Research on the Mechanism of Influencing Factors of the Urban Road Traffic Operation State

Wang, Tianxiao; Wu, Hao; Li, Qiang; Ni, Shaoquan; Qin, Tinghui; Niu, Yifan; Cao, Di

doi:https://doi.org/10.1155/2022/7283841

Wireless Communications and Mobile Computing

On this page

Abstract Introduction Conclusion Data Availability Conflicts of Interest References Copyright Related Articles

Special Issue

Cyber-Physical Mobile Computing, Communications, and Sensing for Industrial Internet of Things and Industry 4.0 2021

View this Special Issue

Research Article | Open Access

Volume 2022 | Article ID 7283841 | https://doi.org/10.1155/2022/7283841

Research on the Mechanism of Influencing Factors of the Urban Road Traffic Operation State

Tianxiao Wang,^1,2Hao Wu,²Qiang Li,³Shaoquan Ni,¹Tinghui Qin,²Yifan Niu,¹and Di Cao⁴

Academic Editor: Shunmei Meng

Received08 Jan 2022

Revised04 Mar 2022

Accepted12 Mar 2022

Published07 Apr 2022

Abstract

Profound understanding of an interaction mechanism among influencing factors in the perspective of urban road traffic operation is of great significance for scientific and effective urban congestion management. In this paper, 6 indicators are proposed for the road traffic operation in the aspects of average speed of road sections, regional traffic index, and number of traffic incidents. Based on this, 4 major influencing factors and 12 measurable subinfluencing factors are proposed according to urban traffic supply and demand, and the SEM (structural equation model) is established to find out their interaction mechanism. And then, several sets of local traffic data collected in Shenzhen are used for model fitting, validation, and path analysis. The mathematical results show that all 6 indicators affiliated to the road traffic operation have a good explanation when it comes to the change of operation status. Among 4 latent variables in the traffic supply and demand aspect, the service equipment operation level and control equipment operation level can positively influence the road traffic operation status, while the urban traffic demand plays a negative role. Complex interactions among four latent variables are further pointed out. Finally, on the basis of the path coefficient relationship among the influencing factors, scientific suggestions and guidance are provided for urban road management control.

1. Introduction

In the current situation of increasingly severe road traffic congestion, comprehensively combining the influencing factors of urban road traffic operation status and understanding and mastering its change interaction mechanism will help people deeply understand the urban road traffic operation mechanism and make practical and effective management and control decisions.

There have been many studies on issues related to the operation status of urban road traffic. Those researches can be roughly divided into three categories: single road sections and intersections, local road networks, and urban macro road traffic. Lang and Fu [1] used indicators such as saturation and vehicle speed to analyze and evaluate the traffic operation status of local sections of expressways based on actual data; Chen et al. [2] proposed innovative traffic index calculation methods accurately calculated and described the street-level traffic operation characteristics of small areas; Wei and Ma [3] regarded urban road traffic as a giant system at the macro level and studied the evaluation system and method of its coordination degree from the two subsystems of supply and demand. In recent years, the research scope has gradually developed from a single level to a combination of macro and micro level. Bai et al. [4] established a comprehensive evaluation method of traffic operation status by looking for the three-level evaluation links of highway points, lines, and surfaces. By summarizing the existing research, most of them discuss the operation status of urban road traffic from the perspective of evaluation, and its role is limited to identifying the quality of road traffic operation status at different levels [5, 6]. Moreover, urban road traffic is a giant system, and traffic supply, demand, and operation status are closely linked and interact. However, existing research often considers them separately, so that it is impossible to accurately grasp the operating state of the traffic state from an overall perspective, it is impossible to conduct in-depth research on its evolution law, and it is impossible to provide specific and effective guidance for the formulation of traffic management and control strategies [7–10].

