Chongqing Jiaotong University

This paper proposes a convolutional neural network (CNN)-based method that learns traffic as images and predicts large-scale, network-wide traffic speed with a high accuracy. Spatiotemporal traffic dynamics are converted to images describing the time and space relations of traffic flow via a two-dimensional time-space matrix. A CNN is applied to the image following two consecutive steps: abstract traffic feature extraction and network-wide traffic speed prediction. The effectiveness of the proposed method is evaluated by taking two real-world transportation networks, the second ring road and north-east transportation network in Beijing, as examples, and comparing the method with four prevailing algorithms, namely, ordinary least squares, k-nearest neighbors, artificial neural network, and random forest, and three deep learning architectures, namely, stacked autoencoder, recurrent neural network, and long-short-term memory network. The results show that the proposed method outperforms other algorithms by an average accuracy improvement of 42.91% within an acceptable execution time. The CNN can train the model in a reasonable time and, thus, is suitable for large-scale transportation networks.

Train–track–bridge dynamic interaction is a fundamental concern in the field of railway engineering, which plays an extremely important role in the optimal design of railway bridges, especially in high-speed railways and heavy-haul railways. This paper systematically presents a state-of-the-art review of train–track–bridge dynamic interaction. The evolution process of train–bridge dynamic interaction model is described briefly, from the simplest moving constant force model to the sophisticated train–track–bridge dynamic interaction model (TTBDIM). The modelling methodology of the key elements in the TTBDIM is systematically reviewed, including the train, the track, the bridge, the wheel–rail contact, the track–bridge interaction, the system excitation and the solution algorithm. The significance of detailed track modelling in the whole system is highlighted. The experimental research and filed test focusing on modelling validation, safety assessment and long-term performance investigation of the train–track–bridge system are briefly presented. The practical applications of train–track–bridge dynamic interaction theory are comprehensively discussed in terms of the system dynamic performance evaluation, the system safety assessment and train-induced environmental vibration and noise prediction. The guidance is provided on further improvement of the train–track–bridge dynamic interaction model and the challenging research topics in the future.

Based on the panel data of 277 cities in China from 2011 to 2018, this paper constructs the digital economy index and the green economy efficiency index. The research found the following: first, the digital economy has significantly improved the efficiency of the green economy in the region. Second, the digital economy has a greater impact on the efficiency of the green economy in the eastern region and large cities than in the central and western regions and small cities. Third, technological innovation is an important way for digital economy to improve the efficiency level of green economy.

Since the last decade, several complex-valued neural networks have been developed and applied in various research areas. As an extension of real-valued recurrent neural networks, complex-valued recurrent neural networks use complex-valued states, connection weights, or activation functions with much more complicated properties than real-valued ones. This paper presents several sufficient conditions derived to ascertain the existence of unique equilibrium, global asymptotic stability, and global exponential stability of delayed complex-valued recurrent neural networks with two classes of complex-valued activation functions. Simulation results of three numerical examples are also delineated to substantiate the effectiveness of the theoretical results.

Abstract This study investigates the mediating effects of environmental and operational performance on the relationship between green supply chain management (GSCM) and financial performance. The proposed relationships are analyzed using survey data from a sample of 126 automobile manufacturers in China. The results suggest that GSCM as an integral supply chain strategy is significantly and positively associated with both environmental and operational performance, which then indirectly leads to improved financial performance. The results indicate the possible complementarity effects between various internal and external GSCM practices.

Abstract Resident tissue macrophages (RTM) can fulfill various tasks during development, homeostasis, inflammation and repair. In the lung, non-alveolar RTM, called interstitial macrophages (IM), importantly contribute to tissue homeostasis but remain little characterized. Here we show, using single-cell RNA-sequencing (scRNA-seq), two phenotypically distinct subpopulations of long-lived monocyte-derived IM, i.e. CD206 + and CD206 − IM, as well as a discrete population of extravasating CD64 + CD16.2 + monocytes. CD206 + IM are peribronchial self-maintaining RTM that constitutively produce high levels of chemokines and immunosuppressive cytokines. Conversely, CD206 − IM preferentially populate the alveolar interstitium and exhibit features of antigen-presenting cells. In addition, our data support that CD64 + CD16.2 + monocytes arise from intravascular Ly-6C lo patrolling monocytes that enter the tissue at steady-state to become putative precursors of CD206 − IM. This study expands our knowledge about the complexity of lung IM and reveals an ontogenic pathway for one IM subset, an important step for elaborating future macrophage-targeted therapies.

