Exploration of Advances in Statistical Methodologies for Crash Count and Severity Prediction Models PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download Exploration of Advances in Statistical Methodologies for Crash Count and Severity Prediction Models PDF full book. Access full book title Exploration of Advances in Statistical Methodologies for Crash Count and Severity Prediction Models by Kai Wang. Download full books in PDF and EPUB format.
Author: Kai Wang
Publisher:
ISBN:
Category : Electronic dissertations
Languages : en
Pages :
Get Book Here
Book Description
This report first describes the use of different copula based models to simultaneously estimate the two crash indicators: injury severity and vehicle damage. The Gaussian copula model outperforms the other copula based model specifications (i.e. Gaussian, Farlie-Gumbel-Morgenstern (FGM), Frank, Clayton, Joe and Gumbel copula models), and the results indicate that injury severity and vehicle damage are highly correlated, and the correlations between injury severity and vehicle damage varied with different crash characteristics including manners of collision and collision types. This study indicates that the copula-based model can be considered to get a more accurate model structure when simultaneously estimating injury severity and vehicle damage in crash severity analyses. The second part of this report describes estimation of cluster based SPFs for local road intersections and segments in Connecticut using socio-economic and network topological data instead of traffic counts as exposure. The number of intersections and the total local roadway length were appropriate to be used as exposure in the intersection and segment SPFs, respectively. Models including total population, retail and non-retail employment and average household income are found to be the best both on the basis of model fit and out of sample prediction. The third part of this report describes estimation of crashes by both crash type and crash severity on rural two-lane highways, using the Multivariate Poisson Lognormal (MVPLN) model. The crash type and crash severity counts are significantly correlated; the standard errors of covariates in the MVPLN model are slightly lower than the other two univariate crash prediction models (i.e. Negative Binomial model and Univariate Poisson Lognormal model) when the covariates are statistically significant; and the MVPLN model outperforms the UPLN and NB models in crash count prediction accuracy. This study indicates that when simultaneously predicting crash counts by crash type and crash severity for rural two-lane highways, the MVPLN model should be considered to avoid estimation error and to account for the potential correlations among crash type counts and crash severity counts.
Author: Kai Wang
Publisher:
ISBN:
Category : Electronic dissertations
Languages : en
Pages :
Get Book Here
Book Description
This report first describes the use of different copula based models to simultaneously estimate the two crash indicators: injury severity and vehicle damage. The Gaussian copula model outperforms the other copula based model specifications (i.e. Gaussian, Farlie-Gumbel-Morgenstern (FGM), Frank, Clayton, Joe and Gumbel copula models), and the results indicate that injury severity and vehicle damage are highly correlated, and the correlations between injury severity and vehicle damage varied with different crash characteristics including manners of collision and collision types. This study indicates that the copula-based model can be considered to get a more accurate model structure when simultaneously estimating injury severity and vehicle damage in crash severity analyses. The second part of this report describes estimation of cluster based SPFs for local road intersections and segments in Connecticut using socio-economic and network topological data instead of traffic counts as exposure. The number of intersections and the total local roadway length were appropriate to be used as exposure in the intersection and segment SPFs, respectively. Models including total population, retail and non-retail employment and average household income are found to be the best both on the basis of model fit and out of sample prediction. The third part of this report describes estimation of crashes by both crash type and crash severity on rural two-lane highways, using the Multivariate Poisson Lognormal (MVPLN) model. The crash type and crash severity counts are significantly correlated; the standard errors of covariates in the MVPLN model are slightly lower than the other two univariate crash prediction models (i.e. Negative Binomial model and Univariate Poisson Lognormal model) when the covariates are statistically significant; and the MVPLN model outperforms the UPLN and NB models in crash count prediction accuracy. This study indicates that when simultaneously predicting crash counts by crash type and crash severity for rural two-lane highways, the MVPLN model should be considered to avoid estimation error and to account for the potential correlations among crash type counts and crash severity counts.