The contributions of this paper include three parts. Firstly, from the perspective of combination of micro and macro, three indicators and factors of traffic supply, traffic demand, and traffic operation status are sorted out. The second is to construct SEM, complete the model modification and evaluation, clarify their relationship through the SEM, explore the interaction mechanism, and then analyze the evolution law of the urban road traffic operation status. Thirdly, according to the analysis results of the model, references are provided for urban traffic managers to formulate feasible control strategies.

This paper is organized as follows. In Section 2, the evaluation index system is proposed in the aspect of road traffic operation status. The structural equation model (SEM) is established in Section 3, and confirmatory factor analysis, correction, and model evaluation are carried out. A conclusion is made in Section 4.

2. Evaluation Index System

The evaluation index system is proposed based on the road traffic operation status. On this basis, starting from the two general directions that affect road traffic operation status, traffic supply, and traffic demand, focusing on service equipment, management and control equipment, urban traffic demand, and surrounding traffic demand in the city, sort out and identify the influencing factors of road traffic operation status.

2.1. Road Traffic Operation Status

Existing studies analyze and measure the road traffic operation status based on the overall average speed of the road section, the average regional traffic index, and the average congestion time [11, 12]. However, the indexes such as the average speed of the whole section and the average traffic index of the whole region ignore the different importance of variable levels of sections and regions to the overall road traffic operation status of the city, resulting in the lack of accuracy of the calculation results. In addition, these indexes are limited to the characteristics of the vehicle itself. On the other hand, the variables that determine the status of road traffic should also take the impact of some incidents connected with safety issues into considerations [13].

Based on this, this paper selects the regional traffic index and the average speed of road sections as evaluation indicators and categorizes them into the average speed of key roads, the average speed of other roads, the traffic index of key areas, and the traffic index of other areas from the spatial perspective. In addition, the two types of indicators that cause accidental damage, including accident alarms and congestion alarms, are included in the evaluation index system [14, 15]. The selection of indicators is shown in Table 1.

2.2. Influencing Factors of Road Traffic Operation Status

From a macro perspective, real-time road traffic demand and road traffic supply are two key factors that affect the state of road traffic. (1)Road traffic supply

The supply of urban road traffic [16] is generally composed of transportation infrastructure, that is, related services and control equipment, e.g., roads, vehicles, stations, transportation organizations, and services [17]. The road traffic supply can be subdivided into static supply and dynamic supply according to the speed and frequency of changes in the supply. The static supply index mainly covers the quantity of infrastructure, such as the length of roads at all levels and the number of control equipment, while the dynamic supply is reflected by some real-time data. These real-time data are mainly retrieved from urban road traffic facilities, which can be further divided into traffic service facilities and traffic management facilities according to their functions. The traffic service facilities include parking lots, sight guidance screens, etc., while the traffic management facilities mainly include traffic signs, traffic lines, physical isolation devices, traffic signal control equipment, traffic violation recognition and capturing equipment, etc. [18]. The traffic operation state is directly affected by the dynamic supply when the static supply of infrastructure and other facilities is determined.

Dynamic supply of road traffic mainly covers two aspects: service equipment operation level and control equipment operation level. Among them, service equipment includes parking, road traffic guidance, and networked coordinated control [19]. Control equipment includes road and intersection monitoring, signal light control, etc. The selection of specific indicators is shown in Table 2. (2)Traffic demand

Real-time traffic demand is reflected by the flow of passenger and freight traffic moving on the road, which indicates strong spatial and temporal characteristics. In terms of spatial characteristics, passenger and freight traffic flows in the scope of intracity and the municipal boundary area are extracted [20].

Intracity traffic measures the total amount of real-time vehicle traffic in the intracity area; municipal boundary traffic measures the passenger and freight traffic in and out of the city and on highways around the city [21, 22]. Among them, the traffic volume entering and leaving the intracity area and volume of high-speed traffic are divided based on the spatial characteristics, while those intracity demands including the volume of expressway, arterial road, and other roads are considered in the aspects of temporal characteristics. The selection of specific indicators is shown in Table 3.

