Monica Menendez

Partial proportional odds model—An alternate choice for analyzing pedestrian crash injury severities

The conventional methods for crash injury severity analyses include either treating the severity data as ordered (e.g. ordered logit/probit models) or non-ordered (e.g. multinomial models). The ordered models require the data to meet proportional odds assumption, according to which the predictors can only have the same effect on different levels of the dependent variable, which is often not the case with crash injury severities. On the other hand, non-ordered analyses completely ignore the inherent hierarchical nature of crash injury severities. Therefore, treating the crash severity data as either ordered or non-ordered results in violating some of the key principles. To address these concerns, this paper explores the application of a partial proportional odds (PPO) model to bridge the gap between ordered and non-ordered severity modeling frameworks. The PPO model allows the covariates that meet the proportional odds assumption to affect different crash severity levels with the same magnitude; whereas the covariates that do not meet the proportional odds assumption can have different effects on different severity levels. This study is based on a five-year (2008-2012) national pedestrian safety dataset for Switzerland. A comparison between the application of PPO models, ordered logit models, and multinomial logit models for pedestrian injury severity evaluation is also included here. The study shows that PPO models outperform the other models considered based on different evaluation criteria. Hence, it is a viable method for analyzing pedestrian crash injury severities.

Evaluation of Presignals at Oversaturated Signalized Intersections

One of the major causes of bus delays in urban environments is signalized intersections. A commonly used solution to give priority to buses at signalized intersections is to dedicate a lane for bus use only. This strategy allows the bus to skip the car queues and minimizes the bus delay experienced at the signal. However, especially for low bus flows, the strategy can waste valuable green time at signals and impose additional car delays. Overall, even when bus passengers enjoy reduced travel times, the total person hours of delay in the system can increase. To avoid this problem and utilize the full capacity of the main signal while still providing bus priority, the use of a presignal has been proposed. The goal of this research was to quantify the benefits of the use of presignals in system wide total person hours of delay, specifically for oversaturated intersections. Theoretical formulas were developed to quantify the effects of a presignal on traffic flow, and the formulas were empirically verified. The theoretical model was used to compare the total delay with a presignal strategy with the total delay with a dedicated bus lane or fully mixed lanes. Bounds on bus-to-car occupancy ratios were quantified for which presignals provided the lowest delay compared with a dedicated lane or mixed lane strategy. The results showed that for oversaturated intersections, presignals were better for the system than dedicated bus lanes. Moreover, presignals could decrease the total person hours of delay compared with mixed lanes for large car demands.

Combining screening and metamodel-based methods: An efficient sequential approach for the sensitivity analysis of model outputs

Abstract Sensitivity analysis (SA) is able to identify the most influential parameters of a given model. Application of SA is usually critical for reducing the complexity in the subsequent model calibration and use. Unfortunately it is hardly applied, especially when the model is in the form of a computationally expensive black-box computer program. A possible solution concerns applying SA to the metamodel (i.e., an approximation of the computationally expensive model) instead. Among the other options, the use of Gaussian process metamodels (also known as Kriging metamodels) has been recently proposed for the SA of computationally expensive traffic simulation models. However, the main limitation of this approach is its dependence on the model dimensionality. When the model is high-dimensional, the estimation of the Kriging metamodel may still be problematic due to its high computational cost. In order to overcome this problem, in the present paper, the Kriging-based approach has been combined with the quasi-optimized trajectory based elementary effects (quasi-OTEE) approach for the SA of high-dimensional models. The quasi-OTEE SA is used first to screen the influential and non-influential parameters of a high-dimensional model; then the Kriging-based SA is used to calculate the variance-based sensitivity indices, and to rank the most influential parameters in a more accurate way. The application of the proposed sequential SA is illustrated with several numerical experiments. Results show that the method can properly identify the most influential parameters and their ranks, while the number of model evaluations is considerably less than the variance-based SA (e.g., in one of the tests the sequential SA requires over 50 times less model evaluations than the variance-based SA).

