H Michael Zhang

Calibration of Microsimulation with Heuristic Optimization Methods

Model calibration is a crucial step in building a reliable microscopic traffic simulation application because it serves as the additional check to ensure that the model parameters accurately reflect the local driving environment so that decisions made on the basis of these results will not be misinformed decisions. Because of its stochastic nature and complexity, the calibration problem, usually formulated as an optimization problem, is often solved by using heuristic methods. To date, calibration is still a time-consuming task because many adopted methods require many simulation runs in search of an optimal solution. Moreover, many aspects of the calibration problem are not fully understood and need further investigation. In this study, another heuristics calibration algorithm is developed on the basis of the simultaneous perturbation stochastic approximation scheme and applied to calibration of several networks coded in Paramics. The results indicate that the new heuristic algorithm can reach the same level of accuracy with considerably fewer iterations and computer processor time than other heuristic algorithms such as genetic algorithms and the trial-and-error iterative adjustment algorithm. Applications of all three heuristic methods in a northern California network also reveal that some model parameters affect the simulation results more significantly than others. These findings can help modelers better choose calibration methods and fine-tune key parameters.

A Polymorphic Dynamic Network Loading Model

A polymorphic dynamic network loading (PDNL) model is developed and discretized to integrate a variety of macroscopic traffic flow and node models. The polymorphism, realized through a general node-link interface and proper discretization, offers several promi- nent advantages. First of all, PDNL allows road facilities in the same network to be represented by different traf- fic flow models based on the tradeoff of efficiency and realism and/or the characteristics of the targeted prob- lem. Second, new macroscopic link/node models can be easily plugged into the framework and compared against existing ones. Third, PDNL decouples links and nodes in network loading, and thus opens the door to parallel computing. Finally, PDNL keeps track of individual ve- hicular quanta of arbitrary size, which makes it possible to replicate analytical loading results as closely as desired. PDNL, thus, offers an ideal platform for studying both an- alytical dynamic traffic assignment problems of different kinds and macroscopic traffic simulation.

Fundamental Diagram of Traffic Flow: New Identification Scheme and Further Evidence from Empirical Data

A systematic approach is developed to identify the bivariate relation of two fundamental traffic variables, traffic volume and density, from single-loop detector data. The approach is motivated by the observation of a peculiar feature of traffic fluctuations. That is, in a short time, traffic usually experiences fluctuations without a significant change in speed. This fact is used to define equilibrium in a new manner, and a mixed integer programming approach is proposed for constructing a piecewise linear fundamental diagram (FD) accordingly. By construction, the proposed method is data adaptive and optimal in the sense of least absolute deviation. This method is used to perform a case study with data from one section of a multilane freeway. The results indicate that both capacity drop and concave–convex FD shapes abound in practice. Differences in traffic behavior across freeway lanes and along freeway sections revealed through the FD are discussed.

Increasing the Capacity of Signalized Intersections with Dynamic Use of Exit Lanes for Left-Turn Traffic

Many congested intersections have a heavy traffic volume on movements for which capacity is insufficient because of geometric limitations. An unconventional approach that increases the capacity of heavily congested intersections is presented: this approach opens up exit lanes for left-turn traffic dynamically with the help of an additional traffic light installed at the median opening (the presignal); this situation is referred to as exit lanes for left-turn (EFL) control. An optimization problem for EFL control was formulated as a mixed-integer nonlinear program, in which the geometric layout, main signal timing, and presignal timing were integrated. The mixed-integer nonlinear program was solved by transformation into a series of mixed-integer linear programs. The latter problem can be solved with the standard branch-and-bound technique. The results of extensive numerical analysis and VISSIM simulation showed that the EFL approach could increase intersection capacity and reduce traffic delay substantially, especially under high left-turn demand. Moreover, the EFL control can be applied to one or multiple legs simultaneously; thus the control is particularly useful for intersections with an unbalanced left demand and a degree of saturation in travel directions.

Driving simulator evaluation of drivers’ response to intersections with dynamic use of exit-lanes for left-turn

With the worsening of urban traffic congestion in large cities around the world, researchers have been looking for unconventional designs and/or controls to squeeze more capacity out of intersections, the most common bottlenecks of the road network. One of these innovative intersection designs, known as the exit-lanes for left-turn (EFL), opens up exit-lanes to be used by left-turn traffic with the help of an additional traffic light installed at the median opening (the pre-signal). This paper studies how drivers respond to EFL intersections with a series of driving simulator experiments. In our experiments, 64 drivers were recruited and divided into two groups. One group is trained to use the EFL while the other group is not. In addition, four scenarios were considered with different sign and marking designs and traffic conditions in the experiments. Results indicate that drivers show certain amount of confusion and hesitation when encountering an EFL intersection for the first time. They can be overcome, however, by increasing exposure through driver education or by cue provided from other vehicles. Moreover, drivers unfamiliar with EFL operation can make a left turn using the conventional left-turn lanes as usual. The EFL operation is not likely to pose any serious safety risk of the intersection in real life operations.

