A computationally efficient two-stage method for short-term traffic prediction on urban roads

Abstract

Short-term traffic prediction plays an important role in intelligent transport systems. This paper presents a novel two-stage prediction structure using the technique of Singular Spectrum Analysis (SSA) as a data smoothing stage to improve the prediction accuracy. Moreover, a novel prediction method named Grey System Model (GM) is introduced to reduce the dependency on method training and parameter optimisation. To demonstrate the effects of these improvements, this paper compares the prediction accuracies of SSA and non-SSA model structures using both a GM and a more conventional Seasonal Auto-Regressive Integrated Moving Average (SARIMA) prediction model. These methods were calibrated and evaluated using traffic flow data from a corridor in Central London under both normal and incident traffic conditions. The prediction accuracy comparisons show that the SSA method as a data smoothing step before the application of machine learning or statistical prediction methods can improve the final traffic prediction accuracy. In addition, the results indicate that the relatively novel GM method outperforms SARIMA under both normal and incident traffic conditions on urban roads.     

Estimation of missing traffic counts using factor,genetic, neural,and regression techniques

Analyses from some of the highway agencies show that up to 50% permanent traffic counts (PTCs) have missing values. It will be difficult to eliminate such a significant portion of data from traffic analysis. Literature review indicates that the limited research uses factor or autoregressive integrated moving average (ARIMA) models for predicting missing values. Factor-based models tend to be less accurate. ARIMA models only use the historical data. In this study, genetically designed neural network and regression models, factor models, and ARIMA models were developed. It was found that genetically designed regression models based on data from before and after the failure had the most accurate results. Average errors for refined models were lower than 1% and the 95th percentile errors were below 2% for counts with stable patterns. Even for counts with relatively unstable patterns, average errors were lower than 3% in most cases.     

Comparison of parametric and nonparametric models for traffic flow forecasting

Single point short-term traffic flow forecasting will play a key role in supporting demand forecasts needed by operational network models. Seasonal autoregressive integrated moving average (ARIMA), a classic parametric modeling approach to time series, and nonparametric regression models have been proposed as well suited for application to single point short-term traffic flow forecasting. Past research has shown seasonal ARIMA models to deliver results that are statistically superior to basic implementations of nonparametric regression. However, the advantages associated with a data-driven nonparametric forecasting approach motivate further investigation of refined nonparametric forecasting methods. Following this motivation, this research effort seeks to examine the theoretical foundation of nonparametric regression and to answer the question of whether nonparametric regression based on heuristically improved forecast generation methods approach the single interval traffic flow prediction performance of seasonal ARIMA models.     

Estimating potential reductions in externalities from rail-road substitution in Trans-European freight transport corridors

This paper develops a method for analysing and estimating savings in externalities that could be achieved by substituting truck with rail freight services in a given Trans-European freight transport corridor. The externalities affected include energy consumption, emissions of greenhouse gases, noise, congestion, and traffic incidents/accidents. The European Commission transport policy aims to provide an institutional framework for the medium- to long-term sustainable development of the transport sector. An important aspect of this policy is to stimulating the modal shift from truck to rail freight transport in inland Trans-European corridors.     

Discrete dynamic models for freeway incident detection systems

This paper considers the problem of freeway incident detection within the general framework of computer‐based freeway surveillance and control. A new approach to the detection of freeway traffic incidents is presented based on a discrete‐time stochastic model of the form ARIMA (0, 1, 3) that describes the dynamics of traffic occupancy observations. This approach utilizes real‐time estimates of the variability in traffic occupancies as detection thresholds, thus eliminating the need for threshold calibration and lessening the problem of false‐alarms. Because the moving average parameters of the ARIMA (0, 1, 3) model change over time, these parameters can be updated occasionally. The performance of the developed detection algorithm has been evaluated in terms of detection rate, false‐alarm rate, and average time‐lag to detection, using a total of 1692 minutes of occupancy observations recorded during 50 representative traffic incidents.     

A hybrid approach to integrate fuzzy C-means based imputation method with genetic algorithm for missing traffic volume data estimation

Although various innovative traffic sensing technologies have been widely employed, incomplete sensor data is one of the most major problems to significantly degrade traffic data quality and integrity. In this study, a hybrid approach integrating the Fuzzy C-Means (FCM)-based imputation method with the Genetic Algorithm (GA) is develop for missing traffic volume data estimation based on inductance loop detector outputs. By utilizing the weekly similarity among data, the conventional vector-based data structure is firstly transformed into the matrix-based data pattern. Then, the GA is applied to optimize the membership functions and centroids in the FCM model. The experimental tests are conducted to verify the effectiveness of the proposed approach. The traffic volume data collected at different temporal scales were used as the testing dataset, and three different indicators, including root mean square error, correlation coefficient, and relative accuracy, are utilized to quantify the imputation performance compared with some conventional methods (Historical method, Double Exponential Smoothing, and Autoregressive Integrated Moving Average model). The results show the proposed approach outperforms the conventional methods under prevailing traffic conditions.     