Administration of drugs via the buccal route has attracted much attention in recent years. However, developing systems with satisfactory adhesion under wet conditions and adequate drug bioavailability still remains a challenge. Here, we propose a mussel-inspired mucoadhesive film. Ex vivo models show that this film can achieve strong adhesion to wet buccal tissues (up to 38.72 ± 10.94 kPa). We also demonstrate that the adhesion mechanism of this film relies on both physical association and covalent bonding between the film and mucus. Additionally, the film with incorporated polydopamine nanoparticles shows superior advantages for transport across the mucosal barrier, with improved drug bioavailability (~3.5-fold greater than observed with oral delivery) and therapeutic efficacy in oral mucositis models (~6.0-fold improvement in wound closure at day 5 compared with that observed with no treatment). We anticipate that this platform might aid the development of tissue adhesives and inspire the design of nanoparticle-based buccal delivery systems.

In this paper, the exponential stability problem is investigated for a class of Cohen–Grossberg-type bidirectional associative memory neural networks with time-varying delays. By using the analysis method, inequality technique and the properties of an M-matrix, several novel sufficient conditions ensuring the existence, uniqueness and global exponential stability of the equilibrium point are derived. Moreover, the exponential convergence rate is estimated. The obtained results are less restrictive than those given in the earlier literature, and the boundedness and differentiability of the activation functions and differentiability of the time-varying delays are removed. Two examples with their simulations are given to show the effectiveness of the obtained results.

Purpose – The purpose of this paper is to extend previous green supply chain management (GSCM) research by developing and empirically testing a conceptual framework that investigates the relationships between three dimensions of integrated green supply chain management (iGSCM) and multiple dimensions of operational performance. Design/methodology/approach – The study is based on survey data collected from 126 automotive manufacturers in China. The relationships between theoretical constructs are analysed using structural equation modelling. Findings – This study generates important findings of the significant and positive relationships between iGSCM (internal GSCM, GSCM with customers and GSCM with suppliers) and operational performance in terms of flexibility, delivery, quality and cost. Practical implications – It is important for managers to simultaneously consider internal GSCM and GSCM with customers and suppliers when implementing environmental sustainability in the supply chains. Overlooking either internal GSCM or external GSCM may hinder their efforts to improve operational performance. Originality/value – This study contributes to the literature by defining iGSCM that combines three main dimensions, namely, internal GSCM, GSCM with customers and GSCM with suppliers, and empirically testing its impact on multiple operational performance dimensions.

Real-time object detection is significant for industrial and research fields. On edge devices, a giant model is difficult to achieve the real-time detecting requirement and a lightweight model built from a large number of the depth-wise separable convolutional could not achieve the sufficient accuracy. We introduce a new lightweight convolutional technique, GSConv, to lighten the model but maintain the accuracy. The GSConv accomplishes an excellent trade-off between the accuracy and speed. Furthermore, we provide a design suggestion based on the GSConv, Slim-Neck (SNs), to achieve a higher computational cost-effectiveness of the real-time detectors. The effectiveness of the SNs was robustly demonstrated in over twenty sets comparative experiments. In particular, the real-time detectors of ameliorated by the SNs obtain the state-of-the-art (70.9% AP50 for the SODA10M at a speed of ~ 100FPS on a Tesla T4) compared with the baselines. Code is available at https://github.com/alanli1997/slim-neck-by-gsconv