3. Structural Equation Model (SEM)

3.1. In-Line Style

The structural equation model, also known as structural equation modeling, is a statistical method based on the covariance matrix of variables to analyze the relationship between variables. Compared with traditional statistical methods such as regression analysis, the structural equation model has the following advantages [23–25]. (1)Capable of processing multiple dependent variables simultaneously. In traditional multiple regression or path analysis, the coefficients of the dependent variables are calculated independently, which neglects the influences and relationships with other factors. The SEM considers multiple factors at the same time, and the analysis effect is more accurate(2)Error tolerant between dependent and independent variables. Many social and psychological studies involve variables that cannot be accurately and directly measured, which are known as latent variables. When analyzing the relationship between latent variables, neither the independent nor the dependent variable can be accurately measured. When traditional regression models deal with such problems, they will only consider the error of the dependent variable. SEM analysis allows measurement errors in both independent and dependent variables and has advantages in analyzing the correlation between latent variables and other issues(3)Simultaneous analysis of factor structures and relationships. In SEM analysis, the correlation between latent variables and their underlying factors is calculated at the same time. Compared with the traditional analysis method that calculates factor structures and relationships independently, SEM can comprehensively consider the interaction and role of all coexistence factors, appropriately adjust the factor structure, and obtain a more comprehensive correlation(4)More complex factor relationships are considered. Traditional factor analysis is difficult to deal with models that have more complicated subordination relationships such as one factor subordinate to multiple factors or consider higher-order factors, while SEM can make up for this deficiency(5)Able to evaluate the rationality of factor structure and affiliation. SEM can evaluate its rationality by calculating the overall fitting degree of different factor structures and affiliations to the same sample data, so as to obtain the closest relationship to the data

3.2. Model Structure

The SEM consists of two parts: measurement equation and structural equation [26]. The measurement equation describes the relationship between latent variables and indicators, while the structural equation describes the relationship between the latent variables. The measurement equation and structural model are shown in (1) and (2).

where are the exogenous and endogenous index vectors, respectively; are exogenous and endogenous latent variables, respectively; are the relationships between exogenous indicators and exogenous latent variables and the relationships between endogenous indicators and endogenous latent variables, respectively; and are the error terms of exogenous and endogenous indexes, respectively.

where are the exogenous and endogenous latent variables. Bare the relationship between the endogenous latent variables, are the influence of the exogenous variables on the endogenous latent variables, and are the structural equation residuals.

3.3. SEM Construction

The construction and analysis process of the SEM is shown in Figure 1. The process of SEM construction is as follows: (1) propose theoretical assumptions about subordination and correlation; (2) introduce the definition of related variables and classify them in different categories; (3) based on the proposed theoretical assumptions and variables, construct a SEM; (4) collect data; (5) perform confirmatory factor analysis and adjust the model based on the analysis results so as to meet the related tests of reliability and validity; (6) output the model evaluation results; and (7) perform path analysis and output the analysis results of the affiliation and correlation between the key variables.

The realization of the above process is based on AMOS, which is powerful visualization software for SEM analysis. Firstly, the conjecture model is constructed, and the visual conjecture model is constructed according to the theoretical hypothesis and variable determination. The elliptic variable represents the latent variable that is not easy to measure directly [27]. Rectangular variables represent measurement variables that measure latent variables; the circular variable represents the allowable error between the latent variable and the measured variable; directed edges represent the interaction between variables. Then, the model was tested and the data were input into the conjecture model for confirmatory factor analysis. The model was modified according to the modification suggestions given by AMOS, which mainly focused on the presence and direction of the interaction between variables and specifically modified the form of directed edges. Finally, the path analysis is carried out based on the modified model [28].

3.3.1. Model Establishment

According to the evaluation indicators of road traffic operation status and related influencing factors constructed, combined with the relevant rules of the SEM structural equation model, the model variables are listed in Table 4.