Author: Irfan Uddin Ahmed
Publisher:
ISBN:
Category : Automobile driving in bad weather
Languages : en
Pages : 222
Get Book Here
Book Description
The primary objective of practitioners working on traffic safety is to reduce the number and severity of crashes. The Highway Safety Manual (HSM) provides practitioners with analytical tools and techniques to estimate the expected crash frequency and severity with the aim to identify and evaluate safety countermeasures. Expected crash frequency can be estimated using Safety Performance Functions (SPFs) provided in Part C of the HSM. The HSM provides simple SPFs which are developed using the most frequently used crash counts model, the negative binomial regression model. The rural nature of Wyoming highways coupled with the mountainous terrain (i.e., challenging roadway geometry) make the HSM basic SPFs unsuitable to determine crash contributing factors for Wyoming conditions. In this regard, the objective of this study is to implement advanced statistical methods such as the different functional forms of Negative Binomial, and Bayesian approach, to develop crash prediction models, investigate crash contributing factors, and determine the impact of safety countermeasures. Bayesian statistics in combination with the power of Markov Chain Monte Carlo (MCMC) sampling techniques provide frameworks to model small sample datasets and complex models at the same time, where the traditional Maximum Likelihood Estimation (MLE) based methods tend to fail. As such, a novel No-U-Turn Sampler for Hamiltonian Monte Carlo (NUTS HMC) sampling technique in a Bayesian framework was utilized to investigate the crash frequency, injury severity of crashes on the interstate freeways and some rural highways in Wyoming. The Poisson and the Negative Binomial (NB) models are the most commonly used regression models in traffic safety analysis. The advantage of the NB model can be further enhanced by providing different functional forms of the variance and the dispersion structure. The NB-2 is the most common form of the NB model, typically used in developing safety performance functions (SPFs) largely due to the mean-variance quadratic relationship. However, studies in the literature have shown that the mean-variance relationship could be unrestrained. Another introduced formulation of the NB model is NB-1, which assumes that there is a constant ratio linking the mean and the variance of the crash frequencies. A more general type of the NB model is the NB-P model, which does not constrain the mean-variance relationship. Thus, leveraging the power of this unrestrained mean-variance relationship, more accurate safety models could be developed, and these would lead to more accurate estimation of crash risk and benefits of potential solutions. This study will help practitioners to implement advanced methodologies to solve traffic safety problems of rural highways that have plagued the researchers for a long time now. The methodologies proposed in this study will help practitioners to replace the outdated and inefficient traditional models and obtain more accurate traffic safety models to predict crashes and the resulting crash injury severity. Moreover, this research quantified the safety effectiveness of some unique countermeasures on rural highways.
Author:
Publisher:
ISBN:
Category : Traffic accidents
Languages : en
Pages : 91
Get Book Here
Book Description
Author: National Research Council (U.S.). Transportation Research Board
Publisher:
ISBN:
Category : Traffic accidents
Languages : en
Pages : 148
Get Book Here
Book Description
Transportation Research Record contains the following papers: Method for identifying factors contributing to driver-injury severity in traffic crashes (Chen, WH and Jovanis, PP); Crash- and injury-outcome multipliers (Kim, K); Guidelines for identification of hazardous highway curves (Persaud, B, Retting, RA and Lyon, C); Tools to identify safety issues for a corridor safety-improvement program (Breyer, JP); Prediction of risk of wet-pavement accidents : fuzzy logic model (Xiao, J, Kulakowski, BT and El-Gindy, M); Analysis of accident-reduction factors on California state highways (Hanley, KE, Gibby, AR and Ferrara, T); Injury effects of rollovers and events sequence in single-vehicle crashes (Krull, KA, Khattack, AJ and Council, FM); Analytical modeling of driver-guidance schemes with flow variability considerations (Kaysi, I and Ail, NH); Evaluating the effectiveness of Norway's speak out! road safety campaign : The logic of causal inference in road safety evaluation studies (Elvik, R); Effect of speed, flow, and geometric characteristics on crash frequency for two-lane highways (Garber, NJ and Ehrhart, AA); Development of a relational accident database management system for Mexican federal roads (Mendoza, A, Uribe, A, Gil, GZ and Mayoral, E); Estimating traffic accident rates while accounting for traffic-volume estimation error : a Gibbs sampling approach (Davis, GA); Accident prediction models with and without trend : application of the generalized estimating equations procedure (Lord, D and Persaud, BN); Examination of methods that adjust observed traffic volumes on a network (Kikuchi, S, Miljkovic, D and van Zuylen, HJ); Day-to-day travel-time trends and travel-time prediction form loop-detector data (Kwon, JK, Coifman, B and Bickel, P); Heuristic vehicle classification using inductive signatures on freeways (Sun, C and Ritchie, SG).