Mechanism of Early-Onset Breakdown at On-Ramp Bottlenecks on Shanghai, China, Expressways

The mechanism of early-onset breakdowns was studied at on-ramp bottlenecks on expressways in Shanghai, China. From four on-ramp breakdown events captured on video, key parameters were extracted: prequeue flow, queue-discharge flow, speed variation per minute, lane change (LC) times in the mainline lanes and the acceleration lane, LC types, and LC locations (longitudinal and lateral). A total of 1,583 LC samples were analyzed. The findings showed a great difference in LC patterns when breakdowns occurred earlier than normal (i.e., before the bottleneck reaches expected capacity). In the case of an early breakdown, most LCs were forced LCs that occurred near the downstream end of the bottleneck, which spread laterally rather quickly. In contrast, in normal breakdowns in the United States, LCs were mostly free LCs that occurred evenly along the bottleneck longitudinally but were concentrated in rightmost lanes laterally.

Computing Individual Path Marginal Cost in Networks with Queue Spillbacks

“Individual path marginal cost” (IPMC) is defined as the change in travel cost of one unit of flow on a time-dependent path caused by one unit of flow on another time-dependent path. Knowledge of IPMC is central to dynamic transportation modeling, for instance, to compute system-optimal network performance, to solve a dynamic origin–destination (O-D) estimation problem, and to analyze equity issues for travelers with different origins and destinations. This paper proposes a method of approximating IPMC for general networks, in which a cell transmission model–based kinematic wave model is used to model traffic dynamics. By tracing the changes in the cumulative flow curves of the bottleneck links on which queues form during dynamic network loading, an approximation method is developed to obtain the IPMC for the cases of merge junctions, diverge junctions, and general junctions. This method was applied to compute the total path marginal cost in a network. The results showed that vehicles at the beginning of the congestion duration had significantly larger marginal travel costs than other vehicles. The method was then applied to solve a dynamic O-D estimation problem with partial link-flow counts and historical O-D trip tables. With the incorporation of IPMC into the estimation procedure, both the O-D demands and the observed path travel times were successfully reproduced.

Local Synchronization Control Scheme for Congested Interchange Areas in Freeway Corridor

Congestion that initiates at closely spaced highway junctions and intersections, particularly freeway interchange areas, may spread and severely degrade the operational efficiency of the whole network if not handled in a timely and proper manner. A local synchronization traffic control scheme is proposed to manage queues at those critical locations through coordination of neighboring intersection traffic signals and freeway on-ramp meters. By reducing the amount of traffic feeding into and increasing the amount of traffic discharging from heavily queued sections, the scheme can prevent a queue from evolving into gridlock and thus improve overall system performance. With the help of a network kinematic wave traffic flow model, the local synchronization scheme is implemented and tested on a computer for two sample networks, one small synthetic corridor network and one large, real corridor network. The numerical results indicate that this control scheme can improve the overall operational efficiency in both corridors considerably, with as much as 50% travel time savings. This control scheme appears to perform best under incident conditions and, somewhat surprisingly, compares favorably with a more complex global optimal control scheme.

Bus Drivers' Responses to Real-Time Schedule Adherence and the Effects on Transit Reliability

Bus drivers have responded positively toward real-time bus information in various surveys. However, empirical studies on their actual responses are limited. On the basis of actual automatic vehicle location data, this study quantified bus drivers’ responses to real-time schedule adherence and their effects on transit reliability. Bus trips that were ahead of and behind schedule were analyzed separately at timepoint stops, regular stops, and along the roadways between stops. Results revealed that bus drivers would use real-time information to keep on schedule. Early buses were found to be more likely to make adjustments in response to information than were late buses along the roadways. Moreover, bus drivers’ responses to real-time information was found to improve transit reliability: 50% of the improvement was the result of drivers’ responses to schedule adherence at timepoint stops and 50% was the result mainly of drivers’ responses to schedule adherence along the roadways. The likelihood that drivers would make adjustments at regular stops to adhere to schedule was low.

Dynamic Holding Strategy to Prevent Buses from Bunching

This study proposed a robust dynamic control strategy to regulate bus headways and prevent buses from bunching by holding them at bus stops. The proposed strategy monitors bus locations in real time and estimates the time gaps between consecutive buses at a desired frequency. The holding times of all buses at their respective immediately downstream stops are determined simultaneously on the basis of the estimated time gaps. A procedure that consists of a discrete quadratic dynamic control program and a quadratic static optimization program was developed to produce a unique dynamic control law of holding times. Numerical investigations on an operational bus route revealed that the proposed strategy produced greater system reliability than did some existing control strategies and that the bus system under the control of the proposed strategy recovered promptly from large system disruptions.