Sc Wong

URBAN TRAFFIC FLOW PREDICTION USING A FUZZY-NEURAL APPROACH

Abstract This paper develops a fuzzy-neural model (FNM) to predict the traffic flows in an urban street network, which has long been considered a major element in the responsive urban traffic control systems. The FNM consists of two modules: a gate network (GN) and an expert network (EN). The GN classifies the input data into a number of clusters using a fuzzy approach, and the EN specifies the input–output relationship as in a conventional neural network approach. While the GN groups traffic patterns of similar characteristics into clusters, the EN models the specific relationship within each cluster. An online rolling training procedure is proposed to train the FNM, which enhances its predictive power through adaptive adjustments of the model coefficients in response to the real-time traffic conditions. Both simulation and real observation data are used to demonstrative the effectiveness of the method.

Reserve capacity of a signal-controlled road network

The concept of reserve capacity has been used extensively for performance measure and timing design of individual signal-controlled intersections. In this paper, we extend this concept to a general signalcontrolled road network under time-stationary conditions. The whole capacity of a road network is controled by intersections whose capacity depends on traffic signal settings. We first define and formulate the reserve capacity of a whole signal-controlled road network, and then show how a bi-level programming method can be used to determine signal setting for maximization of the network reserve capacity. A numerical example is presented to illustrate the concept and the calculation procedure.

A multi-class traffic flow model - An extension of LWR model with heterogeneous drivers

In this paper, a multi-class traffic flow model as an extension of the Lighthill, Whitham and Richards (LWR) model with heterogeneous drivers is formulated. The model takes into account the distribution of heterogeneous drivers characterized by their choice of speeds in a traffic stream, which models the dynamic behavior of heterogeneous users whereby faster vehicles could overtake slower ones under uncongested condition as well as congested condition (though less easily), and slower vehicles would slow down the faster ones. Numerical simulations show that the model can predict many of the puzzling traffic flow phenomena such as the two-capacity (or reverse-lambda) regimes occurred in the fundamental diagram, hysteresis and platoon dispersion.

An Aggregation Approach to Short-Term Traffic Flow Prediction

In this paper, an aggregation approach is proposed for traffic flow prediction that is based on the moving average (MA), exponential smoothing (ES), autoregressive MA (ARIMA), and neural network (NN) models. The aggregation approach assembles information from relevant time series. The source time series is the traffic flow volume that is collected 24 h/day over several years. The three relevant time series are a weekly similarity time series, a daily similarity time series, and an hourly time series, which can be directly generated from the source time series. The MA, ES, and ARIMA models are selected to give predictions of the three relevant time series. The predictions that result from the different models are used as the basis of the NN in the aggregation stage. The output of the trained NN serves as the final prediction. To assess the performance of the different models, the naive, ARIMA, nonparametric regression, NN, and data aggregation (DA) models are applied to the prediction of a real vehicle traffic flow, from which data have been collected at a data-collection point that is located on National Highway 107, Guangzhou, Guangdong, China. The outcome suggests that the DA model obtains a more accurate forecast than any individual model alone. The aggregation strategy can offer substantial benefits in terms of improving operational forecasting.

Lane-based optimization of signal timings for isolated junctions

This paper presents a lane-based optimization method for the integrated design of lane markings and signal settings for isolated junctions. Both traffic and pedestrian movements are considered in a unified framework. Capacity maximization and cycle length minimization problems are considered. The problems are formulated as Binary-Mix-Integer-Linear-Programs (BMILP), which are solvable by any standard branch-and-bound routine. The integer variables include the permitted movements on traffic lanes and successor functions to govern the order of signal displays, whereas the continuous variables include the assigned lane flows, common flow multiplier, cycle length, and starts and durations of green for traffic movements and lanes and pedestrian crossings. A set of constraints are set up to ensure feasibility and safety of the resulting optimal lane markings and signal settings. Numerical examples are given to demonstrate the effectiveness of the proposed method.

Demand-supply equilibrium of taxi services in a network under competition and regulation

Abstract This paper investigates the nature of demand–supply equilibrium in a regulated market for taxi service. Distinguished from conventional economic analysis, a network model is used to describe the demand and supply equilibrium of taxi services under fare structure and fleet size regulation in an either competitive or monopoly market. The spatial structure of the market such as the form of road network and the customer origin–destination demand pattern are explicitly considered. The model can determine a number of system performance measures at equilibrium such as utilization rate for taxi and level of service quality, and predict the effects of alternative regulations on system performance. The model can thus be used as a policy tool by the regulator to ascertain appropriate taxi regulations such as the selection of taxi fleet size and fare structure. A case study in Hong Kong was conducted to illustrate some interesting findings.

A Network Model of Urban Taxi Services

A network model is developed to describe how vacant and occupied taxis will cruise in a road network to search for customers and provide transportation services. The model can determine a number of system performance measures at equilibrium, such as vacant taxi movements and taxi utilization for a given road network and the customer origin-destination demand pattern. The effects of the taxi fleet size and the uncertainty on the system performances are explicitly taken into account. The model offers some interesting insights into the nature of the equilibrium of taxi services, and offers some policy-relevant results for decision making. These include the findings that the average taxi utilization decreases sharply with the number of taxis operating, and that the higher the taxi utilization, the larger the average customer waiting time.

A stochastic transit assignment model using a dynamic schedule-based network

Using the schedule-based approach, in which scheduled time-tables are used to describe the movement of vehicles, a dynamic transit assignment model is formulated. Passengers are assumed to travel on a path with minimum generalized cost which consists of four components: in-vehicle time; waiting time; walking time; and a time penalty for each line change. With the exception of in-vehicle time, each of the other cost components is weighted by a sensitivity coefficient which varies among travelers and is defined by a density function. This time-dependent and stochastic minimum path is generated by a specially developed branch and bound algorithm. The assignment procedure is conducted over a period in which both passenger demand and train headways are varying. Due to the stochastic nature of the assignment problem, a Monte Carlo approach is employed to solve the problem. A case study using the Mass Transit Railway System in Hong Kong is given to demonstrate the model and its potential applications.

A predictive dynamic traffic assignment model in congested capacity-constrained road networks

In this paper, a predictive dynamic traffic assignment model in congested capacity-constrained road networks is formulated. A traffic simulator is developed to incrementally load the traffic demand onto the network, and updates the traffic conditions dynamically. A time-dependent shortest path algorithm is also given to determine the paths with minimum actual travel time from an origin to all the destinations. The traffic simulator and time-dependent shortest path algorithm are employed in a method of successive averages to solve the dynamic equilibrium solution of the problem. A numerical example is given to illustrate the effectiveness of the proposed method.

A DYNAMIC SCHEDULE-BASED MODEL FOR CONGESTED TRANSIT NETWORKS

In this paper we propose a model and algorithm for solving the equilibrium assignment problem in a congested, dynamic and schedule-based transit network. We assume that the time varying origin-destination trip demand is given. All travelers have full predictive information (that have been gained through past experience) about present and future network conditions and select paths that minimize a generalized cost function that encompasses four components: (a) in-vehicle time; (b) waiting time; (c) walking time; and (d) a line change penalty. All transit vehicles have a fixed capacity and operate precisely as specified in pre-set timetables. Passengers queue at platforms according to the single channel first-in-first-out discipline. By using time-increment simulation, the passenger demand is loaded onto the network and the available capacity of each vehicle is updated dynamically. After each simulation run, the passenger arrival and departure profiles at all stations are recorded and these are used to predict dynamic queuing delays. From such delays, minimum paths are updated and used for the next simulation run. The user equilibrium assignment problem is solved iteratively by the method of successive averages.