H. M. Abdul Aziz

Exploring the determinants of pedestrian-vehicle crash severity in New York City

Pedestrian-vehicle crashes remain a major concern in New York City due to high percentage of fatalities. This study develops random parameter logit models for explaining pedestrian injury severity levels of New York City accounting for unobserved heterogeneity in the population and across the boroughs. A log-likelihood ratio test for joint model suitability suggests that separate models for each of the boroughs should be estimated. Among many variables, road characteristics (e.g., number of lanes, grade, light condition, road surface, etc.), traffic attributes (e.g., presence of signal control, type of vehicle, etc.), and land use (e.g., parking facilities, commercial and industrial land use, etc.) are found to be statistically significant in the estimated model. The study also suggests that the set of counter measures should be different for different boroughs in the New York City and the priority ranks of countermeasures should be different as well.

Random parameter model used to explain effects of built-environment characteristics on pedestrian crash frequency

Pedestrian safety has been a major concern for megacities such as New York City. Although pedestrian fatalities show a downward trend, these fatalities constitute a high percentage of overall traffic fatalities in the city. Data from New York City were used to study the factors that influence the frequency of pedestrian crashes. Specifically, a random parameter, negative binomial model was developed for predicting pedestrian crash frequencies at the census tract level. This approach allows the incorporation of unobserved heterogeneity across the spatial zones in the modeling process. The influences of a comprehensive set of variables describing the sociodemographic and built-environment characteristics on pedestrian crashes are reported. Several parameters in the model were found to be random, which indicates their heterogeneous influence on the numbers of pedestrian crashes. Overall, these findings can help frame better policies to improve pedestrian safety.

Integration of Environmental Objectives in a System Optimal Dynamic Traffic Assignment Model

Maintaining air-quality standards is a major concern for transportation planners and policy makers in the United States. This necessitates considering nontraditional emission objectives in transportation systems modeling. In this research, the authors integrate emission-based objectives into the traditional travel time based dynamic assignment framework. Carbon monoxide (CO) emissions from vehicles are computed as functions of space mean speed (determined from an embedded mesoscopic traffic flow model). Different performance metrics (CO emission, system wide travel time, and speed profiles) from the integrated model are compared with traditional dynamic assignment model (with travel time minimization objective). In addition, results indicate changes in route choice behavior of the road users when emission objective is integrated to dynamic assignment framework.

Unified Framework for Dynamic Traffic Assignment and Signal Control with Cell Transmission Model

This research presents a unified framework for dynamic traffic assignment and signal control optimization. An objective function based on the system-optimal approach with an embedded traffic flow model (the cell transmission model) for dynamic traffic assignment was considered. The optimization model (a mixed-integer program) explicitly considered intersection delay and lost time from phase switches in addition to a traditional travel time objective. Two test networks were used to demonstrate the applicability of the proposed model. Results showed better performance of the models when they were compared with fixed-signal timing plans. The formulation of signal control design also accounted for the variation of cycle length, and results showed the variation of cycle lengths with different objective functions under different levels of congestion.

Network Traffic Control in Cyber-Transportation Systems Accounting for User-Level Fairness

This research identifies the advantage of connected vehicle environment in cyber-transportation systems (CTS) and proposes control strategies that can yield benefits from system and user perspectives. Particularly, the focus is to minimize the total stopped delay at signalized intersections for the entire trip of a user. Further, we define and assess the notion of fairness from the user perspective. We develop and implement a vehicle-to-infrastructure (V2I)-based signal controller within the framework of CTS. The test results from a network containing 15 signalized intersections show that the proposed fair queuing (FQ) algorithm performs better than the fixed control and real-time adaptive control in terms of both efficiency (system travel time) and fairness (minimizing the difference between maximum and mean intersection delay) across users.

An analytical framework for vehicular traffic signal control integrated with dynamic traffic assignment using cell transmission model

This work presents two analytical models for the traffic signal control problem within dynamic traffic assignment framework. The bi-level formulation is based on a Stackelberg Duopoly game where the upper level objective is to determine optimal signal timing plans with the lower level representing a dynamic user equilibrium conditions. In addition, a single level system optimal model is proposed. Cell transmission model is adapted as the embedded traffic flow model in the form of a mathematical program. The single level program is solved and tested on a network containing multiple signalized intersections. It is found that the proposed model performs better than pre-timed signal timing plans.

An Approach to Assess the Impact of Dynamic Congestion in Vehicle Routing Problems

This research proposes an integrated framework of capacitated vehicle routing problems (CVRP) and traffic flow model (cell transmission model in this research) to assess the effect of time-varying congestion. We develop a framework consisting sequence of mixed integer programs solving the CVRP with updated cost obtained from the traffic flow model. A real-world network with 15 cities and towns is tested with the framework and results show significant travel time reduction from the case where time-varying congestion is not considered. In addition, we consider system optimal type of route choice behavior within the traffic flow model.