An Exploratory Study of Two Efficient Approaches for the Sensitivity Analysis of Computationally Expensive Traffic Simulation Models

One of the main challenges arising when calibrating a complex traffic simulation model concerns the selection of the most important input parameters. The quasi-optimized trajectory-based elementary effects (quasi-OTEE) and the Kriging-based sensitivity analysis (SA) are two recently developed efficient approaches for the SA of computationally expensive simulation models. In this paper, two experimental studies using two different traffic simulation models (i.e., Aimsun and VISSIM) are presented to compare these two approaches and to better understand their advantages and disadvantages. Results show that both approaches are able to identify, to a good degree, the important parameters. In particular, the quasi-OTEE is better for screening the parameters, whereas the Kriging-based SA has higher precision in ranking the parameters. These findings suggest the following rule of thumb for the SA of computationally expensive traffic simulation models: the quasi-OTEE SA can be used first to screen the parameters and to decide which parameters to discard. Then, the Kriging-based SA can be used to refine the analysis and calculate first-order indexes to identify the correct rank of the important parameters.

Assessment of the Impact of Incidents near Bottlenecks: Strategies to Reduce Delay

How the location and duration of an incident affect delays near a recurrent bottleneck is evaluated in this study. With conventional kinematic wave theory and some dimensional analysis, tools are provided to determine whether an incident will cause generalized delays (i.e., delays that have a lingering effect for the whole length of the peak hour) according to the incidents magnitude, location, and duration. The results apply to a broad range of cases, encompassing many types of facilities and incidents. Furthermore, the results can be used as a foundation for developing and implementing new strategies to obtain significant reductions in delay. The value of fault-free surveillance is also analyzed and presented as part of an optimization problem for locating roadside assistance vehicles. It is found that this value is very high, which could justify installing advanced traffic-monitoring schemes near major bottlenecks.

Generalized effects of on-street parking maneuvers on the performance of nearby signalized intersections

An on-street parking maneuver can often start a temporary bottleneck that leads to additional delay of following vehicles. When the maneuver occurs upstream near a signalized intersection, the scope of the impact can be magnified. In this case, the intersection might be starved for traffic because of the parking maneuver, so a portion of the intersection capacity is wasted. The authors of this paper estimate the reduction in service rate attributable to the delay from an upstream parking maneuver at different locations and provide suggestions for avoiding such a reduction. The perturbation caused by the parking maneuver is calculated from the hydrodynamic theory of traffic flow. By using dimensional analysis, the authors analyze the relationship between background conditions (i.e., traffic volume, traffic saturation condition, signal control settings) time and location of the parking maneuver, and reduction in service rate at the intersection. Combined with a previous study, this research completes an analysis of the effects of parking maneuvers at different locations near a signalized intersection (as both upstream and downstream maneuvers are included). The results show that, generally, for an undersaturated intersection, parking downstream of the intersection causes fewer negative effects on the service rate than does parking upstream. However, for an oversaturated intersection, parking downstream near the intersection might reduce the service rate significantly. These conclusions may help guide parking planning–management under different background conditions and assist the development of parking regulations (such as dynamic parking supply according to real-time traffic conditions).

Exploring the application of latent class cluster analysis for investigating pedestrian crash injury severities in Switzerland

One of the major challenges in traffic safety analyses is the heterogeneous nature of safety data, due to the sundry factors involved in it. This heterogeneity often leads to difficulties in interpreting results and conclusions due to unrevealed relationships. Understanding the underlying relationship between injury severities and influential factors is critical for the selection of appropriate safety countermeasures. A method commonly employed to address systematic heterogeneity is to focus on any subgroup of data based on the research purpose. However, this need not ensure homogeneity in the data. In this paper, latent class cluster analysis is applied to identify homogenous subgroups for a specific crash type-pedestrian crashes. The manuscript employs data from police reported pedestrian (2009-2012) crashes in Switzerland. The analyses demonstrate that dividing pedestrian severity data into seven clusters helps in reducing the systematic heterogeneity of the data and to understand the hidden relationships between crash severity levels and socio-demographic, environmental, vehicle, temporal, traffic factors, and main reason for the crash. The pedestrian crash injury severity models were developed for the whole data and individual clusters, and were compared using receiver operating characteristics curve, for which results favored clustering. Overall, the study suggests that latent class clustered regression approach is suitable for reducing heterogeneity and revealing important hidden relationships in traffic safety analyses.

