover
Rail	Infrastructure	Railroad mileage	N/A
		Major rail stations	N/A
		Trains	N/A
	Development scale	Rail passenger volume	N/A
		Rail freight volume	N/A
	Transport efficiency	Passenger efficiency	Rail passenger turnover/Passenger trains
		Freight efficiency	Rail freight turnover/Freight trains
	Sustainability	Urban rail transit passenger volume	N/A
		Railway electrification rate	N/A
		Road-to-rail freight increment	N/A
	Economic efficiency	Profits of key service companies	N/A
Water	Infrastructure	Inland waterway mileage	N/A
		Port berths	N/A
		Ships	N/A
	Development scale	Water passenger volume	N/A
		Water freight volume	N/A
	Transport efficiency	Passenger efficiency	Inland water passenger turnover/Passenger capacity
		Freight efficiency	Inland water freight turnover/Net load
Air	Infrastructure	Civil airline mileage	N/A
		Airports	N/A
		Takeoffs and landings	N/A
	Development scale	Air passenger volume	N/A
		Air freight volume	N/A
	Transport efficiency	Passenger efficiency	Air passenger turnover/Takeoffs
		Freight efficiency	Air freight turnover/Takeoffs
	Economic efficiency	Profits of key service companies	N/A

below lists the indicators.

The indicator statistics for 2016 to 2021 were calculated on an annual basis, with 2018 as the base year and 100 as the base value.

3.2. Calculation Formulas

The analytic hierarchy process was used to calculate the TDI values. The 38 indicators were nondimensionalized, assigned with weights, and then used to calculate the values of indexes for road, rail, water, and air transport, respectively. After that, the calculated indexes were given weights and then aggregated into the TDI. The calculation formulas are as follows:

$$TD{I}_{kj}=100{\displaystyle {\sum }_{i=1}^n{w}_i\times I_j(i)}$$

(1)

where $TD{I}_{kj}$ is the index for transport mode k in year j; $w_i$ is the weight for the indicator i; $I_j$($i$) is the standard score of indicator i in year j; n denotes the number of indicators.

$$TD{I}_j={\displaystyle {\sum }_{i=1}^4{W}_k\times TD{I}_{kj}}$$

(2)

where $TD{I}_j$ is the TDI in year j; $W_k$ is the weight for transport mode k; $TD{I}_{kj}$ is the index for transport mode k in year j.

All calculations were completed using the SPSS software. For weight assignment, a critical step for index aggregation, both subjective and objective weighting methods were used to determine the weight for each indicator.

3.3. Weight Assignment

Weights measure the relative importance of each indicator or mode of transport to the TDI or the entire transport industry. The weight for an indicator was determined based on its objective weight obtained through the entropy weight method and its subjective weight obtained through the analytic hierarchy process.

3.3.1. Using the Entropy Weight Method to Obtain Objective Weights

The entropy weight method measures indicator dispersion through information entropy and a greater entropy value means a lower degree of dispersion, less information that the indicator can provide, and a smaller impact on the entire evaluation system, that is to say, a lower weight should be given to the indicator, and vice versa. The weights given by this method are relatively objective and reliable, without involving any personal opinions or bias.

First, we standardized the matrix $X=\left(x_{ij}\right)$ (i = 1, …, m; j = 1, …, n), where m is the number of years evaluated (data from 2017 to 2019 was used) and n is the number of indicators, to obtain the matrix $\tilde{Z}=\left(z_{ij}\right)$:

$$z_{ij}=\frac{x_{ij}}{\sqrt{{\displaystyle {\sum }_{i=1}^n{x}_{ij}^2}}}$$

(3)

$${\tilde{z}}_{ij}=\frac{x_{ij}-\min \left\{x_j\right\}}{\max \left\{x_j\right\}-\min \left\{x_j\right\}}$$

(4)