Identifying and quantifying the influential factors on incident clearance time can benefit incident management for accident causal analysis and prediction, and consequently mitigate the impact of non-recurrent congestion. Traditional incident clearance time studies rely on either statistical models with rigorous assumptions or artificial intelligence (AI) approaches with poor interpretability. This paper proposes a novel method, gradient boosting decision trees (GBDTs), to predict the nonlinear and imbalanced incident clearance time based on different types of explanatory variables. The GBDT inherits both the advantages of statistical models and AI approaches, and can identify the complex and nonlinear relationship while computing the relative importance among variables. One-year crash data from Washington state, USA, incident tracking system are used to demonstrate the effectiveness of GBDT method. Based on the distribution of incident clearance time, two groups are categorized for prediction with a 15-min threshold. A comparative study confirms that the GBDT method is significantly superior to other algorithms for incidents with both short and long clearance times. In addition, incident response time is found to be the greatest contributor to short clearance time with more than 41% relative importance, while traffic volume generates the second greatest impact on incident clearance time with relative importance of 27.34% and 19.56%, respectively.

High flammability of polymers has become a major issue which has restricted its applications. Recently, highly crystalline materials and metal-organic frameworks (MOFs), which consisted of metal ions and organic linkers, have been intensively employed as novel fire retardants (FRs) for a variety of polymers (MOF/polymer). The MOFs possessed abundant transition metal species, fire-retardant elements and potential carbon source accompanied with the facile tuning of the structure and property, making MOF, its derivatives and MOF hybrids promising for fire retardancy research. The recent progress and strategies to prepare MOF-based FRs are emphasized and summarized. The fire retardancy mechanisms of MOF/polymer composites are explained, which may guide the future design for efficient MOF-based FRs. Finally, the challenges and prospects related to different MOF-based FRs are also discussed and aim to provide a fast and holistic overview, which is beneficial for researchers to quickly get up to speed with the latest development in this field.

Vibration signals under the same health state often have large differences due to changes in operating conditions. Likewise, the differences among vibration signals under different health states can be small under some operating conditions. Traditional deep learning methods apply fixed nonlinear transformations to all the input signals, which have a negative impact on the discriminative feature learning ability, i.e., projecting the intraclass signals into the same region and the interclass signals into distant regions. Aiming at this issue, this article develops a new activation function, i.e., adaptively parametric rectifier linear units, and inserts the activation function into deep residual networks to improve the feature learning ability, so that each input signal is trained to have its own set of nonlinear transformations. To be specific, a subnetwork is inserted as an embedded module to learn slopes to be used in the nonlinear transformation. The slopes are dependent on the input signal, and thereby the developed method has more flexible nonlinear transformations than the traditional deep learning methods. Finally, the improved performance of the developed method in learning discriminative features has been validated through fault diagnosis applications.

Abstract We investigate the shadows and photon spheres of the four-dimensional Gauss–Bonnet black hole with the static and infalling spherical accretions. We show that, for both cases, there always exist shadows and photon spheres. The radii of the shadows and photon spheres are independent of the profiles of accretion for a fixed Gauss–Bonnet constant, implying that the shadow is a signature of the spacetime geometry and it is hardly influenced by accretion. Because of the Doppler effect, the shadows of the infalling accretion are found to be darker than in the static case. We also investigate the effect of the Gauss–Bonnet constant on the shadow and photon spheres, and we find that the larger the Gauss–Bonnet constant is, the smaller the radii of the shadow and photon spheres will be. In particular, the observed specific intensity increases as the Gauss–Bonnet constant grows.

This paper addresses the multistability issue for quaternion-valued neural networks (QVNNs) with time delays. By using the inequality technique, sufficient conditions are proposed for the boundedness and the global attractivity of delayed QVNNs. Based on the geometrical properties of the activation functions, several criteria are obtained to ensure the existence of equilibrium points, of which are locally stable. Two numerical examples are provided to illustrate the effectiveness of the obtained results.