Author: Houjun Tang
Publisher:
ISBN:
Category :
Languages : en
Pages :
Get Book Here
Book Description
Roadway safety is a high priority among transportation agencies domestically and internationally. In order to improve traffic safety, the research community has been working to develop and apply more advanced statistical modeling methods in an effort to identify relationships between crash frequency or severity and various driver, roadway, and vehicle characteristics. For example, crash frequency models have evolved from Poisson and negative binomial (NB) regression models, to models that handle excess zero counts (e.g., zero-inflated Poisson or NB models), and to models that address unobserved heterogeneity (e.g., latent class and random parameters models). Examples of crash severity models include multinomial, ordered, and nested logit models. These statistical models usually contain assumptions that are not straight-forward to assess, or put limitations on variables and data. To alleviate these concerns, researchers have recently introduced data mining techniques into traffic safety research. Examples of these methods include classification and regression trees (CART), random forests (RF), support vector machines (SVM) and neural networks (NN). Each has shown promising results to predict the frequency or severity of traffic crashes, which is essential to managing the safety performance of a roadway network.Most safety studies that have applied data mining algorithms have focused on model outcomes of specific crash types and have not addressed parameter sensitivity and its influence on model performance. Furthermore, the application of more sophisticated algorithms has been limited in the published traffic safety literature. In light of this, the purpose of the research is to fill these gaps by systematically evaluating the predictive power of CART and RF algorithms in crash severity analysis and the MOB algorithm in a crash frequency context, and by exploring the sensitivity of key parameters that affect model performance within each algorithm. In addition, guidelines for applying select data mining models are generalized for application purposes. Empirical data collected from Pennsylvania are used to accomplish the research tasks. The major findings from this research are as follows: 1. The RF model produced better predictive power than the CART model, while the CART model performed slightly better than the binary logit model, in the crash severity context. 2. When increasing the parameter values in the CART algorithm (i.e., minimum node size and prior probability of fatal and injury crashes), the sensitivity (i.e., prediction accuracy) of fatal and injury crashes increases and the precision decreases.3. The RF model is robust to parameter tuning. The evaluation metrics are stabilized when more than three variables are considered at each node splitting. The model can produce satisfactory performance with limited number of trees (i.e., far less than the default value). The variable importance ranking is not sensitive to different parameter settings.4. The MOB-NB model shows better data fitness than traditional NB models, in terms of log-likelihood and AIC, in the crash frequency context.5. The estimated confidence intervals and elasticities of independent variables suggest that the MOB-NB model can efficiently identify variable effect heterogeneity under different subgroup patterns in the dataset. In the crash frequency context, the existence of passing zones and the posted speed limit were identified as subgroups in the present study.6. The MOB-NB model produces the minimum mean square error among candidate models, including standard NB model and adjusted NB models, which incorporate splitting variables identified by MOB-NB model. The difference in prediction accuracy among the models is relatively small.
Author: Lisa Kaunitz
Publisher:
ISBN:
Category :
Languages : en
Pages : 47
Get Book Here
Book Description
This study explores factors that effect vehicle accidents, predicts the severity of accidents through logistic regression, and forecasts the number of future accidents to occur using time-series analysis. From insights gathered during exploration, a final dataset is prepared for the use of a logistic regression model. The final model predicts whether or not an accident will be severe with an accuracy of 82%, and reveals the three main features that statistically contribute to the odds of an accident having a severe impact on traffic. Finally, a time-series analysis is run in order to model the number of accidents that can occur on a given day using historical data. This paper evaluates the dataset in ways that have yet to be explored, and provides a great baseline understanding of what is possible for the future of transportation.
Author: National Research Council (U.S.). Transportation Research Board
Publisher:
ISBN:
Category : Traffic accident investigation
Languages : en
Pages : 212
Get Book Here
Book Description
Covers empirical approaches to outlier detection in intelligent transportation systems data, modeling of traffic crash-flow relationships for intersections, profiling of high-frequency accident locations by use of association rules, analysis of rollovers and injuries with sport utility vehicles, and automated accident detection at intersections via digital audio signal processing.
