Wai Wong

Group-based optimization of a time-dependent TRANSYT traffic model for area traffic control

In this paper, a time-dependent TRANSYT traffic model is developed for the evaluation of a performance index, which is a weighted combination of the estimated delay and number of stops. The group-based specification of signal timings is employed for the traffic model and the set of constraints on these variables is formed. The derivatives of the performance index with respect to the group-based variables are derived, based on which optimization heuristics are developed to solve the time-dependent problem. A medium size signal-controlled network from the Shatin New Town in Hong Kong is employed as a numerical example to illustrate the effectiveness of the heuristics. Two other scenarios are also analyzed, one is based on the average traffic flow and the other takes the traffic flows in individual time intervals for independent analyses. A well-known microscopic simulation model NETSIM is used to evaluate the performance indices for the signal plans derived from these three scenarios. With the proposed heuristics, a remarkable improvement over the average flow scenario is obtained, and when compared with the signal plans from independent analyses, a good improvement is also found. Sensitivity analyses are conducted to illustrate the superiority of the proposed methodology under different levels of congestion in the network.

Network topological effects on the macroscopic Bureau of Public Roads function

Abstract Cost flow functions are a central class of models in the transport field because they are an essential ingredient of static user equilibrium traffic assignment and transport policy and planning analysis. Macroscopic cost flow (MCF) functions, which model travel cost at different levels of network use, have gained much attention in recent decades due to their potential applications in area-wide traffic management and control, and initial land-use planning. They are the collective outcomes of the responses of road users to the existing transportation network and the interactions between road users at different levels of traffic demand. Network topological effects have long been anticipated to be the primary factors governing the shape and performance of a network. However, few studies have investigated network topology. This paper aims to unveil the direct link between network topological metrics and the parameters of a specific MCF, in the form of macroscopic Bureau of Public Roads (MBPR) function, with the support of real-world data. Seventy-one 1 km × 1 km urban built-up regions were sampled in Hong Kong. The MBPR functions of the selected regions were calibrated using taxi global positioning system data. Intensive investigations revealed that the average number of junctions per unit distance and the road density were topological features correlating with the free-flow travel time and congestion sensitivity parameters of the MBPR function, respectively. A spatially variable MBPR function was established.

SHEARED DELAY FORMULAE FOR THE TRANSYT TRAFFIC MODEL: A REVIEW AND CALIBRATION

The paper summarizes the general theory of sheared curves and gives a general expression for these curves. The use of sheared delay formulae in different versions of TRANSYT, a well-known procedure for area traffic control, is reviewed. A new set of sheared formulae for the TRANSYT traffic model to estimate random-and-oversaturation queues and delays has recently been proposed that opens the way to extend the TRANSYT program to solve time-dependent problems. However, the choice of the key parameter in the sheared formulae remains an open question that needs further investigation. In this paper, the key parameter is calibrated by means of a well-known microscopic simulation package, NETSIM, which was validated for a medium-sized signal-controlled network in Hong Kong. Useful results were obtained.

A time-dependent TRANSYT traffic model for area traffic control

Traffic signals have been used in many countries and found to be one of the most effective ways to resolve conflicting traffic movements. Conventional methods for the calculation of signal settings use stage-based specification of timings and assume the traffic demand is time-stationary. With the advent of microprocessor controller technology, a higher degree of flexibility for the specification of signal settings becomes possible by using a group-based representation of signal timings. In this paper, the original TRANSYT traffic model is modified to deal with the problem of time-varying demand. This time-dependent TRANSYT traffic model employs a newly calibrated set of sheared formulae for queues and delays, based on groupbased signal specification. This opens the ways of extending the traffic model for solving the time-dependent signal setting problem in road networks for area traffic control. The performance index evaluated from the time-dependent TRANSYT traffic model is compared with the results obtained from a microscopic simulation model NETSIM.

Estimation of Scaling Factors for Traffic Counts Based on Stationary and Mobile Sources of Data

To combine mobile sources and stationary sources, a modeling approach to quantify the variability of the linear projection function using a non-linear regression method is established in this study. Weights that vary spatial-temporally are assigned to neighboring scaling factors. Together with a normalized weighted average function, the subject scaling factor is determined. The framework is applied to a case study in Hong Kong combining Global Positioning System data and the annual traffic counts from 85 fixed stations in Annual Traffic Census database. The performance of the models is assessed based on relative root mean square error and Akaike information criterion.

Bootstrap standard error estimations of nonlinear transport models based on linearly projected data

ABSTRACT Linear data projection is a commonly leveraged data scaling method for unbiased traffic data estimation. However, recent studies have shown that model estimations based on linearly projected data would certainly result in biased standard errors. Although methods have been developed to remove such biases for linear regression models, many transport models are nonlinear regression models. This study outlines the practical difficulties of the traditional approach to standard error estimation for generic nonlinear transport models, and proposes a bootstrapping mean value restoration method to accurately estimate the parameter standard errors of all nonlinear transport models based on linearly projected data. Comprehensive simulations with different settings using the most commonly adopted nonlinear functions in modeling traffic flow demonstrate that the proposed method outperforms the conventional method and accurately recovers the true standard errors. A case study of estimating a macroscopic fundamental diagram that illustrates situations necessitating the proposed method is presented.