Then we used the deviation matrix $P=\left(p_{ij}\right)$:

$$p_{ij}=\frac{{\tilde{z}}_{ij}}{{\displaystyle {\sum }_{i=1}^n{\tilde{z}}_{ij}}}$$

(5)

to calculate the information entropy of each indicator:

$$E_i=-\frac{1}{\ln \left(n\right)}{\displaystyle {\sum }_{i=1}^n{p}_{ij}\ln \left(p_{ij}\right)}$$

(6)

and the weight for each indicator ${\alpha }_i$:

$${\alpha }_i=\frac{1-E_i}{{\displaystyle {\sum }_{i=1}^{n}1-E_i}}$$

(7)

3.3.2. Using the Analytic Hierarchy Process to Obtain Subjective Weights

The analytic hierarchy process relies on the qualitative analysis and judgments of decision-makers.

In this study, electronic questionnaires were distributed to 10 transport administration departments, 2 large transport companies, 3 industry associations, and 3 university research institutions via the Internet, and 121 feedback questionnaires were collected, of which 108 were effective. Among the questionnaire participants, 50.00% were industry managers, 36.11% were company employees, and 13.89% were researchers; 49.18% of them were specialized in road transport, 18.03% in rail transport, 16.39% in water transport, and 16.39% in air transport.

Based on the scoring of n indicators by m experts, we constructed m judgment matrices: $A_k$=($a_{ij,k})$ (i, j = 1, …, n; k = 1, …, m), and obtained judgment matrix:$A={(a_{ij})}_{m\times m}$

$$a_{ij}={\displaystyle \prod }_{k=1}^m{a}_{ij,k}^{1/m}$$

(8)

Then we normalized the judgment matrix $A={(a_{ij})}_{m\times m}$ to obtain the matrix $B={(b_{ij})}_{m\times m}$, where

$$b_{ij}=\frac{a_{ij}}{{{\displaystyle \sum }}_{i=1}^m{a}_{ij}}$$

(9)

The normalized feature vector ${\beta }_i$ was calculated as follows:

$${\beta }_i=\frac{{\displaystyle {\sum }_{j=1}^m{b}_{ij}}}{m}$$

(10)

The consistency ratio $\mathit{CR}$ was verified as follows:

$$CR=\frac{CI}{RI}=\frac{\frac{\lambda -n}{n-1}}{RI}$$

(11)

$$\lambda =\frac{1}{n}{\displaystyle {\sum }_{i=1}^n\frac{{(A\beta )}_i}{{\beta }_i}}$$

(12)

$\mathit{CI}$ is the average consistency indicator; $\mathit{RI}$ is the random consistency indicator, whose value can be obtained by looking up the table; $\lambda$ is the maximum characteristic root of matrix A. If $\mathit{CR}$ < 0.1, the inconsistency degree of matrix A would be considered to be within the tolerance range, and ${\beta }_i$ would be used as the subjective weight of the indicator.

According to the results, the contribution of each transport mode determined by empirical weights is in line with objective facts.

3.3.3. Final Weights for the Indicators

The formula for calculating the final weights $w_i$ is as follows:

$$w_i=0.5{\alpha }_i+0.5{\beta }_i$$

(13)

Table 3. Weights for the indicators.

Mode of Transport	Indicator	$\mathbf{Objective} \mathbf{Weight} {\mathit{\alpha }}_{\mathit{i}}$	$\mathbf{Subjective} \mathbf{Weight} {\mathit{\beta }}_{\mathit{i}}$	$\mathbf{Final} \mathbf{Weight} {\mathit{w}}_{\mathit{i}}$
Road	Length of highways	8.54%	9.29%	8.92%
	Tier-3 and above highway passenger stations	9.34%	7.18%	8.26%
	Passengers and trucks	7.43%	5.67%	6.55%
	Road passenger volume	7.24%	6.23%	6.74%
	Road freight volume	7.08%	10.46%	8.77%
	Road passenger efficiency	7.08%	8.98%	8.03%
	Road freight efficiency	7.44%	10.29%	8.86%
	Bus lines	6.62%	3.29%	4.96%
	New-energy buses	6.93%	2.97%	4.95%
	The utilization rate of ETC on expressways	10.71%	9.19%	9.95%
	Total profit of key highway service companies	14.98%	9.37%	12.17%
	The ratio of the number of bus accident claims to road passenger turnover	6.61%	17.08%	11.84%
Rail	Railroad mileage	10.06%	13.00%	11.53%
	Major rail stations	7.88%	9.42%	8.65%
	Trains	10.69%	6.04%	8.37%
	Rail passenger volume	9.30%	12.96%	11.13%
	Rail freight volume	10.60%	10.15%	10.38%
	Rail passenger efficiency	9.40%	11.45%	10.43%
	Rail freight efficiency	8.21%	8.68%	8.44%
	Urban rail transit passenger volume	7.74%	7.29%	7.52%
	Road-to-rail freight increment	8.71%	2.62%	5.66%
	Railway electrification rate	7.83%	3.21%	5.52%
	Total profit of key railway service companies	9.58%	15.18%	12.37%
Water	Inland waterway mileage	25.33%	12.82%	19.08%
	Port berths	25.33%	10.10%	17.72%
	Ships	10.32%	11.05%	10.69%
	Water passenger volume	9.67%	12.26%	10.97%
	Water freight volume	9.99%	19.64%	14.81%
	Inland water passenger efficiency	9.68%	13.12%	11.39%
	Inland water freight efficiency	9.68%	21.01%	15.34%
Air	Civil airline mileage	15.68%	6.83%	11.26%
	Airports	0.00%	11.96%	5.98%
	Takeoffs and landings	14.39%	9.77%	12.08%
	Air passenger volume	13.90%	16.03%	14.96%
	Air freight volume	13.05%	9.34%	11.20%
	Air passenger efficiency	16.43%	16.71%	16.57%
	Air freight efficiency	13.77%	8.77%	11.27%
	Total profit of key aviation service companies	12.79%	20.59%	16.68%

below lists the weights for the 38 indicators.

The final weights for the indicators and Formula (1) were used to calculate the TDI values for Yunnan’s road, rail, water, and air transport from 2016 to 2020, as shown in

Table 4. TDI values for each mode of transport in 2016 to 2020.

Mode of Transport	2016	2017	2018	2019	2020
TDI-road	93.80	98.68	100.00	105.11	92.21
TDI-rail	80.95	93.31	100.00	111.58	90.99
TDI-water	101.13	99.98	100.00	96.55	83.16
TDI-air	101.57	100.09	100.00	108.10	83.09

.

Based on the TDI values for road, rail, water, and air transport from 2016 to 2020, the entropy weight method and analytic hierarchy process were used again to determine the weights for these four modes of transport, as listed in

Table 5. Weights for the four modes of transport.

Mode of Transport	$\mathbf{Objective} \mathbf{Weight} {\mathit{\alpha }}_{\mathit{j}}$	$\mathbf{Subjective} \mathbf{Weight} {\mathit{\beta }}_{\mathit{j}}$	$\mathbf{Final} \mathbf{Weight} {\mathit{\omega }}_{\mathit{j}}$
Road	33.90%	31.35%	32.62%
Rail	26.90%	28.85%	27.88%
Water	19.40%	17.78%	18.59%
Air	19.80%	22.02%	20.91%

.

The analytic hierarchy process was used to determine the final weights for the five dimensions of each mode of transport, as shown in

Table 6. Final weights for the five dimensions.

Index	Infrastructure	Development Scale	Transport Efficiency	Sustainability	Safety and Profitability
Weight	0.33	0.03	0.29	0.20	0.28

.

4. Results

4.1. TDI Values

Table 7. Yunnan’s TDI values from 2016 to 2021.

TDI	2016	2017	2018	2019	2020	2021
Total	93.20	97.72	100.00	105.95	88.28	100.55
TDI-road	93.80	98.68	100.00	105.11	92.21	101.51
TDI-rail	80.95	93.31	100.00	111.58	90.99	115.55
TDI-water	101.13	99.98	100.00	96.55	83.16	87.51
TDI-air	101.57	100.09	100.00	108.10	83.09	90.66

and Figure 2 below show the calculated TDI values of Yunnan.

The total TDI shows a steady upward trend between 2016 and 2019, following a dramatic decline. After reaching the trough in 2020, it starts to recover but fails to reach the 2019 peak in 2021. TDI-road has the most similar values and trends to the total TDI, and it is slightly higher than the total TDI in 2020. TDI-rail rises rapidly between 2016 and 2019, declines in 2020, but surges back up to a new high in 2021. TDI-water continues to decline between 2016 and 2020, with a steep drop in 2020, and then starts to recover but fails to reach its 2019 level in 2021. TDI-air shows an upward trend between 2016 and 2019, and a slow recovery in 2021 after a decline in 2020.

4.2. Values of the Dimensional Indexes

Table 8. Values of the dimensional indexes.

Index	2016	2017	2018	2019	2020	2021
Infrastructure	94.75	98.03	100.00	103.79	110.15	112.36
Development scale	93.83	97.18	100.00	105.03	76.32	81.91
Transport efficiency	96.53	95.97	100.00	107.97	70.61	81.15
Sustainability	39.09	54.35	60.50	72.70	66.12	81.77
Safety and profitability	82.80	84.42	81.41	80.70	56.79	75.18

below lists the values of the indexes for the infrastructure, development scale, transport efficiency, sustainability, and safety and profitability, and Figure 3 shows their trends compared to the TDI trend.

The infrastructure index shows a year-on-year increase. The development scale and transport efficiency indexes move in line with the TDI. They almost overlap with the TDI curve between 2016 and 2019 but drop below the TDI curve after 2020. The sustainability index increases year by year, despite a slight decline in 2020. However, the overall sustainability index curve lies below the TDI curve. The safety and profitability index moves in line with but below the TDI.

4.3. Relationship between the TDI and Economic Development

To verify the linkage between the TDI and regional macroeconomy, we conducted a correlation analysis between the TDI and Yunnan’s GDP, and the result was 0.913, meaning a strong positive correlation between them. Figure 4 below shows their trends.

The two curves follow similar trends. To some extent, the TDI proves the positive effect of transport development on the macroeconomy and can be used to warn of possible risks.

5. Analysis and Discussion

5.1. Characteristics of Different Modes of Transport in Yunnan

COVID-19 has a significant impact on transport activities in various modes of transport [25]. Since the sudden outbreak of COVID-19 in 2020, strict control measures against COVID-19 have led to a significant reduction in passenger travel and freight movement, and even the suspension of some transport routes. After April 2020, the nationwide control measures were gradually lifted. However, as the outbreak continued to spread on a small scale across the country, local controls were adopted to restrict travel. Considering the rapid spread of the Omicron variant, China did not liberalize the control until December 2022.

Road transport is the dominant mode of transport in Yunnan. This can be confirmed by the fact that TDI-road has the most similar values and trends to the total TDI and that the weight for TDI-road is 32.62%, larger than the weight for any other mode of transport. In addition, road transport accounts for approximately 68% of the province’s total transport volume (rail 17%, air 13%, and water 2%) [40]. That is, road transport undertakes the majority of the province’s transport activities. With great technical and economic flexibility, road transport carries the largest number of passengers and freight in the province, providing powerful support to ensure smooth economic and social development even during COVID-19.

Rail transport is the fastest-growing mode of transport in Yunnan. Thanks to the vigorous construction of railway infrastructure in recent years, the province’s railway mileage in operation has grown considerably, from 1930 km in 2012 to 5000 km, with lines extended abroad [40]. Due to its significant advantages in carrying capacity and transport distance, along with the continuously improved infrastructure, rail transport is used more for passenger and freight transport.