There are 5 latent variables of road traffic operation status, service equipment operation level, management and control equipment operation level, municipal boundary demand, and intracity demand, which further contain 18 subordinate indicators that are taken as measured variables. Based on the relevant theoretical knowledge and the above variables, a hypothetical SEM is constructed as shown in Figure 2.

3.3.2. Data Collection

Sample data of the measured variables are required to be input into the SEM so as to perform confirmatory factor analysis and path analysis, and there are specific requirements for the sample data volume. Existing researches [29, 30] show that the difference between the sample data volume and the number of parameters to be estimated in the hypothetical model needs to be greater than 50. Thus, the number of samples between 100 and 200 is suitable for estimating the SEM.

This paper uses the real traffic data collected in Shenzhen from June 1, 2020, to June 7, 2020. The frequency of retrieving data is 1 hour, and the sample data volume for each measurement indicator is 168. The dataset is shown in Table 5.

SPSS software is used to standardize the data. Afterwards, AMOS software is used to draw a hypothetical model and bring in sample data to complete the model confirmatory factor analysis, correction, and evaluation.

3.3.3. Confirmatory Factor Analysis, Correction, and Model Evaluation

This paper uses AMOS software, brings in sample data to the constructed hypothetical SEM, and carries out confirmatory factor analysis. In this process, the maximum likelihood estimation method is used to fit the model [31–33], and according to the analysis results, the defects of the hypothetical model are corrected, and the revised model structure is shown in Figure 3.

In the application of the SEM, after the hypothetical model is fitted with the sample data, the degree of fitness between the hypothetical model and the actual data must be tested through the corresponding structural equation fitness evaluation indicators [34, 35]. There are three commonly used evaluation indicators, namely, absolute fit index, value-added fit index, and parsimony fit index. Absolute fit index measures the degree of fitting between the constructed model and sample data [36–38]. Commonly used indicators include GFI (goodness-of-fit index), RMSEA (root mean square of approximate error), SRMR (root mean square of normalized residual error), etc. A value-added fit index, which is also called a relative fitness index, is a statistic obtained by comparing the theoretical model with the benchmark model which indicates the degree of improvement. Benchmark models are usually the models with the most limitations and the worst fitness. Commonly used indicators include NFI (normed fit index) [40], TFI (nonnormed fit index) [39], and CTI (comparative fit index) [40]. Parsimony fit is a kind of indicator derived from the previous two types of fit [42]. Based on the idea of parsimony ratio of and , which are the degree of freedom of the theoretical model and benchmark model, respectively, the commonly used indicators include the chi-square degree of freedom (/DF), parsimony normed fit index (PNFI), and parsimony relative noncentrality index (PCFI). This paper uses 6 indicators including GFI, NFI, TLI, chi-square freedom ratio (/DF), PNFI, and PCFI [43–45]. The fitting results are shown in Table 6. Table 6 shows that all indicators of the revised model meet the fitting standard.

In addition, it is necessary to carry out a variation extraction value (AVE) and combination reliability (CR) test for latent variables. where is the path coefficient, is the number of measured variables, and is the residual error [39].

Based on the calculation methods above, the results are shown in Table 7.

It indicates ideal convergence validity when the AVE is greater than 0.5 and the combination reliability is greater than 0.8. As shown in Table 6, the SEM has good performances in the two aspects. The above results show that the revised SEM of road traffic operation status meets the relevant requirements and standards, effective factor division measurement, good fit of the sample data, and reasonable structure. It can accurately reflect the relationship between relevant influencing factors in the perspective of the urban road traffic operation state.

3.3.4. Path Analysis

Based on the basic structure of the SEM of the road traffic operating state, the path analysis is carried out and the path diagram is shown in Figure 4.

