A new mixed traffic flow involving human-driven vehicles (HDVs) and connected and automated vehicles (CAVs) has emerged as a result of developments in autonomous driving and wireless communication. Deep reinforcement learning (DRL) is a promising method for addressing complex traffic flow control issues. However, the majority of DRL-based studies concentrated on optimizing either traffic signals or vehicle trajectories, often overlooking the interaction between these two elements. In this paper, we propose a cooperative control approach for traffic signals and vehicle trajectories using a multi-agent actor-critic (CCTV-MAC) in a mixed traffic flow environment. Our approach incorporates traffic signal control (TSC) agents that optimize signal cycles and green ratios by extracting trajectory features from HDVs and CAVs, as well as vehicle trajectory control (VTC) agents that optimize CAV trajectories based on information from the self-vehicle, the preceding vehicle, and the approaching signal timing provided by the TSC agents. A transformer-based spatio-temporal attention network and a reward-sharing mechanism are incorporated to enhance coordinated decision-making among adjacent TSC agents. Additionally, a vehicle-road-cloud integration structure (VRCIS) is designed for CCTV-MAC, where model training tasks are deployed on the cloud side to accelerate the learning process via parallel computing formulas and algorithms, while decision-making tasks are executed on the vehicle side and roadside to minimize response latency. With the support of VRCIS, cooperative control of TSC and VTC agents is achieved through effective information exchange. Simulations on real-world road networks show that our proposed CCTV-MAC outperforms traditional and other DRL methods.

In recent years, with a remarkable increase in urban rail transit operations, the issue of energy efficiency in train operations has attained increasing attention. In this study, a two-stage optimization model is proposed to optimize driving strategies and schedules. We comprehensively consider the optimization of train running curves, running time allocations to a whole line, and utilization of regenerative braking energy, to reduce the net energy consumption of train operations. In the first stage, a multi-objective jellyfish search (MOJS) optimization algorithm is used to optimize a switching sequence at each inter-station, and Pareto fronts are obtained corresponding to energy-saving train running curves. In the second stage, a grey wolf optimizer (GWO) is adopted to optimize running times between adjacent stations, dwell times at stations, and headway time. This stage aims to coordinate the operations of multiple trains, to reduce the traction energy consumption of a whole line, and to increase the utilization of regenerative braking energy. The optimality is discussed for the proposed two-stage optimization processes. Numerical experiments are conducted based on train and infrastructure data of the Beijing Yizhuang metro line. The results show that the proposed optimization model and solution algorithms have a considerable energy-saving effect.

As a major use of electricity, the energy efficiency of urban railways is of great concern. To reduce the operational energy consumption of a metro line, this paper proposes a two-stage optimization model for energy-efficient train operations based on three aspects: train operation strategies at inter-stations, running time allocation within an operating cycle, and utilization of regenerative braking energy of multiple collaborative trains in the same power supply area. The first stage is based on a multi-objective optimization algorithm, which optimizes the position-speed curves at each inter-station to obtain a series of energy-saving running curves and the corresponding Pareto fronts. In the second stage, a single-objective optimization algorithm and parallel computing method are used to solve the integrated optimization model for train operations with the minimum net energy consumption as the objective function, to reduce the net energy consumption in an operating cycle. Finally, numerical experiments are carried out with data on the Beijing Yizhuang line. The results show that the proposed bi-level programming model and solution methods have a good energy-saving effect.

The escalating air pollution resulting from traffic congestion has necessitated a shift in traffic control strategies towards green and low-carbon objectives. In this study, a graph convolutional network and self-attention value decomposition-based multi-agent actor-critic (GSAVD-MAC) approach is proposed to cooperative control traffic network flow, where vehicle carbon emission and traffic efficiency are considered as reward functions to minimize carbon emissions and traffic congestions. In this method, we design a local coordination mechanism based on graph convolutional network to guide the multi-agent decision-making process by extracting spatial topology and traffic flow characteristics between adjacent intersections. This enables distributed agents to make low-carbon decisions which not only account for their own interactions with the environment but also consider local cooperation with neighboring agents. Further, we design a global coordination mechanism based on self-attention value decomposition to guide multi-agent learning process by assigning various weights to distributed agents with respect to their contribution degrees. This enables distributed agents to learn a globally optimal low-carbon control strategy in a cooperative and adaptive manner. In addition, we design a cloud computing-based parallel optimization algorithm for the GSAVD-MAC model to reduce calculation time costs. Simulation experiments based on real road networks have verified the advantages of the proposed method in terms of computational efficiency and control performance.

