Panagiotis Ch. Anastasopoulos

A note on modeling vehicle accident frequencies with random-parameters count models.

In recent years there have been numerous studies that have sought to understand the factors that determine the frequency of accidents on roadway segments over some period of time, using count data models and their variants (negative binomial and zero-inflated models). This study seeks to explore the use of random-parameters count models as another methodological alternative in analyzing accident frequencies. The empirical results show that random-parameters count models have the potential to provide a fuller understanding of the factors determining accident frequencies.

An empirical assessment of fixed and random parameter logit models using crash- and non-crash-specific injury data

Traditional crash-severity modeling uses detailed data gathered after a crash has occurred (number of vehicles involved, age of occupants, weather conditions at the time of the crash, types of vehicles involved, crash type, occupant restraint use, airbag deployment, etc.) to predict the level of occupant injury. However, for prediction purposes, the use of such detailed data makes assessing the impact of alternate safety countermeasures exceedingly difficult due to the large number of variables that need to be known. Using 5-year data from interstate highways in Indiana, this study explores fixed and random parameter statistical models using detailed crash-specific data and data that include the injury outcome of the crash but not other detailed crash-specific data (only more general data are used such as roadway geometrics, pavement condition and general weather and traffic characteristics). The analysis shows that, while models that do not use detailed crash-specific data do not perform as well as those that do, random parameter models using less detailed data still can provide a reasonable level of accuracy.

Tobit analysis of vehicle accident rates on interstate highways

There has been an abundance of research that has used Poisson models and its variants (negative binomial and zero-inflated models) to improve our understanding of the factors that affect accident frequencies on roadway segments. This study explores the application of an alternate method, tobit regression, by viewing vehicle accident rates directly (instead of frequencies) as a continuous variable that is left-censored at zero. Using data from vehicle accidents on Indiana interstates, the estimation results show that many factors relating to pavement condition, roadway geometrics and traffic characteristics significantly affect vehicle accident rates.

A study of factors affecting highway accident rates using the random-parameters tobit model.

A large body of previous literature has used a variety of count-data modeling techniques to study factors that affect the frequency of highway accidents over some time period on roadway segments of a specified length. An alternative approach to this problem views vehicle accident rates (accidents per mile driven) directly instead of their frequencies. Viewing the problem as continuous data instead of count data creates a problem in that roadway segments that do not have any observed accidents over the identified time period create continuous data that are left-censored at zero. Past research has appropriately applied a tobit regression model to address this censoring problem, but this research has been limited in accounting for unobserved heterogeneity because it has been assumed that the parameter estimates are fixed over roadway-segment observations. Using 9-year data from urban interstates in Indiana, this paper employs a random-parameters tobit regression to account for unobserved heterogeneity in the study of motor-vehicle accident rates. The empirical results show that the random-parameters tobit model outperforms its fixed-parameters counterpart and has the potential to provide a fuller understanding of the factors determining accident rates on specific roadway segments.

A multivariate tobit analysis of highway accident-injury-severity rates

Relatively recent research has illustrated the potential that tobit regression has in studying factors that affect vehicle accident rates (accidents per distance traveled) on specific roadway segments. Tobit regression has been used because accident rates on specific roadway segments are continuous data that are left-censored at zero (they are censored because accidents may not be observed on all roadway segments during the period over which data are collected). This censoring may arise from a number of sources, one of which being the possibility that less severe crashes may be under-reported and thus may be less likely to appear in crash databases. Traditional tobit-regression analyses have dealt with the overall accident rate (all crashes regardless of injury severity), so the issue of censoring by the severity of crashes has not been addressed. However, a tobit-regression approach that considers accident rates by injury-severity level, such as the rate of no-injury, possible injury and injury accidents per distance traveled (as opposed to all accidents regardless of injury-severity), can potentially provide new insights, and address the possibility that censoring may vary by crash-injury severity. Using five-year data from highways in Washington State, this paper estimates a multivariate tobit model of accident-injury-severity rates that addresses the possibility of differential censoring across injury-severity levels, while also accounting for the possible contemporaneous error correlation resulting from commonly shared unobserved characteristics across roadway segments. The empirical results show that the multivariate tobit model outperforms its univariate counterpart, is practically equivalent to the multivariate negative binomial model, and has the potential to provide a fuller understanding of the factors determining accident-injury-severity rates on specific roadway segments.