The range of path coefficient of latent variables is . The path coefficient approaching to +1 indicates a stronger positive correlation between variables, while the path coefficient approaching to -1 indicates a stronger negative correlation. And the closer the coefficient is to 0, the weaker the correlation between the variables [46, 47]. The specific values are shown in Table 8. The analysis shows that for the road traffic operation status, the operation level of service equipment and control equipment at the traffic supply level have a positive influence on it, between which the operation level of service equipment has a relative higher influence. Thus, the high operation level of service equipment will positively improve the operation status of road traffic. At the traffic demand level, the intracity demand generated by the driving of vehicles on the roads in the city and municipal boundary has a negative impact on the status of road traffic, while the impact from the intracity area is relatively greater. In addition to the status of road traffic operation, there are also obvious correlations between traffic supply and demand related factors, which are concentrated in the promotion of the generation and change of municipal boundary and intracity road traffic demand by the operation level of service equipment. It can be seen that higher service equipment operation level will not only improve the traffic operation status but also stimulate more traffic demand. There is also a correlation between the municipal boundary and intracity area in the aspect of road traffic demand. The main reason is that the two are strongly related at the spatial level. The dynamic traffic flow changes within the intracity area and municipal boundary will make influences on each other.

The relationship between the latent variables and their key measurement variables (the absolute value of the path coefficient is greater than 0.8) is shown in Table 9. The analysis shows that the key points of the measured variables, the average speed and key points of other roads, and the traffic index of other regions have a strong ability to explain the road traffic operation status. It can be seen that the average speed and traffic index can be distinguished according to the spatial scope to be more detailed and accurate. Show the real situation of road traffic operation status. The networked coordinated control of the ratio of intersections and the online rate of road checkpoint monitoring have the strongest ability to explain the two types of latent variables related to traffic supply. Except for the traffic volume for leaving the city, the traffic volume on all levels and types of roads in other cities has a strong correlation with traffic demand.

3.3.5. Suggestions on Road Traffic Control Strategies

By combining the path coefficients between the latent variables and the latent variables and key measurement variables, the following road traffic-related management and control strategy are proposed. (a)Using the form of spatial division to split the macroscopic average speed of the whole road section or the whole area traffic index and taking factors such as safety into consideration can describe the road traffic operation status in a more comprehensive and detailed manner(b)The interaction mechanism between the related factors of road traffic operation status is complicated. In order to improve the road traffic operation status, it is necessary to comprehensively consider key factors related to traffic supply and demand at the same time and clarify the internal interaction relationship between supply and demand(c)Improving the operation level of road traffic service equipment can directly promote the operation of road traffic, but it will also stimulate more traffic demand. The two types of promotion exist at the same time, and there is a certain balance between them(d)Focus on improving the operation level of road management and control equipment or restrain the road traffic demand to a certain extent. For example, reducing the traffic volume of private cars can relatively more effectively improve the overall road traffic operation status

3.3.6. Limitations of the Model

Due to the limitation of datasets, the selected influencing factors can be still enriched, and the research samples can be expanded. Further studies concentrating on the interaction mechanism among related influencing factors can be carried out with enriched data considering cases in different environments and seasons.

4. Conclusion

In summary, this paper establishes SEM, which can conduct a comprehensive and in-depth analysis and discussion on the interaction mechanism among influencing factors of road traffic operation status from the data perspective. It can quantify their interactions, provide new ideas for a comprehensive and in-depth understanding of the changes in road traffic operation status, and provide scientific support for formulating targeted, practical, and effective management and control strategies aimed at improving road traffic operation status.

Data Availability

No data were used to support this study.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