Sustainable and resilient pavement infrastructure is critical for current economic and environmental challenges. In the past 10 years, the pavement infrastructure strongly supports the rapid development of the global social economy. New theories, new methods, new technologies and new materials related to pavement engineering are emerging. Deterioration of pavement infrastructure is a typical multi-physics problem. Because of actual coupled behaviors of traffic and environmental conditions, predictions of pavement service life become more and more complicated and require a deep knowledge of pavement material analysis. In order to summarize the current and determine the future research of pavement engineering, Journal of Traffic and Transportation Engineering (English Edition) has launched a review paper on the topic of “New innovations in pavement materials and engineering: A review on pavement engineering research 2021”. Based on the joint-effort of 43 scholars from 24 well-known universities in highway engineering, this review paper systematically analyzes the research status and future development direction of 5 major fields of pavement engineering in the world. The content includes asphalt binder performance and modeling, mixture performance and modeling of pavement materials, multi-scale mechanics, green and sustainable pavement, and intelligent pavement. Overall, this review paper is able to provide references and insights for researchers and engineers in the field of pavement engineering.

In this paper, the boundedness and complete stability of complex-valued neural networks (CVNNs) with time delay are studied. Some conditions to guarantee the boundedness of the CVNNs are derived using local inhibition. Moreover, under the boundedness conditions, a compact set that globally attracts all the trajectories of the network is also given. Additionally, several conditions in terms of real-valued linear matrix inequalities (LMIs) for complete stability of the CVNNs are established via the energy minimization method and the approach that converts the complex-valued LMIs to real-valued ones. Examples with simulation results are given to show the effectiveness of the theoretical analysis.

Abstract This study explores the links of implementing customer‐centric green supply chain management (GSCM) with its antecedent factors (i.e. customer pressure) and performance outcomes (i.e. operational performance and customer satisfaction). Data for this study were obtained through a survey of 126 automobile manufacturers in China. Results suggest that customer pressure has a positive effect on the implementation of customer‐centric GSCM, which, in turn, leads to multiple operational performance improvements (i.e. flexibility, delivery, quality and cost). While production flexibility and cost appear to have no significant impact on customer satisfaction, product quality and delivery are significantly and positively associated with customer satisfaction. On the practical front, this paper provides guidelines for managers in implementing customer‐centric GSCM to respond to customer pressures and improve firm performance, and for policy‐makers to encourage partner‐focused GSCM efforts in environmental policy. Copyright © 2014 John Wiley & Sons, Ltd and ERP Environment

Abstract We investigate the effects of quintessence dark energy on the shadows of black hole, surrounded by various profiles of accretions. For the thin-disk accretion, the images of the black hole comprises the dark region and bright region, including direct emission, lensing rings and photon rings. Although their details depend on the form of the emission, generically, direct emission plays a major role for the observed brightness of the black hole, while the lensing ring makes a small contribution and the photon ring makes a negligible contribution. The existence of a cosmological horizon also plays an important role in the shadows, since the observer in the domain of outer communications is near the cosmological horizon. For spherically symmetric accretion, static and infalling matters are considered. We find that the positions of photon spheres are the same for both static and infalling accretions. However, the observed specific intensity of the image for infalling accretion is darker than for static accretion, due to the Doppler effect of the infalling motion.

A phosphorus abundant biomass phytic acid (PA)-functionalized metal–organic framework (MOF) UiO-66-NH2 (PA-UiO66-NH2) was synthesized successfully as a novel fire retardant (FR) to reduce fire hazards and suppress smoke for the epoxy (EP) resin. The complexing and salt formation reaction between phosphate groups and both Zr species or amine groups on UiO-66-NH2 were verified by X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy. The EP/5PA-UiO66-NH2 sample exhibited excellent fire retardancy and toxic gases suppression property with a 41% decrease in the peak heat release rate and a 42% reduction in total smoke production. The slightly modified thermal degradation path of the EP/5PA-UiO66-NH2 sample was evidenced by means of pyrolysis gas chromatography–mass spectrometry, which led to the reinforced char with a strong barrier and a higher polyaromatic structure. Moreover, the amine and phosphorus groups may react with EP, thus providing a better interfacial strength between FRs and EP matrix. Therefore, the enhanced mechanical property was also observed by dynamic mechanical analysis with a light increment of storage modulus (8%). In perspective, functionalization of MOFs to modify the thermal decomposition of FR may provide a possible way for MOFs to become efficient FRs.