Shaw-Pin Miaou

Modeling Traffic Crash-Flow Relationships for Intersections: Dispersion Parameter, Functional Form, and Bayes Versus Empirical Bayes Methods

Statistical relationships between traffic crashes and traffic flows at roadway intersections have been extensively modeled and evaluated in recent years. The underlying assumptions adopted in the popular models for intersections are challenged. First, the assumption that the dispersion parameter is a fixed parameter across sites and time periods is challenged. Second, the mathematical limitations of some functional forms used in these models, particularly their properties at the boundaries, are examined. It is also demonstrated that, for a given data set, a large number of plausible functional forms with almost the same overall statistical goodness of fit (GOF) is possible, and an alternative class of logical formulations that may enable a richer interpretation of the data is introduced. A comparison of site estimates from the empirical Bayes and full Bayes methods is also presented. All discussions and comparisons are illustrated with a set of data collected for an urban four-legged signalized intersection in Toronto, Ontario, Canada, from 1990 to 1995. In discussing functional forms, the need for some goodness-of-logic measures, in addition to the GOF measure, is emphasized and demonstrated. Finally, analysts are advised to be mindful of the underlying assumptions adopted in the popular models, especially the assumption that the dispersion parameter is a fixed parameter, and the limitations of the functional forms used. Promising directions in which this study may be extended are also discussed.

Pitfalls of Using R2 To Evaluate Goodness of Fit of Accident Prediction Models

In developing statistical models of traffic accidents, flow, and roadway design, the R2 goodness-of-fit measure has been used for many years to (a) determine the overall quality and usability of the model, (b) select covariates for inclusion in the model, (c) make decisions as to whether it would be worthwhile to collect additional covariates, and (d) compare the relative quality of models developed from different studies. The pitfalls of using R2 to make these decisions and comparisons are demonstrated through computer simulations of commonly used accident prediction models, including the Poisson and negative binomial regression models. Because accident prediction models are nonnormal and functional forms are typically nonlinear, it is shown that R2 is not an appropriate measure to make any of the decisions and comparisons mentioned. Also, three properties are identified as desirable for any alternative measure to appropriately evaluate these models: (a) it should be bounded between 0 and 1—a value of 0 ...

REAL-TIME PREDICTION OF TRAFFIC FLOWS USING DYNAMIC GENERALIZED LINEAR MODELS

In previous real-time flow prediction studies, the emphasis was placed on the prediction accuracy of the model. The accuracy of the prediction bounds (or limits), on the other hand, was largely ignored. Prediction bounds are, however, important input parameters in such applications as real-time stochastic traffic control, incident detection, and route guidance in the context of dynamic traffic assignment. The objectives of this study are to explore the statistical nature of traffic flows when aggregated at short time intervals and to examine the potential of using the generalized linear model in the dynamic setting to predict traffic flows and provide prediction bounds. Specifically, this study derives recursive algorithms based on the quasi-likelihood principle and performs on-line, multiple-step-ahead predictions of short-term arrival flows for signalized intersections. Preliminary results are presented using a simulated data set from CORSIM and a real data set collected from signalized intersections.

Part 1: Roadside Safety Design: Developing Guidelines for Median Barrier Installation: Benefit-Cost Analysis with Texas Data

Guidelines for the installation of median barriers presented in the AASHTO Roadside Design Guide have remained essentially unchanged for more than 30 years. In recent years, the need for improved guidance has prompted several states to reevaluate their guidelines and has also precipitated a nationwide research project administered by the Transportation Research Board. The objective of the study, on which this paper is based, was to develop improved guidelines for the use of median barriers on new and existing high-speed, multilane, divided highways in Texas. The purpose here is to present some modeling and benefit-cost analysis results from that study, with a focus on the results from a particular data set developed under a cross-sectional with-without study design. The highways of interest are those classified as Interstates, freeways, and expressways with four or more lanes and posted speed limits of 55 mph (88 km/h) or higher. The models employed to estimate median-related crash frequencies and severit...

Estimating Vehicle Roadside Encroachment Frequencies by Using Accident Prediction Models

The existing data to support the development of roadside encroachment-based accident-prediction models are limited and largely outdated. Under FHWA and TRB sponsorship, several roadside safety projects have attempted to address this issue by proposing rather comprehensive data collection plans and conducting pilot data collection efforts. It is clear from these studies that the required cost for the proposed roadside field data-collection efforts will be very high. Furthermore, the validity of any field-collected roadside encroachment data may be questionable because of the technical difficulty of distinguishing intentional (or controlled) from unintentional (or uncontrolled) encroachments. A method to estimate some of the basic roadside encroachment parameters, including vehicle roadside encroachment frequency and the probability distribution of lateral extent of encroachments, using existing accident-based prediction models is proposed. The method is developed by utilizing the probabilistic relationships between a roadside encroachment event and a run-off-the-road accident event. With some assumptions, the method is capable of providing a wide range of basic encroachment parameters from conventional accident-based prediction models. To illustrate the concept and use of such a method, some basic encroachment parameters are estimated for rural two-lane undivided roads. In addition, the estimated encroachment parameters are compared with those estimated from the existing encroachment data. The illustration indicates that this method can be a viable approach to estimating basic encroachment parameters of interest and, thus, has the potential of reducing the roadside data collection cost.

Some Limitations of the Models in the Highway Safety Manual to Predict Run-off-Road Crashes

Crash-prediction models in the current edition of the Highway Safety Manual (HSM) have been developed to predict crash frequency by collision type and severity level for specific types of roadways and sites. Each model is made up of three major components: safety performance functions (SPFs), crash modification factors, and calibration factors. The objective of this study was to identify the limitations of the prediction models in estimating single-vehicle, run-off-road (SVROR) crashes for roadside safety analyses and suggest needed changes and developments. The paper presents a review of the state of the models in HSM and focuses on SPFs. Data from FHWAs safety effects of cross-section design for two-lane roads database were used to gain insight about the characteristics of SVROR crashes and total crashes, and to identify the limitations of the current models in predicting the frequency, type, and severity of SVROR crashes. Three major areas of limitations of SPFs are discussed: (a) assumptions involved in development, (b) variables that are potentially important to roadside design but not considered, and (c) statistical bias and uncertainty of the model equations.