References

L. Haipeng and F. Jingjing, “On the traffic operation evaluation for basic road sections of an expressway,” Technology & Economy in Areas of Communications, vol. 15, no. 3, pp. 95-96, 2013.
View at: Google Scholar
C. Xi, S. Guohua, and Z. Xi, “Design and analysis of traffic indicator calculation method for residential districts,” Journal of Highway and Transportation Research and Development, vol. 36, no. 7, pp. 136–142, 2019.
View at: Google Scholar
W. Lianyu and M. Yongfeng, “Supply-demand coordination development of urban road traffic system,” Journal of Traffic and Transportation Engineering, vol. 4, pp. 58–61, 2004.
View at: Google Scholar
B. Hua, W. Jianjun, and J. Yimei, “Evaluation of highway network traffic operation state based on point, line, and area,” China Journal of Highway and Transport, vol. 31, no. 11, pp. 197–204, 2018.
View at: Google Scholar
Z. Cai and X. Zheng, “A private and efficient mechanism for data uploading in smart cyber-physical systems,” IEEE Transactions on Network Science and Engineering., vol. 7, no. 2, pp. 766–775, 2020.
View at: Publisher Site | Google Scholar
X. Xiaolong, H. Li, X. Weijie, Z. Liu, L. Yao, and F. Dai, “Artificial intelligence for edge service optimization in internet of vehicles: a survey,” Tsinghua Science and Technology, vol. 27, no. 2, pp. 270–287, 2022.
View at: Google Scholar
X. Zheng and Z. Cai, “Privacy-preserved data sharing towards multiple parties in industrial IoTs,” IEEE Journal on Selected Areas in Communications., vol. 38, no. 5, pp. 968–979, 2020.
View at: Publisher Site | Google Scholar
Q. He, C. Wang, G. Cui et al., “A Game-Theoretical Approach for MitigatingEdge DDoS Attack,” IEEE Transactions on Dependable and Secure Computing, p. 1, 2021.
View at: Publisher Site | Google Scholar
C. Hu, W. Fan, E. Zeng et al., “Digital twin-assisted real-time traffic data prediction method for 5G-enabled Internet of vehicles,” IEEE Transactions on Industrial Informatics, vol. 18, no. 4, pp. 2811–2819, 2022.
View at: Publisher Site | Google Scholar
X. Xu, Z. Fang, L. Qi, X. Zhang, Q. He, and X. Zhou, “TripRes,” ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), vol. 17, no. 2, pp. 1–21, 2021.
View at: Publisher Site | Google Scholar
X. Jianmin, W. Jia, and S. Yanfang, “Comprehensive evaluation of urban road traffic operation status based on game theory-cloud model,” Journal of Guangxi Normal University: Natural Science Edition, vol. 38, no. 4, pp. 1–10, 2020.
View at: Google Scholar
J. Mabrouki, M. Azrour, G. Fattah, D. Dhiba, and S. E. Hajjaji, “Intelligent monitoring system for biogas detection based on the Internet of things: Mohammedia, Morocco city landfill case,” Big Data Mining and Analytics, vol. 4, no. 1, pp. 10–17, 2021.
View at: Publisher Site | Google Scholar
A. Guezzaz, Y. Asimi, M. Azrour, and A. Asimi, “Mathematical validation of proposed machine learning classifier for heterogeneous traffic and anomaly detection,” Big Data Mining and Analytics, vol. 4, no. 1, pp. 18–24, 2021.
View at: Publisher Site | Google Scholar
L. Wang, X. Zhang, T. Wang et al., “Diversified and Scalable Service Recommendation With Accuracy Guarantee,” IEEE Transactions on Computational Social Systems, vol. 8, no. 5, pp. 1182–1193, 2021.
View at: Publisher Site | Google Scholar
L. Qi, C. Hu, X. Zhang et al., “Privacy-aware data fusion and prediction with spatial-temporal context for smart city industrial environment,” IEEE Transactions on Industrial Informatics, vol. 17, no. 6, pp. 4159–4167, 2021.
View at: Publisher Site | Google Scholar