When dealing with traffic big data under the background of Internet of Things (IoT), traffic control under the single-machine computing environment is difficult to adapt to the massive and rapid analysis and decision-making. To tackle this problem, we propose a parallel computing approach of traffic network flow control based on the mechanism of model predictive control (MPC). A non-analytical rule-based traffic flow model is developed to forecast vehicle movements in the prediction horizon according to the real-time feedback information of traffic flow, and evaluate the performances of candidate control strategies. Furthermore, to accelerate the solution process of obtaining the optimal control schemes in the prediction horizon, a two-level hierarchical parallel genetic algorithm (HPGA) based on Spark cloud computing is designed. Through the parallel computing architecture, the computationally intensive optimization tasks are decomposed into multiple parallel sub-tasks with the aid of resilient distributed datasets (RDDs), which improves the computational efficiency. The simulation results demonstrate the validity of the proposed methodology for traffic network flow predictive control with respect to unsaturated and oversaturated traffic scenarios. The Spark-based parallel optimization approach has the potential to satisfy the computing requirements of online optimization when dealing with the big data of traffic network flow control while keeps favorable control performances.

Traffic flow in a road network is mutually interactive and interdependent with each other. It is challenging to describe the dynamics of traffic network flow by using analytical methods. In this article, the deep convolutional neural network (DCNN) model is employed to address traffic network flow prediction. To improve the parameter learning efficiency confronting traffic big data, a parallel training approach is developed for the DCNN prediction model. The theoretical foundation is developed for the parallel training algorithm of the DCNN model. A master-slave parallel computing solution for traffic network flow prediction is implemented on the Spark cloud. Real data of traffic network flow are applied to verify the effectiveness of the DCNN prediction model and the parallel training algorithm. The experimental results demonstrate that the DCNN prediction model for traffic network flow outperforms the typical prediction models based on backpropagation neural networks, support vector regressions, radial basis functions, and decision tree regressions. The proposed parallel training method can improve the training efficiency and obtain global features of the entire dataset from local learning with regard to the respective data subsets.

Based on the life cycle of business process reengineering (BPR), BPR is decomposed into business reengineering (BR) at the strategic and tactical levels and process reengineering at the operational level, including process structural and parametric optimization, respectively. Quality-function-deployment-centered integrated tools are proposed to support the rational analysis and decision making in BR. The analytic hierarchy process is utilized to evaluate the intangible attributes. Questionnaire and stochastic simulation are used to determine the strategic goals. Correlation analysis is employed to appraise the effects of each reengineering objective on customer requirements and other reengineering objectives. The slack-adjusted assurance region data envelopment analysis model is developed to estimate the performance levels of benchmarked organizations and to predict the possible investment and incurred performances. Finally, the decision of reengineering objectives and their performance levels in a specified phase is made through the 0–1 integer programming model considering cost-effective predictive results, which brings about the formation of tactical goals. The proposed tools provide structured approaches to set strategic and tactical goals, and organically link them. Thus, they can provide a powerful support for aligning business strategies, information technology tactics, and business processes of reengineering organization with customer requirements. An illustrative example demonstrates the application of the systematic analysis and decision-making methodology supported by the proposed analytic tools.

The development of modern information technology makes it possible to timely and actively navigate traffic flow to avoid congestion. This paper outlines the architecture of the system of mobile transportation information service, and some critical techniques to realize the system are elaborated such as mobile positioning, XML-based information organization, traffic-flow information collection and processing, grid computing, flow media service and Web services, etc. The route navigation based on predictive control is proposed, and the realization study of route navigation demonstrates how the critical techniques are holistically utilized.

Business performance prediction draws a vision that accounts for the measures of business process improvement through benchmarking which is to find best practices in an industry or organization and suggest solutions for business process improvement. Data envelopment analysis (DEA), a nonparametric mathematical programming, is utilized for business performance prediction that gives the best benchmark of business process improvement. The DEA-based benchmarking procedure and SA (slacks-adjusted)-AR (assurance region)-DEA model-based performance predictive design models are developed in this paper. The predictive design method considering business constraints on efficient frontiers of decision-making units (DMUs) is a kind of quantitative business knowledge discovery based on historical business data that synthetically takes both successful and impolitic business practices of various organizations or within an organization into consideration.

The quality of business process is on one hand determined by the structure of business process chain, and on the other hand the configuration of resources and organization. This paper mainly discusses how to appropriately assign resources according to the historical information of resources in the execution of activities so as to ensure the high performances of business processes. The quality of resource assignment is described by the average approximation degree to the ideal resource assignment based on the TOPSIS method and the uniformity degree of resource assignment based on the concept of information entropy, which can be regarded as the predictive quality of business process. The multi-objective evaluation-combined optimization models of business processes have been developed of intra- and inter-enterprise business processes. The mixed non-dominated sorting genetic algorithm (NSGA) is utilized to solve the multi-objective optimization problem.