Methodology to Evaluate Cost and Accuracy of Parking Patrol Surveys

Patrol surveys are frequently used to estimate average parking duration. However, error in estimating is unavoidable and yet unpredictable. It is, therefore, hard for surveyors to decide on survey budgets; lower ones may generate unusable results and higher ones are costly. Therefore, a method to evaluate and improve survey accuracy without increasing (and possibly even lowering) cost could be highly beneficial. In this paper, an analytical model is built to determine the effects of survey budget on survey accuracy in a systematic and generalized way. The relationship between the survey budget, the estimated average parking duration (i.e., the survey result), and the survey error is illustrated through dimensional analysis. On the basis of this analysis, objective criteria are found to evaluate patrol surveys, a budget test is provided to find the optimal survey budget, and a correction method is proposed to improve survey accuracy. Moreover, simulations are run on the basis of more realistic assumptions, so that the conclusions from the analytical model can be extended to real life situations. The model is well validated with real parking data. An example application is also provided that illustrates how to use the proposed approaches to choose a survey budget and to improve survey accuracy. It is hoped that this study will help surveyors to better understand patrol surveys and obtain high-quality results, while keeping costs to a minimum.

Exploratory Analysis of Signal Coordination Impacts on Macroscopic Fundamental Diagram

The macroscopic fundamental diagram (MFD) of urban traffic is a recently developed tool to describe traffic on large-scale urban networks. A network’s MFD can be significantly affected by various properties, including signal settings, block lengths, travel speeds, and routing behaviors. However, current understanding of the impact of signal offsets (i.e., signal coordination) on the MFD is limited to idealized one-dimensional networks, and little is known about how signal coordination affects the MFDs of more realistic two-dimensional networks. To overcome this gap, the present study examined the impacts of signal coordination on the MFD of an idealized two-dimensional grid network using both an analytical method and a microscopic traffic simulator (Aimsun). Seven coordination strategies were considered, including the simultaneous, alternating, double-alternating, one-dimensional green-wave, two-dimensional green-wave, MAXBAND, and random strategies. In general, the impacts of coordination were highly sensitive to the signal cycle length chosen. The results also revealed that poor coordination can significantly decrease the network capacity and the free-flow travel speed. However, good coordination offers little advantage over simultaneous offsets, even with directional travel demands moving in the prioritized directions. The reason is that the benefits provided to vehicles in the prioritized directions are negated by the disadvantages to vehicles traveling in the nonprioritized directions. These results suggest that the insights from other idealized simulations with simultaneous offsets may be generalizable to more realistic situations with more realistic coordination strategies.

Use of Microsimulation for Examination of Macroscopic Fundamental Diagram Hysteresis Patterns for Hierarchical Urban Street Networks

This study used microsimulation to analyze traffic performance on various idealized hierarchical urban street networks. Each network consisted of local streets and arterial streets, which represented the micro- and macro-structure of an urban network, respectively. Arterials were differentiated from local streets through additional green time at intersections and additional travel lanes. An idealized peak period was simulated for completely uniform demand patterns. Observed relationships between average network flow and density—known as the macroscopic fundamental diagram (MFD)—were used to compare the performance of arterial structures. Specifically, the size and shape of hysteresis loops that emerged in the MFD were used, since the networks were found to have more uniform congestion patterns during the onset of congestion than during congestion recovery. The presence of arterials was found to affect significantly the spatial distribution of congestion on the network. Arterials that passed through the congested center of the network and arterials that divided the network resulted in increased congestion inhomogeneity and larger hysteresis loops in the MFD. Arterials placed near the periphery of the network helped to attract vehicles to less-used areas of the network and reduce congestion in the center, which reduced congestion inhomogeneity. Furthermore, limited opportunities to access the arterials also contributed to dense pockets of congestion on nearby streets. For the network and demand conditions studied here, arterial ring roads appeared to distribute congestion more evenly and have had better network performance than arterial grids.