Cost Savings Analysis of Performance-Based Contracts for Highway Maintenance Operations

Highway agencies around the world are undergoing major changes in their traditional maintenance practices, including the privatization of entire sections of highway routine maintenance activities. Performance-based contracts (PBC) are an option in such privatizing efforts. This paper presents a methodology to estimate the likelihood and amount of cost savings associated with the application of PBC for highway maintenance operations. Using data on maintenance contracts from around the world, we develop models that can be used to compare several contracting methods and include variables such as contract duration, activity type, and contract size. We find that large projects with strong competition, long duration and extension periods, long outsourced road sections that incorporate crack sealing, pothole repair, illumination repair/maintenance, and mowing activities, favor outsourcing under PBC. Our methodology can be useful to transportation agencies for making decisions about the use of PBC and other methods of maintenance outsourcing at the preplanning phase.

Three-Stage Least-Squares Analysis of Time and Cost Overruns in Construction Contracts

Construction cost overrun and time overrun (delay) are a significant problem in highway-construction project delivery. Previous research studies have provided insight into the factors that affect overruns; however the findings may have been limited because they do not explicitly consider the simultaneous relationship between cost and time overruns. In this paper, we use data from Indiana highway projects to provide empirical evidence that a simultaneous relationship exists between cost and time overruns and that analysis of these two contractual outputs need to take due cognizance of such simultaneity. Using the three-stage least-squares technique, we identify a number of factors that significantly affect cost overrun and time overrun and we show how the effect of these variables vary by attributes such as project type and results of the bidding process. The models developed in this paper can help agencies enhance the estimation of the expected overruns of final cost and the delay in completion time for their planned projects.

Empirical Assessment of the Likelihood and Duration of Highway Project Time Delays

Delays in the completion of highway construction and maintenance projects are important concerns to state highway agencies and contractors alike because such time delays can have a number of adverse consequences, such as extending the duration of active work zones, contributing to road-user dissatisfaction and increasing the risk of litigation regarding delay responsibility. In this paper, using data from 1,722 highway projects in Indiana, random-parameter statistical models are estimated to study the factors that contribute to the likelihood of encountering a project time delay and its duration. The model estimation results show that the likelihood and duration of project time delays are significantly influenced by factors such as project cost (contract bid amount), project type, planned project duration, and the likelihood of adverse weather.

Household Automobile and Motorcycle Ownership Analyzed with Random Parameters Bivariate Ordered Probit Model

This paper investigates the factors that affect household automobile and motorcycle ownership in large metropolitan areas. Extensive geocoded trip data from Athens, Greece, were modeled with the random parameters bivariate ordered probit model. This model accounts for unobserved heterogeneity in the data population and commonly shared characteristics with automobile and motorcycle ownership. The random parameters bivariate probit model provided a statistically superior fit compared with its fixed parameters counterpart. The studys results indicate that vehicle (automobile and motorcycle) ownership is determined by a number of factors, such as traveler characteristics, the population density at the origin and destination, the distance and time to the destination for several trip purposes, and access to public transit.

Analysis of Pavement Overlay and Replacement Performance Using Random Parameters Hazard-Based Duration Models

The effectiveness of pavement overlays and pavement replacements in terms of their impact on pavement life is not well understood. This is complicated further by data collection limitations and by the effect that physical deterioration, load volumes, weather, geology, and other factors may have on their effectiveness. Understanding the survivability of overlays and replacements has the potential to provide improved resource allocation and more effective use of state funds. In this paper, pavement overlays and replacements are assessed for their effectiveness on pavement life for urban roads. Using data from Indiana, seemingly unrelated regression equations first are estimated to predict the pavement performance over time. Using these forecasts and historical thresholds, the service life of the pavement is determined and random parameter duration models are estimated to identify influential factors affecting pavement service life. The model-estimation results provide some new insights into the interrelationships among pavement rehabilitation, pavement condition, pavement service life, road functional class, traffic loads and trucks, weather and soil condition, and rehabilitation expenditures.