Author:
Publisher:
ISBN:
Category :
Languages : en
Pages : 226
Get Book Here
Book Description
Author: Gaurav Satish Mehta
Publisher:
ISBN:
Category : Electronic dissertations
Languages : en
Pages : 158
Get Book Here
Book Description
Providing safe travel from one point to another is the main objective of any public transportation agency. The recent publication of the Highway Safety Manual (HSM) has resulted in an increasing emphasis on the safety performance of specific roadway facilities. The HSM provides tools such as crash prediction models that can be used to make informed decisions. The manual is a good starting point for transportation agencies interested in improving roadway safety in their states. However, the models published in the manual need calibration to account for the local driver behavior and jurisdictional changes. The method provided in the HSM for calibrating crash prediction models is not scientific and has been proved inefficient by several studies. To overcome this limitation this study proposes two alternatives. Firstly, a new method is proposed for calibrating the crash prediction models using negative binomial regression. Secondly, this study investigates new forms of state-specific Safety Performance Function SPFs using negative binomial techniques. The HSM's 1st edition provides a multiplier applied to the univariate crash prediction models to estimate the expected number of crashes for different crash severities. It does not consider the distinct effect unobserved heterogeneity might have on crash severities. To address this limitation, this study developed a multivariate extension of the Conway Maxwell Poisson distribution for predicting crashes. This study gives the statistical properties and the parameter estimation algorithm for the distribution. The last part of this dissertation extends the use of Highway Safety Manual by developing a multivariate crash prediction model for the bridge section of the roads. The study then compares the performance of the newly proposed multivariate Conway Maxwell Poisson (MVCMP) model with the multivariate Poisson Lognormal, univariate Conway Maxwell Poisson (UCMP) and univariate Poisson Lognormal model for different crash severities. This example will help transportation researchers in applying the model correctly.
![Analyses of Crash Occurence [sic] and Inury [sic] Severities on Multi Lane Highways Using Machine Learning Algorithms PDF](https://books.google.com/books/content/images/frontcover/StusnQAACAAJ?fife=w80-h100)
Author: Abhishek Das
Publisher:
ISBN:
Category : Crash injuries
Languages : en
Pages : 199
Get Book Here
Book Description
Reduction of crash occurrence on the various roadway locations (mid-block segments; signalized intersections; un-signalized intersections) and the mitigation of injury severity in the event of a crash are the major concerns of transportation safety engineers. Multi lane arterial roadways (excluding freeways and expressways) account for forty-three percent of fatal crashes in the state of Florida. Significant contributing causes fall under the broad categories of aggressive driver behavior; adverse weather and environmental conditions; and roadway geometric and traffic factors. The objective of this research was the implementation of innovative, state-of-the-art analytical methods to identify the contributing factors for crashes and injury severity. Advances in computational methods render the use of modern statistical and machine learning algorithms. Even though most of the contributing factors are known a-priori, advanced methods unearth changing trends. Heuristic evolutionary processes such as genetic programming; sophisticated data mining methods like conditional inference tree; and mathematical treatments in the form of sensitivity analyses outline the major contributions in this research. Application of traditional statistical methods like simultaneous ordered probit models, identification and resolution of crash data problems are also key aspects of this study. In order to eliminate the use of unrealistic uniform intersection influence radius of 250 ft, heuristic rules were developed for assigning crashes to roadway segments, signalized intersection and access points using parameters, such as 'site location', 'traffic control' and node information. Use of Conditional Inference Forest instead of Classification and Regression Tree to identify variables of significance for injury severity analysis removed the bias towards the selection of continuous variable or variables with large number of categories. For the injury severity analysis of crashes on highways, the corridors were clustered into four optimum groups. The optimum number of clusters was found using Partitioning around Medoids algorithm. Concepts of evolutionary biology like crossover and mutation were implemented to develop models for classification and regression analyses based on the highest hit rate and minimum error rate, respectively. Low crossover rate and higher mutation reduces the chances of genetic drift and brings in novelty to the model development process. Annual daily traffic; friction coefficient of pavements; on-street parking; curbed medians; surface and shoulder widths; alcohol / drug usage are some of the significant factors that played a role in both crash occurrence and injury severities. Relative sensitivity analyses were used to identify the effect of continuous variables on the variation of crash counts. This study improved the understanding of the significant factors that could play an important role in designing better safety countermeasures on multi lane highways, and hence enhance their safety by reducing the frequency of crashes and severity of injuries. Educating young people about the abuses of alcohol and drugs specifically at high schools and colleges could potentially lead to lower driver aggression. Removal of on-street parking from high speed arterials unilaterally could result in likely drop in the number of crashes. Widening of shoulders could give greater maneuvering space for the drivers. Improving pavement conditions for better friction coefficient will lead to improved crash recovery. Addition of lanes to alleviate problems arising out of increased ADT and restriction of trucks to the slower right lanes on the highways would not only reduce the crash occurrences but also resulted in lower injury severity levels.
