Traditionally, data have been collected to measure and improve the performance of incident management (IM). While these data are less detailed than crash records, they are timelier and contain useful attributes typically not reported in the crash database. This paper proposes the use of Getis–Ord (Gi*) spatial statistics to identify hot spots on freeways from an IM database while selected impact attributes are incorporated into the analysis. The Gi* spatial statistics jointly evaluate the spatial dependency effect of the frequency and attribute values within the framework of the conceptualized spatial relationship. The application of the method was demonstrated through a case study by using the incident database from the Houston, Texas, Transportation Management Center (TranStar). The method successfully identified the clusters of high-impact accidents from more than 30,000 accident records from 2006 to 2008. The accident duration was used as a proxy measure of its impact. The proposed method could be modified, however, to identify the locations with high-valued impacts by using any other attributes, provided that they were either continuous or categorical in nature and could provide meaningful implications. With improved intelligent transportation system infrastructure and communication technology, hot spot analyses performed with IM data of freeway network and arterials in the vicinity have become a much more promising alternative. Freeway management agencies can use the results of hot spot analysis to provide visualized information to aid the decision-making process in the design, evaluation, and management of IM strategies and resources. The limitations of the method and possible future research are discussed in the closing section of the paper.

This paper presents a mathematical model for realtime queue estimation using connected vehicle (CV) technology from wireless sensor networks. The objective is to estimate the queue length for queue-based adaptive signal control. The proposed model can be applied without signal timing, traffic volume, or queue characteristics as basic inputs. The model is also developed so that it can work with both fixed-time signals and actuated signals. Furthermore, a discrete wavelet transform (DWT) is applied to the queue estimation algorithm in this paper for the first time. The purpose of the DWT is to enhance the proposed queue estimation to be more accurate and consistent regardless of the randomness in the penetration ratio. Experimental results are provided to validate the proposed model in both pretimed control and actuated control with a microscopic simulator, i.e., VISSIM. The results indicate that the proposed algorithm is able to estimate the queue length from VISSIM in the test case with pretimed signal control reasonably well. The results in actuated control cases, which have not been studied previously, showed that the proposed algorithm remains as accurate as the pretimed control cases. The accuracy of the proposed queue estimation algorithm is obtained without relying on basic inputs that other models typically require but are often impractical to obtain. Therefore, it is expected that the proposed queue estimation model is applicable for adaptive signal control using CV technology in practice.Index Terms-Adaptive signal control, discrete wavelet transform, traffic queue length estimation, wireless vehicle-to-vehicle communications. 1524-9050

Transit signal priority (TSP) strategy gives transit vehicles preferential treatments to move through an intersection with minimum delay. To produce a good TSP timing, advance planning with enough look-ahead time is the key. This, however, means added uncertainty about bus arrival time at stop bar. In this paper, we proposed a stochastic mixed-integer nonlinear program (SMINP) model as the core component of a real-time TSP control system. The model adopts a novel approach to capture the impacts of the priority operation to other traffic by using the deviations of the phase split times from the optimal background split times. In addition, the model explicitly accounts for the randomness of the bus' arrival time by considering the bus stop dwell time and the delay caused by standing vehicle queues. The SMINP is implemented in a simulation evaluation platform developed using a combination of a microscopic traffic simulator and a commercial optimization solver. Comparison analyses were performed to compare the proposed control model with the state-of-the-practice TSP system [i.e., ring-barrier controller (RBC)-TSP]. The results showed the SMINP has yielded as much as 30% improvement of bus delay compared with RBC-TSP in a single-bus case. In a multiple-bus case, SMINP handles the bus priority request much more effectively under congested traffic conditions. Index Terms-Degree of saturation, mixed-integer nonlinear model, near-side bus stop, rolling horizon, stochastic optimization, transit signal priority (TSP).

The artificial neural network (ANN) is one advance approach to freeway travel time prediction. Various studies using different inputs have come to no consensus on the effects of input selections. In addition, very little discussion has been made on the temporal–spatial aspect of the ANN travel time prediction process. In this study, we employ an ANN ensemble technique to analyze the effects of various input settings on the ANN prediction performances. Volume, occupancy, and speed are used as inputs to predict travel times. The predictions are then compared against the travel times collected from the toll collection system in Houston. The results show speed or occupancy measured at the segment of interest may be used as sole input to produce acceptable predictions, but all three variables together tend to yield the best prediction results. The inclusion of inputs from both upstream and downstream segments is statistically better than using only the inputs from current segment. It also appears that the magnitude of prevailing segment travel time can be used as a guideline to set up temporal input delays for better prediction accuracies. The evaluation of spatiotemporal input interactions reveals that past information on downstream and current segments is useful in improving prediction accuracy whereas past inputs from the upstream location do not provide as much constructive information. Finally, a variant of the state‐space model (SSNN), namely time‐delayed state‐space neural network (TDSSNN), is proposed and compared against other popular ANN models. The comparison shows that the TDSSNN outperforms other networks and remains very comparable with the SSNN. Future research is needed to analyze TDSSNN's ability in corridor prediction settings.

No abstract