Browsing by Author "Zeng, Xiaosi"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Development and Evaluation of An Adaptive Transit Signal Priority System Using Connected Vehicle Technology(2014-12-15) Zeng, XiaosiTransit signal priority (TSP) can be a very effective preferential treatment for transit vehicles in congested urban networks. There are two problems with the current practice of the transit signal priority. First, random bus arrival time is not sufficiently accounted for, which?ve become the major hindrance in practice for implementing active or adaptive TSP strategies when a near-side bus stop is present. Secondly, most research focuses on providing bus priority at local intersection level, but bus schedule reliability should be achieved at route level and relevant studies have been lacking. In the first part of this research, a stochastic mixed-integer nonlinear programming (SMINP) model is developed to explicitly to account for uncertain bus arrival time. A queue delay algorithm is developed as the supporting algorithm for SMINP to capture the delays caused by the interactions between vehicle queues and buses entering and exiting near-side bus stops. A concept of using signal timing deviations to approximate the impacts of TSP operations on other traffic is proposed for the first time in this research. In the second part of the research, the deterministic version of the SMINP model is extended to the arterial setting, where a route-based TSP (R-TSP) model is develop to optimize for schedule-related bus performances on the corridor level. The R-TSP model uses the real-time data available only from the connected vehicle communications technology. Based on the connected vehicle technology, a real-time signal control system that implements the proposed TSP models is prototyped in the simulation environment. The connected vehicle technology is also used as the main detection and monitoring mechanism for the real-time control of the adaptive TSP signal system. The adaptive TSP control module is designed as a plug-in module that is envisioned to work with a modern fixed-time or adaptive signal controller with connected vehicle communications capabilities. Using this TSP-enabled signal control system, simulation studies were carried out in both a single intersection setting and a five-intersection arterial setting. The effectiveness of the SMINP model to handle uncertain bus arrival time and the R-TSP model to achieve corridor-level bus schedule reliability were studied. Discussions, conclusions and future research on the topic of adaptive TSP models were made.Item Dynamically Predicting Corridor Travel Time Under Incident Conditions Using a Neural Network Approach(2011-02-22) Zeng, XiaosiThe artificial neural network (ANN) approach has been recognized as a capable technique to model the highly complex and nonlinear problem of travel time prediction. In addition to the nonlinearity, a traffic system is also temporally and spatially dynamic. Addressing the temporal-spatial relationships of a traffic system in the context of neural networks, however, has not received much attention. Furthermore, many of the past studies have not fully explored the inclusion of incident information into the ANN model development, despite that incident might be a major source of prediction degradations. Additionally, directly deriving corridor travel times in a one-step manner raises some intractable problems, such as pairing input-target data, which have not yet been adequately discussed. In this study, the corridor travel time prediction problem has been divided into two stages with the first stage on prediction of the segment travel time and the second stage on corridor travel time aggregation methodologies of the predicted segmental results. To address the dynamic nature of traffic system that are often under the influence of incidents, time delay neural network (TDNN), state-space neural network (SSNN), and an extended state-space neural network (ExtSSNN) that incorporates incident inputs are evaluated for travel time prediction along with a traditional back propagation neural network (BP) and compared with baseline methods based on historical data. In the first stage, the empirical results show that the SSNN and ExtSSNN, which are both trained with Bayesian regulated Levenberg Marquardt algorithm, outperform other models. It is also concluded that the incident information is redundant to the travel time prediction problem with speed and volume data as inputs. In the second stage, the evaluations on the applications of the SSNN model to predict snapshot travel times and experienced travel times are made. The outcomes of these evaluations are satisfactory and the method is found to be practically significant in that it (1) explicitly reconstructs the temporalspatial traffic dynamics in the model, (2) is extendable to arbitrary O-D pairs without complete retraining of the model, and (3) can be used to predict both traveler experiences and system overall conditions.