A dynamic traffic assignment model for the assessment of moving bottlenecks

Moving bottlenecks in highway traffic are defined as a situation in which a slow-moving vehicle, be it a truck hauling heavy equipment or an oversized vehicle, or a long convey, disrupts the continuous flow of the general traffic. The effect of moving bottlenecks on traffic flow is an important factor in the evaluation of network performance. This effect, though, cannot be assessed properly by existing transportation tools, especially when the bottleneck travels relatively long distances in the network.This paper develops a dynamic traffic assignment (DTA) model that can evaluate the effects of moving bottlenecks on network performance in terms of both travel times and traveling paths. The model assumes that the characteristics of the moving bottleneck, such as traveling path, physical dimensions, and desired speed, are predefined and, therefore, suitable for planned conveys.The DTA model is based on a mesoscopic simulation network-loading procedure with unique features that allow assessing the special dynamic characteristics of a moving bottleneck. By permitting traffic density and speed to vary along a link, the simulation can capture the queue caused by the moving bottleneck while preserving the causality principles of traffic dynamics.     

Short-term,real-time prediction of the extreme ambient carbon monoxide concentrations due to vehicular exhaust emissions using transfer function-noise model

The Box–Jenkins transfer function-noise (TFN) models (Box, G.E.P., Jenkins, G.M., Reinsel, G.C., 1994. Time Series Analysis: Forecasting and Control, 3rd ed. Prentice-Hall, Englewood Cliffs, NJ.) have been used to provide short-term, real-time forecast of the extreme carbon monoxide for an air quality control region (AQCR) comprising a major traffic intersection in the centre of the capital city of Delhi. The time series of the surface wind speed and ambient temperature have been used as “explaining” exogenous variables in the TFN models. When compared with the results of univariate ARIMA model of the endogenous series, the forecast performance is found to improve with the inclusion of the wind speed as input series; however, no significant improvement is observed in the forecast with the inclusion of temperature as input series.     

Anisotropic second-order models and associated fundamental diagrams

According to Banks [Investigation of some characteristics of congested flow. Transportation Research Record, 1999], traffic heterogeneity explains the data scattering on the flow–density plane and positive transferences within the congested phase (a transference is a line connecting adjacent points in the time series). This heterogeneity results from a traffic mixture, made up of various vehicles and drivers, or different traffic conditions such as meteorological conditions. This paper only deals with traffic mixture and more particularly with vehicle classes such as passenger car and truck, which are correlated to the vehicle length. When considering a macroscopic model, the mean vehicle length, which is measured by sensors, is associated with the truck percentage. Then the Generic Second Order Model (GSOM) by Lebacque [Lebacque, J.P., Mammar, S., Haj-Salem, H., 2007a. Generic second-order traffic flow modeling. In: Proceedings of the 17th International Symposium on Transportation and Traffic Theory, London, 23–25 July 2007, 749–770.] provides a rigorous mathematical framework for traffic heterogeneity modeling. The added value in this paper is that admissible invariants which characterize generic fundamental diagrams, possibly depending on the mean vehicle length, are interpreted and debated. Aw–Rascle–Zhang’s [Aw, A., Rascle, M., 2000. Resurrection of second-order models of traffic flow. SIAM Journal of Applied Mathematics, 60 (3), 916–938; Zhang, H.M., 2002. A non equilibrium traffic model devoid of gas-like behavior. Transportation Research Part B, 36, 275–290.] and Colombo’s [Colombo, R.M., 2002. A 2 × 2 hyperbolic traffic flow model. Mathematical and Computer Modeling, 35, 683–688.] anisotropic models are deeply analyzed from a traffic point of view. At last an extended GSOM equation system provides a full parameterization of fundamental diagrams which is needed to traffic heterogeneity modeling.     

Memory properties and fractional integration in transportation time-series

In transportation analyses, autoregressive integrated moving average (ARIMA) and generalized autoregressive conditional heteroskedasticity (GARCH) models have been widely used mainly because of their well established theoretical foundation and ease of application. However, they lack the ability to capture long memory properties and do not jointly treat the mean and variance (variability) of a time-series. We employ fractionally integrated dual memory models and compare results to classical time-series models in a traffic engineering context. Results indicate that dual memory models offer better representation of the original time-series than classical models; further, forcing the differentiation parameter of ARIMA model to equal 1 leads to over-inflated moving average terms and, consequently, to questionable models with artificial correlation structures.     