C. B. Li, Q. Y. Zhao, R. X. Guo, and L. L. Tian, “Research on the measurements of the disequilibrium degree of the urban agglomeration traffic supply and demand,” Advanced Materials Research, vol. 912-914, 2014.
View at: Publisher Site | Google Scholar
X. Xu, Z. Fang, J. Zhang et al., “Edge content caching with deep spatiotemporal residual network for IoV in smart city,” ACM Transactions on Sensor Networks (TOSN), vol. 17, no. 3, pp. 1–33, 2021.
View at: Publisher Site | Google Scholar
J. Jin, X. Zhu, B. Wu, J. Zhang, and Y. Wang, “A dynamic and deadline-oriented road pricing mechanism for urban traffic management,” Tsinghua Science and Technology, vol. 27, no. 1, pp. 91–102, 2022.
View at: Publisher Site | Google Scholar
Y. Liu, Z. Song, X. Xiaolong et al., “Bidirectional GRU networks-based next POI category prediction for healthcare,” International Journal of Intelligent Systems, pp. 1–21, 2021.
View at: Publisher Site | Google Scholar
K. Chandan, A. Seco, and A. Silva, “A real-time traffic signal control strategy under partially connected vehicle environment,” Promet Traffic & Transportation, vol. 31, no. 1, pp. 61–73, 2019.
View at: Publisher Site | Google Scholar
H. Fujii, “Multi-agent based traffic simulation at merging section using coordinative behavior model,” Computer Modeling in Engineering and Sciences, vol. 63, no. 3, 2010.
View at: Google Scholar
X. Xia, F. Chen, Q. He et al., “Data, user and power allocations for caching in multi-access edge computing,” IEEE Transactions on Parallel and Distributed Systems, vol. 33, no. 5, pp. 1144–1155, 2022.
View at: Publisher Site | Google Scholar
Q. Zhenghao, Principles and Applications of Structural Equation Modelling, 2009.
Z. Tong, F. Ye, M. Yan, H. Liu, and S. Basodi, “A survey on algorithms for intelligent computing and smart city applications,” Big Data Mining and Analytics, vol. 4, no. 3, pp. 155–172, 2021.
View at: Publisher Site | Google Scholar
L. Yuan, Q. He, S. Tan et al., “Coopedge: a decentralized blockchain-based platform for cooperative edge computing,” in Proceedings of the Web Conference 2021, pp. 2245–2257, Vitual, 2021.
View at: Publisher Site | Google Scholar
A. A. Igolkina and M. G. Samsonova, “SEM: structural equation modeling in molecular biology,” Biophysics, vol. 63, no. 2, pp. 139–148, 2018.
View at: Publisher Site | Google Scholar
A. Amraeinia, Y. Zuo, and J. Zheng, “Interface modification of TiO₂ electron transport layer with PbCl 2 for perovskiote solar cells with carbon electrode,” Tsinghua Science and Technology, vol. 27, no. 4, pp. 741–750, 2022.
View at: Publisher Site | Google Scholar
X. Xu, H. Tian, X. Zhang, L. Qi, Q. He, and W. Dou, “DisCOV: distributed COVID-19 detection on X-ray images with edge-cloud collaboration,” IEEE Transactions on Services Computing, 2022.
View at: Publisher Site | Google Scholar
Q. Wentao and K. Lingzhen, “Path analysis of PPP project performance based on SEM model,” Journal of Wuhan University of Technology: Information & Management Engineering, vol. 39, no. 2, pp. 202–207, 2017.
View at: Google Scholar
X. Zhou, W. Liang, W. Li, K. Yan, S. Shimizu, and K. Wang, “Hierarchical adversarial attacks against graph neural network based IoT network intrusion detection system,” IEEE Internet of Things Journal, 2021.
View at: Publisher Site | Google Scholar
F. Wang, H. Zhu, G. Srivastava, S. Li, M. R. Khosravi, and L. Qi, “Robust collaborative filtering recommendation with user-item-trust records,” IEEE Transactions on Computational Social Systems, pp. 1–11, 2021.
View at: Publisher Site | Google Scholar
X. Zhou, X. Xu, W. Liang, Z. Zeng, and Z. Yan, “Deep-learning-enhanced multitarget detection for end-edge-cloud surveillance in smart IoT,” IEEE Internet of Things Journal, vol. 8, no. 16, pp. 12588–12596, 2021.