For this mountainous inland province with less developed water infrastructure, water transport is the least important mode of transport and can be easily replaced by other more advanced and efficient modes.

Air transport has been seriously affected by COVID-19. From 2020 to 2022, most air routes, particularly international routes, were grounded, hindering the development of air transport.

5.2. Different Aspects of Yunnan’s Transport Services

The infrastructure index shows a year-on-year increase, without being much affected by the epidemic. Between 2016 and 2021, the four modes of transport saw varying degrees of growth in mileage, with roads up 26%, rail up 39%, waterways up 19%, and civil airlines up 38% [40]. The continuously improved transport infrastructure facilitates transport development.

The development scale and transport efficiency indexes move in line with the TDI. They almost overlap with the TDI curve before the outbreak, but drop below the TDI curve after the outbreak, due to a significant reduction in travel.

The sustainability of transport has improved year by year, despite a slight decline due to the epidemic. Thanks to the promotion and use of new modes of transport, new energy means of transport, and new technologies, public transport trips increased between 2016 and 2021, with bus trips increasing by 8.16% and metro trips increasing by 93.66% [40]. More people choose public transport to meet their travel needs, which now includes more new-energy vehicles. In 2021, the number of new-energy buses in the province was 1.5 times that in 2016, accounting for 65.43% of the total number of buses [40]. Moreover, Yunnan has made it clear that it will phase out diesel buses in favor of new-energy vehicles so that the number of new-energy buses will continuously increase. However, considering that the overall sustainability is below the TDI average, further efforts are needed to improve the transport structure and promote new-energy and clean-energy transport equipment as well as eco-friendly and low-carbon travel modes, so as to help achieve the carbon emission peak and carbon neutrality goals.

The safety and profitability index moves in line with but below the TDI, indicating that there are still safety risks and that most transport services are not profitable.

Overall, the continuously improved transport infrastructure in Yunnan facilitates its transport services, which depend mainly on travel demand and activities. Its sustainable transport development has great potential if more efforts would be made to promote eco-friendly and low-carbon transport modes and improve transport safety, efficiency, and service quality.

5.3. TDI-Based Linkage between Transport and Macroeconomy

Due to its technical and functional versatility, transport plays a leading and supporting role in the development of modern society, as supported by the strong positive correlation between the TDI and the regional macroeconomy. For one thing, as one of the productive sectors in the national economy, transport provides services for passenger travel and freight movement and creates additional value for the economy. For another, as an important transmission sector, transport facilitates the intermediate links between other production sectors and makes social activities easier. Connectivity and service performance determine the circulation and quality of resources required for economic and social development. Furthermore, the transport industry provides a platform to develop and promote innovative, eco-friendly applications and new, sustainable, low-carbon technologies.

5.4. TDI Has Advantages among the Existing Research Results

The TDI was compared with CTSI [2] and TOI [4] in the existing research results, as shown in

Table 9. Comparison of TDI, CTSI and TOI.

Index	Covering Mode of Transport	Second-Order Index	Method	Frequency	Indicator	Dimension	Research Area
TDI	Rail; Road; Water; air	TDI-Rail; TDI-Road; TDI-Wate; TDI-air	Analytic Hierarchy Process and Entropy Weight method	Year	38	Infrastructure; development scale; Efficiency; Sustainability; safety and profitability	Yunnan Province, China
CTSI	Rail; Road; Water; air	CITS-passenger transport; CTSI-freight transport	Fisher-ideal	Month	9	Transport volume	China
TOI	Rail; Road; Water; air; Urban public transport; pipeline	TOI-passenger transport; TOI-freight transport	Fisher-ideal	Month	8	Transport volume; profit	the US

.

It is found that the TDI has the following advantages:

(1)

The TDI index includes the four most widely used modes of transport (Rail, Road, Water, and Air). There are Second-order indexes to measure each mode of transport, which can more directly reflect the development of various modes of transport.