Characteristics of truck traffic in Singapore

This paper presents the results of a project conducted to study the characteristics of truck traffic in Singapore. Detailed traffic surveys recording counts of vehicles by axle-configuration were performed at 219 sites over a period of nearly two years. The surveys covered 5 different road classes, namely expressways, arterials, collectors, industrial roads and local roads. It was found that the time distribution of truck travel were not the same among the five road classes. The peaking characteristics of truck traffic were less pronounced compared to passenger car traffic. The peak hour truck volume varied from 67.0% to 9.7% of the daily truck traffic as compared to 13.8% for passenger car traffic. The lane distribution pattern of truck traffic was studied in detail by road class, and was found to be a function of total directional traffic volume, total directional truck volume and the number of traffic lanes. Composition analysis was also carried out to study the lane use characteristics of single- and multiple-unit trucks.     

Combining driver response with cyclical queue optimisation over selected corridors

A case study located in Auckland, New Zealand, was used to quantify the magnitude of savings that may result if the SCATS adaptive traffic control system contains an explicitly combined queue estimation and offset adjustment on a cycle‐by‐cycle basis. A validated SATURN traffic model was used to evaluate five scenarios that represent the short‐run and long‐run efficiency gains resulting from progressive signal adaption with an objective of queue minimisation on the main corridors. Optimisation was applied both area‐wide, and on selected arterial corridors, using a combined split/offset optimisation routine with responsive driver behaviour to achieve a network‐wide and corridor‐specific efficiency gain. The modelling heuristic evaluates the efficiency of both the Equisat and P₀ optimisation policies that would mimic a more progressive adaption of signals under SCATS. Results for the long‐run area‐wide optimisation can produce network‐wide travel‐time savings in the order of 20% and a reduction in transient queues of 28% if only selected corridors are optimised, with a 5% reduction in journey time over an average 8‐min journey. Copyright © 2016 John Wiley & Sons, Ltd.     

Evaluation of green wave policy in real-time railway traffic management

In order to face the expected growth of transport demand in the next years, several new traffic control policies have been proposed and analyzed both to generate timetables and to effectively manage the traffic in real-time. In this paper, a detailed optimization model is used to analyze one such policy, called green wave, which consists in letting trains wait at the stations to avoid speed profile modifications in open corridors. Such policy is expected to be especially effective when the corridors are the bottleneck of the network. However, there is a lack of quantitative studies on the real-time effects of using this policy. To this end, this work shows a comparison of the delays obtained when trains are allowed or not to change their speed profile in open corridors. An extensive computational study is described for two practical dispatching areas of the Dutch railway network.     

Empirical analysis of short period traffic counts and their efficiency: the case of Indian traffic

Short period traffic counts (SPTCs) are conducted routinely to estimate the annual average daily traffic (AADT) at a particular site. This paper uses Indian traffic volume data to methodically and extensively study the effect of four aspects related to the design of SPTCs. These four aspects are: (i) for how long, (ii) on which days should SPTCs be carried out, (iii) how many times, and (iv) on which months should SPTCs be carried out? The analyses indicate that the best durations for conducting SPTCs are 3 days (starting with a Thursday) and 7 days, for total traffic and truck traffic, respectively. Further, these counts should be repeated twice a year keeping a separation of two months between the counts to obtain good estimates of AADT at minimal cost. An additional outcome of this study has been the determination of seasonal factor values for roads in developing economies, like India.     

Are HOV/eco-lanes a sustainable option to reducing emissions in a medium-sized European city?

Innovative traffic management measures are needed to reduce transportation-related emissions. While in Europe, road lane management has focused mainly on introduction of bus lanes, the conversion to High Occupancy Vehicles (HOV) and eco-lanes (lanes dedicated to vehicles running on alternative fuels) has not been studied comprehensively. The objectives of this research are to: (1) Develop an integrated microscopic modeling platform calibrated with real world data to assess both traffic and emissions impacts of future Traffic Management Strategies (TMS) in an urban area; (2) Evaluate the introduction of HOV/eco-lanes in three different types of roads, freeway, arterial and urban routes, in an European medium-sized city and its effects in terms of emissions and traffic performance. The methodology consists of three distinct phases: (a) Traffic and road inventory data collection; (b) Traffic and emissions simulation using an integrated platform of microscopic simulation; and (c) Evaluation of scenarios. For the baseline scenario, the statistical analysis shows valid results. The results show that HOV and eco-lanes in a medium European city are feasible, and when the Average Occupancy of Vehicles (AOV) increases, on freeways, the majority of vehicles can reduce their travel time (2%) with a positive impact in terms of total emissions (−38% NO_x, −39% HC, −43% CO and −37% CO₂). On urban and arterial corridors, the reduction in emissions could be achieved only if the AOV increases from 1.50 to 1.70 passengers/vehicle. Total emissions of the corridor with an AOV of 1.70 passengers/vehicle can be reduced up to 35–36% for the urban route while the values can be reduced by 36–39% for the arterial road. With the introduction of Hybrid Electric Vehicles (HEV) and Electric Vehicles (EV) it is possible to reduce emissions, although the introduction of eco-lanes did not show significant reductions in emissions. When both policies are simulated together, an emissions improvement is observed for the arterial route and for two of the scenarios.     