View at: Publisher Site | Google Scholar
X. Zhou, W. Liang, K. Wang, and L. T. Yang, “Deep correlation mining based on hierarchical hybrid networks for heterogeneous big data recommendations,” IEEE Transactions on Computational Social Systems, vol. 8, no. 1, pp. 171–178, 2021.
View at: Publisher Site | Google Scholar
Z. Cai, Z. Xiong, X. Honghui, P. Wang, W. Li, and Y. Pan, “Generative adversarial networks,” ACM Computing Surveys., vol. 54, no. 6, pp. 1–38, 2021.
View at: Publisher Site | Google Scholar
X. Shiyuan, X. Chen, and Y. He, “EVchain: an anonymous blockchain-based system for charging-connected electric vehicles,” Tsinghua Science and Technology, vol. 26, no. 6, pp. 845–856, 2021.
View at: Publisher Site | Google Scholar
R. H. Hoyle, “Structural equation modeling: concepts, issues, and applications,” The Statistician, vol. 45, 1996.
View at: Publisher Site | Google Scholar
X. Zhou, X. Yang, J. Ma, and K. Wang, “Energy efficient smart routing based on link correlation mining for wireless edge computing in IoT,” IEEE Internet of Things Journal, 2021.
View at: Publisher Site | Google Scholar
X. Zhou, Y. Li, and W. Liang, “CNN-RNN based intelligent recommendation for online medical pre-diagnosis support,” IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 18, no. 3, pp. 912–921, 2021.
View at: Publisher Site | Google Scholar
P. M. Bentler and D. G. Bonett, “Significance tests and goodness of fit in the analysis of covariance structures,” Psychological Bulletin, vol. 88, no. 3, pp. 588–606, 1980.
View at: Publisher Site | Google Scholar
P. M. Bentler, “Comparative fit indexes in structural models,” Psychological Bulletin, vol. 107, no. 2, pp. 238–246, 1990.
View at: Publisher Site | Google Scholar
R. P. McDonald and H. W. Marsh, “Choosing a multivariate model: noncentrality and goodness of fit,” Psychological Bulletin, vol. 107, no. 2, pp. 247–255, 1990.
View at: Publisher Site | Google Scholar
C. Yang, Y. Shao, J. Zhou, Y. Wang, L. Wang, and P. Du, “Mortgage behavior analysis of family farm land management right based on SEM model,” in Fifth International Conference on Economic and Business Management, China, 2020).
View at: Publisher Site | Google Scholar
F. Dai, P. Huang, X. Xu, L. Qi, and M. R. Khosravi, “Spatio-temporal deep learning framework for traffic speed forecasting in IoT,” IEEE Internet of Things Magazine, vol. 3, no. 4, pp. 66–69, 2020.
View at: Publisher Site | Google Scholar
Y. N. Malek, M. Najib, M. Bakhouya, and M. Essaaidi, “Multivariate deep learning approach for electric vehicle speed forecasting,” Big Data Mining and Analytics, vol. 4, no. 1, pp. 56–64, 2021.
View at: Publisher Site | Google Scholar
J. Zhou, K. Cao, X. Zhou, M. Chen, T. Wei, and H. Shiyan, “Throughput-conscious energy allocation and reliability-aware task assignment for renewable powered in-situ server systems,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 41, no. 3, pp. 516–529, 2022.
View at: Publisher Site | Google Scholar
J. Zhou, J. Sun, M. Zhang, and Y. Ma, “Dependable scheduling for real-time workflows on cyber-physical cloud systems,” IEEE Transactions on Industrial Informatics, vol. 17, no. 11, pp. 7820–7829, 2021.
View at: Publisher Site | Google Scholar
F. Wang, M. Zhu, and M. Wang, “6G-enabled short-term forecasting for large-scale traffic flow in massive IoT based on time-aware locality-sensitive hashing,” IEEE Internet of Things Journal, vol. 8, no. 7, pp. 5321–5331, 2021.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2022 Tianxiao Wang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies