Traffic Process Research Articles

Time-oriented interactive process miner: a new approach for time prediction

Everyday information systems collect a different kind of process instances of a business flow. As time goes on, the size of the collected data builds up speedily and constitutes a huge amount of data. It is a very challenging task to obtain valuable information and features of processes from such big data. Considering in advance, the trend and different features of the ongoing process are essential. Especially, time management is crucial in designing and conducting business processes. In this article, a novel process miner algorithm is proposed for time prediction, named time-oriented İnteractive process miner (T-IPM), which predicts the remaining and completion time of each process in a business workflow. The goal is to develop a process miner algorithm for time prediction that is able to work on a huge amount of event logs, and to incorporate the execution records of ongoing processes into the discovered process model instantly. In this study, three datasets from different domains that include event logs of repair, hospital, and traffic process flows were used to carry out the experimental studies in order to show the effectiveness of the proposed time prediction algorithm. The experimental studies show that the proposed algorithm is capable of performing process analysis on a huge amount of event logs with low memory consumption and high prediction accuracy.

Adaptive DRX Method for MTC Device Energy Saving by Using a Machine Learning Algorithm in an MEC Framework

Machine-type communication (MTC) is recognized as an enabling building block for constructing the Internet of Things (IoT) through cellular networks. Since MTC devices (MTCDs) are usually powered by batteries, reducing the energy consumption of an MTCD is an important requirement in the design of an IoT system. The third-generation partnership project (3GPP) specified discontinuous reception (DRX) to save energy in user equipment (UE) by making a UE device sleep whenever it is not involved in data transmission or reception. However, UE is used for human-type communications (HTCs), and MTCs are different from HTCs in many ways, namely, in terms of traffic arrival patterns, mobility, and information priority. Thus, using the DRX developed for UE to save the energy of an MTCD leads to inefficiency in terms of the amount of power saved and the data transmission delay. In the literature, many efforts have been made to dynamically vary the DRX cycle based on traffic activities to reduce the energy consumption of an MTCD. These assume a specific distribution for the traffic arrival process and predict the next packet arrival time to control the DRX cycle. However, since the characteristics of the MTC traffic arrival pattern vary widely according to the types of applications and the network status, previous approaches are limited in the achievable energy efficiency and the range of MTC applications that they can support. To address these issues, we devise adaptive DRX (ADRX) by adopting a machine learning algorithm called bank-of-experts. Without making any assumption regarding the underlying traffic process, ADRX predicts the next packet arrival time using only the observed packet arrival times. Using the predicted next packet arrival time, ADRX dynamically adjusts the sleep interval. In terms of implementation, ADRX is executed in an eNB by following the Multi-access Edge Computing (MEC) framework. Thus, an MTCD avoids consuming the energy required to execute a sophisticated energy-saving algorithm while obtaining the benefits provided by ADRX. Since different IoT services have different requirements for the amount of saved energy and data transmission delay, we design a loss function by combining these two performance metrics so that ADRX can support a wide range of application services. We first analyze the bound of the cumulative regrets of ADRX, and then through simulations, we compare the performance of ADRX with the standard DRX with optimal sleep duration (OSDRX) and a generalized and autonomous DRX (GDRX) scheme. The results show that ADRX outperforms GDRX in terms of the loss function and has almost the same performance as OSDRX.

Traffic modeling for communications networks: A multifractal approach based on few parameters

In this work, in order to model network traffic processes, an efficient multifractal model is proposed based on fewer parameters than other models present in the literature, called 2PEMV. The 2PEMV model (2-Parameter Exponential Modeling for Multipliers’ Variance) is based on a multiplicative cascade in the wavelet domain capable of synthesizing communication network traffic traces which present characteristics such as self-similarity and wide Multifractal Spectrum Width (MSW). For such a purpose, in the 2PEMV model, the energy decay of the wavelet coefficients is captured by means of an exponential modeling for the multipliers’ variance along the cascade scales. The performance of the 2PEMV model to represent the network traffic characteristics is evaluated in comparison to other models present in the literature through simulations that are carried out using real communication network traffic traces.

Understanding and Modeling Urban Mobility Dynamics via Disentangled Representation Learning

Understanding the underlying patterns of the urban mobility dynamics is essential for both the traffic state estimation and management of urban facilities and services. Due to the coupling relationship of generative factors in spatial-temporal domain, it is challenging to model the citywide traffic dynamics under a structural pattern of critical features such as hours of days, days of weeks and weather conditions. To address this challenge, this article develops a disentangled representation learning framework to learn an interpretable factorized representation of the independent data generative factors. In order to make full use of the knowledge on generative factors, this article proposes spatial-temporal generative adversarial network (ST-GAN) to assign the generative factors of traffic flow to the feature vector in latent space and reconstructs the high-dimensional citywide traffic flow from the given factors. With the help of the disentangled representations, the decomposed feature vector in latent space discloses the relationship between underlying patterns and citywide traffic dynamics. Several comprehensively experiments show that ST-GAN not only effectively improves the prediction accuracy but also promisingly characterize structural properties of the traffic evolution process.

Simultaneous Incomplete Traffic Data Imputation and Similarity Pattern Discovery with Bayesian Nonparametric Tensor Decomposition

A crucial task in traffic data analysis is similarity pattern discovery, which is of great importance to urban mobility understanding and traffic management. Recently, a wide range of methods for similarities discovery have been proposed and the basic assumption of them is that traffic data is complete. However, missing data problem is inevitable in traffic data collection process due to a variety of reasons. In this paper, we propose the Bayesian nonparametric tensor decomposition (BNPTD) to achieve incomplete traffic data imputation and similarity pattern discovery simultaneously. BNPTD is a hierarchical probabilistic model, which is comprised of Bayesian tensor decomposition and Dirichlet process mixture model. Furthermore, we develop an efficient variational inference algorithm to learn the model. Extensive experiments were conducted on a smart card dataset collected in Guangzhou, China, demonstrating the effectiveness of our methods. It should be noted that the proposed BNPTD is universal and can also be applied to other spatiotemporal traffic data.

Explosive synchronization induced by traffic processes in complex networks

An abundance of studies on synchronization focus on the characteristics of synchronization itself, but ignore the guiding role of information transfer between dynamic units on the synchronization process. In reality, continuous interactions between dynamic units are costly and sometimes unnecessary. Here we perform both traffic and synchronization processes in the same network and propose a model based on a constant-density traffic. Each node delivers packets with preference according to frequency mismatch and degree, and its synchronization behavior is only determined by the received packets. Explosive synchronization (ES) can be induced in both heterogeneous and homogeneous networks, as well as in both symmetric and asymmetric frequency distributions. We have concluded that the correlation between the average reception degree and the natural frequency is a key feature for the emergence of ES. Furthermore, at different packet densities, the forward and backward critical couplings and the width of the hysteretic loop in ES all exhibit nonmonotonic behavior and have minimum values. ES can still appear even if the packet density is very low.

Influence of channel reservation on the handover calls flow in mobile network

Abstract In this paper the properties of handover traffic are analyzed in the system with channel reservation. The reservation of several traffic channels only for handover calls leads to probability of handover calls congestion decrease to the level of per mill or even lower. The congestion of all traffic channels in the neighbouring cells is the cause that handover traffic process deviates from the Poissonian distribution and this deviation is negligible, because handover calls congestion probability is very low. This low handover calls probability is also the reason why implementation of one equivalent cell instead of 6 neighbouring cells surrounding the central cell does not model satisfactory the traffic flow in the mobile network. The randomness of handover calls arrival to the central cell is tested by our originally developed simulation program. Coefficient of variation of time between handover calls arrival to the central cell is calculated as the result of simulation. We introduced the ratio of the coefficient of variation of the time between new handover calls appearance in the central cell to the value of this coefficient for surrounding cell as the measure of handover calls randomness: the nearer this ration to 1, the more random is handover calls arrival to the central cell.

Effect of Manual Traffic Control on Evacuation Time Estimates

Manual traffic control (MTC) is a common intersection control strategy where trained personnel, typically police law enforcement officers, allocate intersection right-of-way to approaching vehicles. MTC is frequently used for special events and during emergencies. However, the current state-of-the-practice has shown little research that quantifies or assesses the benefits of MTC during evacuations. This paper describes research to develop microscopic traffic simulation models that were used to assess traffic processes of emergency evacuations of nuclear power plant sites involving MTC. A recently developed model to represent MTC was integrated into the simulation to quantify its impact on clearance time and other operational parameters. The results of this effort showed that in rural or less congested urban areas clearance times experienced no significant benefit from the application of MTC. Conversely, in densely populated regions clearance times increased significantly when MTC was deployed. This suggested that for congested urban areas, with closely spaced intersections, the characteristically long cycle lengths associated with MTC resulted in significant queues. These queues propagated upstream, interfering with traffic operations at neighboring intersections and bottleneck points within the network. This, in turn, triggered even more queues and, ultimately, led to localized gridlock. From the perspective of police officers, longer cycle lengths result in fewer phase changes and less lost-time at the intersection. As such, the officer is incentivized to extend green times and cycle lengths as long as possible. What the officer cannot see, however, is the effect of such actions on queue formation at upstream intersections.

Enhancement of guinea pig cytomegalovirus infection by two endogenously expressed components of the pentameric glycoprotein complex in epithelial cells

A better understanding of the mechanisms underlying cell tropisms and the efficiency of viral infection is critical for the development of vaccines and antiviral drugs for viral diseases. In this study, we worked on the entry mechanisms of guinea pig cytomegalovirus and found that endogenous expression of a combination of two components (GP131 and GP133) of the pentameric glycoprotein complex, which is required for non-fibroblast cell tropisms, enhanced viral infection more than 10-fold. In addition, D138A alteration in GP131 increased this enhancement by an additional 10-fold. Although differences in the efficiency of viral infection among various cell types are usually explained by differences in viral entry or traffic processes, our experimental evidences dismissed such possibilities. Instead, our findings that i) endogenous expression of GP131 and GP133 after nuclear delivery of viral DNA still enhanced infection and ii) an HDAC inhibitor overcame the need of the endogenous expression led us to hypothesize a novel mechanism that controls the efficiency of viral infection through the activation of gene expression from viral DNA delivered to the nuclei. Further studies of this unexpected phenomena warrant to understand novel but also general mechanisms for cell tropisms of viral infection and determinants that control infection efficiency.

RNA Interference Screening Identifies Novel Roles for RhoBTB1 and RhoBTB3 in Membrane Trafficking Events in Mammalian Cells.

In the endomembrane system of mammalian cells, membrane traffic processes require a high degree of regulation in order to ensure their specificity. The range of molecules that participate in trafficking events is truly vast, and much attention to date has been given to the Rab family of small GTPases. However, in recent years, a role in membrane traffic for members of the Rho GTPase family, in particular Cdc42, has emerged. This prompted us to develop and apply an image-based high-content screen, initially focussing on the Golgi complex, using RNA interference to systematically perturb each of the 21 Rho family members and assess their importance to the overall organisation of this organelle. Analysis of our data revealed previously unreported roles for two atypical Rho family members, RhoBTB1 and RhoBTB3, in membrane traffic events. We find that depletion of RhoBTB3 affects the morphology of the Golgi complex and causes changes in the trafficking speeds of carriers operating at the interface of the Golgi and endoplasmic reticulum. In addition, RhoBTB3 was found to be present on these carriers. Depletion of RhoBTB1 was also found to cause a disturbance to the Golgi architecture, however, this phenotype seems to be linked to endocytosis and retrograde traffic pathways. RhoBTB1 was found to be associated with early endosomal intermediates, and changes in the levels of RhoBTB1 not only caused profound changes to the organisation and distribution of endosomes and lysosomes, but also resulted in defects in the delivery of two different classes of cargo molecules to downstream compartments. Together, our data reveal new roles for these atypical Rho family members in the endomembrane system.

Selection of a traffic management scheme at an intersection taking into consideration the traffic flow composition

Traffic flows in large cities have a non-uniformed character of origin and changes. The main parameters are often sensitive to the changes in the environmental conditions, specifically, the time of the day, season, etc. The magnitudes that characterize traffic flows have a stochastic nature and therefore they are difficult to predict. According to this, indicators such as intensity, density and traffic speed over time and space are uneven. Taking this into consideration, during the development of traffic management schemes, there arises a necessity to take into consideration additional factors which are of probabilistic nature. This approach provides more opportunities in the traffic management process and makes it possible to use appropriate ways to choose traffic management methods in certain cases. This process involves detailed studies of traffic indicators. The approaches and methods for studying the main indicators of traffic flows were given. They include both field methods and simulation, which involves accounting of mathematical and physical patterns of traffic flows. It also makes it possible to predict the situation when choosing a traffic management scheme with the use of computer equipment. The results of the studies, which were conducted according to the described methods, were analyzed. These results indicate a close relationship between the relative traffic composition for the types of vehicles and the indicators of queuing and delays within the intersections. According to the results, mathematical dependences characterizing this relationship were obtained. It was established that the choice of a road traffic scheme at intersections should be made taking into consideration the traffic flow composition, as each regulation type is effective for a particular case

Short Term Traffic State Prediction via Hyperparameter Optimization Based Classifiers.

Short-term traffic state prediction has become an integral component of an advanced traveler information system (ATIS) in intelligent transportation systems (ITS). Accurate modeling and short-term traffic prediction are quite challenging due to its intricate characteristics, stochastic, and dynamic traffic processes. Existing works in this area follow different modeling approaches that are focused to fit speed, density, or the volume data. However, the accuracy of such modeling approaches has been frequently questioned, thereby traffic state prediction over the short-term from such methods inflicts an overfitting issue. We address this issue to accurately model short-term future traffic state prediction using state-of-the-art models via hyperparameter optimization. To do so, we focused on different machine learning classifiers such as local deep support vector machine (LD-SVM), decision jungles, multi-layers perceptron (MLP), and CN2 rule induction. Moreover, traffic states are evaluated using traffic attributes such as level of service (LOS) horizons and simple if–then rules at different time intervals. Our findings show that hyperparameter optimization via random sweep yielded superior results. The overall prediction performances obtained an average improvement by over 95%, such that the decision jungle and LD-SVM achieved an accuracy of 0.982 and 0.975, respectively. The experimental results show the robustness and superior performances of decision jungles (DJ) over other methods.

A Hybrid Deep Learning-Based Traffic Forecasting Approach Integrating Adjacency Filtering and Frequency Decomposition

Traffic forecasting in urban area has attracted substantial attention in recent years due to its significant assitance for traffic dispatching and trip planning. However, this task is very challenging due to the complex dependencies inherent in the traffic process. In our study, recent and periodic dependencies are identified and used to describe the corresponding near-term and long-term effects of historical traffic data on future traffic states. Subsequently, during the recent dependency modeling, each road is found to correlate with its adjacent roads through traffic flow diffusion, then the correlation intensities are quantified and used to choose strongly correlated roads to build a critical road sequence. While for the periodic dependency modeling, since the historical speed series of a road segment exhibits multi-frequency attributes (i.e., low-frequency daily period and high-frequency stochastic fluctuation), wavelet transform is conducted to decompose the original speed series into low and high-frequency sub-series. On these bases, we propose a hybrid traffic speed forecasting model, flow and wavelet-integrated spatio-temporal network (FW-STN). In the FW-STN, the recent features are captured by the convolutional neural network (CNN) with near-term traffic data from the derived critical road sequence, and the periodic features are captured by the long short-term memory (LSTM) with the low and high-frequency sub-series. Both the recent and periodic features are then fused to conduct the final prediction. Experimental results on real traffic data show that the proposed approach outperforms nine state-of-the-art methods (with improvements of 3% ~15% in mean average percentage error).

Estimation of risks related to the functioning of railway telecommunication

Railway transport is a complex technical system with telecommunication being one of its constituent parts. Technical condition of the objects of the telecommunication division impacts directly on the traffic process. Therefore, to ensure the required indices of the traffic process reliability it is necessary to provide for a pre-set factor for availability of services that are needed for safe control of train traffic. According to the methodology of management of resources, risks and dependability at life cycle stages (URRAN) (Gapanovich et al. in Dependability 2:4–22, 2015. https://doi.org/10.21683/1729-2646-2015-0-2-4-22) implemented on Russian railways, resources are managed in the telecommunication division based on the risk level criterion: if a risk is tolerable then the objects of the telecommunication division ensure the compliance with the traffic process requirements. If risks exceed the tolerable values it is necessary to take measures aimed at the modernization of the objects, or at the improvement of their technical maintenance. This approach is aimed at the optimization of cost of the life cycle of the objects of the telecommunication division. Order of estimation of risks related to the functional faults of telecommunication objects includes the identification of undesirable events that lead to the violations of availability of services of railway telecommunication, and consequently to train delays. Then the frequencies of these undesirable events are estimated, the amount of damage is defined. This data is initial data used to construct matrices of risks of the functional faults of telecommunication objects. The article contains calculation expressions for the construction of risk matrices and the rules of decision making in relation to tolerability (or intolerability) of the assigned risk level. The article also describes the example of estimation of risk related to the functional faults of telecommunication objects based on the statistics of undesirable events held on the Russian railways during several years of observation, with the description of possible measures that can be taken to control each category of the risk.

Distributed Weighted Balanced Control of Traffic Signals for Urban Traffic Congestion

Since urban traffic congestion has become a major problem for big cities in recent years, we propose a distributed control scheme for traffic lights in the network. First, a new criterion called traffic process ability which implies the balance between the traffic demand and traffic capacity of each road is introduced. Moreover, the congestion of a road is mitigated by utilizing the traffic process ability of the neighbors more effectively, where different weights are assigned to roads in each agent according to their importance or the real-time traffic conditions. As only local information is needed, a distributed control scheme in which the road network is divided among several agents is proposed. Furthermore, in order to accelerate the congestion dissipation process, the aggregated state of each agent is introduced into the performance index and balanced with its neighboring agents. The control signals are calculated by agents in a parallel way and the optimization problem is solved iteratively to reach a convergence. Finally, the effectiveness of the proposed control scheme is evaluated by simulation under different scenarios and the performance is compared with the traffic responsive control method SCOOT.

Elements of the model positioning of aircraft on the apron

The design of airports and the organization of their work requires the recognition of the basic components of the air transport process, consisting of an "aerial" part, including the landing phase and the landing operation itself, as well as take-off, the "ground" part, including the task of taxiing aircraft on apron, ground handling tasks, "terminal" part, including passenger handling tasks. These elements form a cause-and-effect sequence, or a series-parallel structure that determines the quality of services provided by the airport, their efficiency, reliability and price. The article presents the issues of decision support for the operation and maintenance of airport infrastructure and traffic management on the ramp and within the airport, i.e. the operation of allocating aircraft to the gates of "gates" using simulation tools. Aircraft taxiing operations on the tarmac integrate the flight phase (along with its components and its problems, such as arriving and departing sequencing) with the ground handling phase of aircraft and passengers at terminals. The model presented in the article is a single element of a holistic approach to the operation of an airport. The overall model consists of the development of decision models for the organization of aircraft traffic on the apron, algorithms for their solution and the possibility of practical application as a simulation tool for analyzing and assessing aircraft traffic processes in the take-off, taxiing and landing phase. To describe the model, a formal mapping of the structure of the necessary airport elements was proposed. A formal record of boundary conditions and criteria relevant to aircraft allocation processes is presented due to the minimization of travel time of passengers transferring between two aircraft assigned to two different gates. Test results can be used in practice, among others by airspace controllers and airport designers for: analyzing and assessing the possibilities of increasing airport capacity, analyzing and assessing the determination of taxiway lengths, maintaining high safety reserves, etc.

Online Power Management Strategies for Energy Harvesting Mobile Networks

The design of self-sustainable base station (BS) deployments is addressed in this paper. We target deployments featuring small BSs with energy harvesting (EH) and storage capabilities. These BSs can use ambient energy to serve the local traffic or store it for later use. A dedicated power packet grid is utilized to transfer energy across them, compensating for imbalance in the harvested energy or in the traffic load. Some BSs are offgrid , i.e., they can only use the locally harvested energy and that transferred from other BSs, whereas others are ongrid , i.e., they can additionally purchase energy from the power grid. Within this setup, an optimization problem is formulated where: harvested energy and traffic processes are estimated (at runtime) at the BSs through Gaussian processes, and a model predictive control framework is devised for the computation of energy allocation and transfer across BSs. The combination of prediction and optimization tools leads to an efficient and online solution that automatically adapts to EH and load dynamics. Numerical results, obtained using real EH and traffic profiles, show substantial improvements with respect to the case where the optimization is carried out without predicting future system dynamics. The main improvements are in the outage probability (zero in most cases), and in the amount of energy purchased from the power grid, that is more than halved for the same served load.

Modelling the spatial distribution of heavy vehicle loads on long-span bridges based on undirected graphical model

Vehicle load modelling is highly important for bridge design and safety evaluation. Conventional modelling approaches for vehicle loads have limitations in characterizing the spatial distribution of vehicles. This article presents a probabilistic method for modelling the spatial distribution of heavy vehicle loads on long-span bridges by using the undirected graphical model (UGM). The bridge deck is divided into grid cells, a UGM with each node corresponding to each cell is employed to model the location distribution of heavy vehicles, by which probabilities of heavy-vehicle distribution patterns can be efficiently calculated through applying the junction tree algorithm. A Bayesian inference method is also developed for updating the location model in consideration of the non-stationarity of traffic process. Gross weights of heavy vehicles are modelled by incorporating additional random variables to the vehicle-location UGM, corresponding probability distributions are constructed conditioned on ignoring correlation and considering correlation, respectively. Case studies using simulated data as well as field monitoring data have been conducted to examine the method. Compared with previous studies involving vehicle load modelling, the presented method can implement probabilistic analysis for all spatial distribution patterns of heavy vehicles on the entire bridge deck.

Assessment of post-disaster reentry traffic in megaregions using agent-based simulation

Transportation personnel, assets, and systems play important roles during emergencies. Of particular note is the critical function of road networks as lifelines for evacuations, then, after such events, to facilitate the rapid inflow of support personnel and resources needed to assess and repair damage. They also provide access for the safe and rapid return of evacuees. Despite the importance of reentry and repopulation, the role of transportation systems for them has been a relatively lightly researched topic. This paper describes research to analyze the traffic processes associated with post-evacuation reentry and its associated travel time and delay. To examine the most extreme of conditions; this effort also focused on repopulation and reentry after mass evacuations in megaregions. Among the findings of this work were that, as expected, reentry traffic processes generally behave similar to, but in the reverse direction, of evacuations. However, contrary to commonly held belief, they are not identically opposite. Rather, network performance during reentry may often be better than that of evacuation. Another broad finding of the research was that the spatial heterogeneity of traffic distribution across available roadways is a primary factor in determining network performance, meaning the more uniform the spread of on all links, the more efficient the network will operate. A third significant finding of this research was also the importance of the temporal spreading of returnee departure times on network efficiency.

Characteristics of the Locations of Increased Road Traffic Incidents in Urban Road Networks in Poland

The results of studies carried out for years by researchers around the world point to man as the most important and – at the same time – the weakest link in traffic processes, and therefore road traffic incidents are caused by hardly anything other than human error. A majority of such errors leading to traffic incidents are made for specific reasons, such as a faulty road environment. The spatial distribution of road traffic incidents is not random – it follows a specific pattern with most incidents concentrating at the ‘weakest’ points of a road network with identifiable shortcomings, making it more difficult for the road users to assess the circumstances and take appropriate action. Points like these are referred to as ‘hazardous locations’. The purpose of this article is to define and describe such locations where specific types of road incidents concentrate at urban road network intersections and intermediate sections between them in Poland. The analysis utilised a modern IT system, WZDR (Road and Traffic Management Support IT System), currently implemented in a few cities in Poland. The most important analytical part of the study contains a detailed description of the relationship between the characteristics representing an increased hazard to road traffic participants in the hazardous locations and incorrect behaviour of the users caused by the considered faults of road infrastructure. This analysis enables identification of typical faults of selected elements of a road network from the point of view of road traffic safety. It provides a basis for improvement of the current principles of design applied both to intersections and to elements of road infrastructure located between intersections.