(2)

The TDI system has 38 indicators covering the infrastructure, development scale, efficiency, sustainability, and safety and profitability of road, rail, water, and air transport, The TDI provides more accurate and reasonable results based on a large amount of data. It can reflect the comprehensive characteristics of transport services more comprehensively.

(3)

Compared with the CTSI and TOI, the TDI is a meso-level index in which the research area is a province in China (Yunnan Province). The regional characteristics are more obvious and can help the government make accurate decisions.

6. Conclusions

This paper proposed the TDI for evaluating regional transport services in a more comprehensive, objective, and effective manner and revealing the relationship between transport and macroeconomy, so as to help the government to analyze the current situation of transport, judge the macroeconomic trends, and make scientific decisions for sustainable transport development. The TDI system has 38 indicators covering the infrastructure, development scale, efficiency, sustainability, and safety and profitability of road, rail, water, and air transport. The analytic hierarchy process and entropy weight method were used to determine the weights for the indicators. Yunnan was selected as a case study to calculate the TDI values for 2016 to 2021 and analyze the transport performance in each year. The study reached the following conclusions:

(1)

Compared to existing evaluation systems, the TDI system is more comprehensive and objective in terms of its scope and dimensions. It evaluates road, rail, water, and air transport from five dimensions: infrastructure, development scale, efficiency, sustainability, and safety and profitability.

(2)

In addition, the TDI provides more accurate and reasonable results based on a large amount of data from direct and authoritative sources, including extensive questionnaires and government systems, as well as reliable weights obtained through the analytic hierarchy process and entropy weight method.

(3)

The TDI has a strong positive correlation with the regional GDP and is in line with the regional macroeconomic trend. Therefore, the TDI can be used for analyzing the transport services and economic operation status, proving the positive effect of transport development on the macroeconomy or warning of possible risks, and facilitating scientific decision making.

(4)

The TDI of Yunnan shows a steady growth between 2016 and 2019, and an abrupt drop in 2020 due to the outbreak of COVID-19, followed by a recovery, which, however, does not yet reach the pre-pandemic level so far.

(5)

For Yunnan, the development of different modes of transport varies. The index for road transport has the most similar values and trends to the total TDI, showing the dominance of road transport in Yunnan.

(6)

The continuously improved transport infrastructure in Yunnan facilitates its transport services, which depend mainly on travel demand and activities. As for sustainability and safety and profitability, Yunnan’s sustainable transport development has great potential if more efforts would be made to promote eco-friendly and low-carbon transport modes and improve transport safety, efficiency, and service quality.

This paper is innovative in the following aspects:

(1)

The research is interdisciplinary and mainly relates to the issues of economic management and transport management.

(2)

It constructs the TDI for evaluating the performance of transport services.

(3)

It combines the analytic hierarchy process and the entropy weight method to obtain reliable weights for transport service analysis.

(4)

It applies big data in the transport industry to help government departments make scientific decisions.

While we believe that the TDI provides a valid method for evaluating and analyzing transport services, we recognize that refinements may be necessary to improve it. For example:

(1)

As different data is monitored and collected for different modes of transport, the TDI can be further refined by including additional indicators and dimensions.

(2)

The TDI can be improved to distinguish between passenger and freight transport.

(3)

It needs to be further demonstrated how the TDI reflects policy and structural shocks, which can have a profound impact on the transport industry in China.

More valuable studies can be conducted in the future. First, as the transport system is constantly improved, the TDI system can be extended to include indicators for monitoring new transport modes such as multimodal freight transport and intermodal passenger transport, or include more transport sustainability indicators for a more comprehensive evaluation of transport efficiency and sustainability. Second, the linkage between transport and macroeconomy is not limited to economic contribution. More studies can be conducted to explore the relationships between the TDI and industry chains, supply chains, etc. Third, based on this study that focuses on annual transport performance, further studies may use quarterly or even monthly data to analyze transport activities in a more refined manner.