Development and calibration of a long-distance passenger traffic assignment model

This paper studies the assignment of long-distance passenger traffic on a highway corridor network. First, we propose a traditional model for the long-distance traffic assignment considering interactions with local commuter traffic. It addresses the effect of local networks on highway corridors. An iterative algorithm is developed to solve for the exact solution. Then, to address the potential computational issues that arise therein, a decomposition method is proposed by introducing a new concept of corridor elasticity. An assignment procedure for long-distance passenger traffic is developed accordingly. Numerical tests show that the proposed decomposition method makes significant improvements in computational performance at a small loss of optimality. This decomposition method well approximates the exact assignment from the traditional formulation, especially when the highway corridors are near-saturation. The proposed decomposition method appears practical for application.     

Dynamic models of commuter behavior: Experimental investigation and application to the analysis of planned traffic disruptions

This paper reviews the results of a series of experiments aimed at investigating the day-to-day dynamics of commuter behavior in congested traffic systems. The interactive experiments involve actual work commuters in a simulated traffic system, whereby commuters noncooperatively supply their decisions to a traffic simulation model that determines the resulting arrival times and associated trip times; these in turn form the basis of the commuters' decisions on the next day. Models are developed to predict the daily switching of departure time and/or route by individual commuters in response to experienced congestion in the system or to exogenously supplied information. These models are incorporated in a dynamic modelling framework for the analysis of the impacts of planned traffic disruptions, such as those associated with major highway repair and reconstruction activities.     

Mobile mapping technology of wind velocity data along highway for traffic safety evaluation

Traffic safety of highway vehicles under strong crosswind gust and other hazardous driving conditions has become a pressing issue for modern highway transportation and economy. It is known that the actual wind environment at the typical height of a moving vehicle varies considerably from one segment to another even on the same highway, because of influence of the specific terrain and surroundings. Therefore, accurate crosswind velocity data in both time and spatial domains are needed, for a rational assessment of traffic safety risks for various moving vehicles on highways in windy conditions. In addition to site-specific wind data which can be used for most vehicles, vehicle-specific crosswind velocity is often required for an accurate safety assessment of high-sided vehicles with unique shapes. A mobile mapping technology aiming at collecting site-specific as well as vehicle-specific wind velocity data for traffic safety evaluations was developed. Such technology integrates a 3D sonic anemometer and geospatial video mapping system, mounted on a vehicle driven along highways at a normal (cruising) speed. As a result, both vehicle-specific and general site-specific crosswind velocity can be directly “sensed” and collected by using a high-sided vehicle or a streamlined car as the test vehicle. A field test of the developed technology with a high-sided truck driven on mountainous sections of the interstate I-70 (in Colorado) was conducted. The crosswind data at six selected feature locations along I-70, representing different roadside environments, was analyzed. Wind-tunnel investigations employing the scaled models of the truck used in the field test as well as a common streamlined sedan car were conducted to evaluate the accuracy and the feasibility of the developed technology.     

Truck trip generation in small- and medium-sized urban areas

This paper describes procedures to develop truck trip generation (TTG) rates for small- and medium-sized urban areas and its implications. Ordinary least squares models are used to develop separate truck production and attraction equations with the number of employees as the independent variable for three industrial groups – retail, transportation and warehousing, and manufacturing. Results from this research indicate that number of employees is a statistically significant predictor, and has significant explanatory power in predicting the number of truck trips produced and attracted. The rates developed in this study are also found to be significantly different from rates developed in other studies with the implication that caution needs to be taken when transferring TTG rates. The rates are applied in a travel demand model as the initial step of incorporating truck traffic into the modeling process.     

An intermodal freight transport system for optimal supply chain logistics

Complexity in transport networks evokes the need for instant response to the changing dynamics and uncertainties in the upstream operations, where multiple modes of transport are often available, but rarely used in conjunction. This paper proposes a model for strategic transport planning involving a network wide intermodal transport system. The system determines the spatio-temporal states of road based freight networks (unimodal) and future traffic flow in definite time intervals. This information is processed to devise efficient scheduling plans by coordinating and connecting existing rail transport schedules to road based freight systems (intermodal). The traffic flow estimation is performed by kernel based support vector mechanisms while mixed integer programming (MIP) is used to optimize schedules for intermodal transport network by considering various costs and additional capacity constraints. The model has been successfully applied to an existing Fast Moving Consumer Goods (FMCG) distribution network in India with encouraging results.