Macroscopic spatial analysis of pedestrian and bicycle crashes

This study investigates the effect of spatial correlation using a Bayesian spatial framework to model pedestrian and bicycle crashes in Traffic Analysis Zones (TAZs). Aggregate models for pedestrian and bicycle crashes were estimated as a function of variables related to roadway characteristics, and various demographic and socio-economic factors. It was found that significant differences were present between the predictor sets for pedestrian and bicycle crashes. The Bayesian Poisson-lognormal model accounting for spatial correlation for pedestrian crashes in the TAZs of the study counties retained nine variables significantly different from zero at 95% Bayesian credible interval. These variables were – total roadway length with 35 mph posted speed limit, total number of intersections per TAZ, median household income, total number of dwelling units, log of population per square mile of a TAZ, percentage of households with non-retired workers but zero auto, percentage of households with non-retired workers and one auto, long term parking cost, and log of total number of employment in a TAZ. A separate distinct set of predictors were found for the bicycle crash model. In all cases the Bayesian models with spatial correlation performed better than the models that did not account for spatial correlation among TAZs. This finding implies that spatial correlation should be considered while modeling pedestrian and bicycle crashes at the aggregate or macro-level.     

A Poisson-lognormal conditional-autoregressive model for multivariate spatial analysis of pedestrian crash counts across neighborhoods

This work examines the relationship between 3-year pedestrian crash counts across Census tracts in Austin, Texas, and various land use, network, and demographic attributes, such as land use balance, residents’ access to commercial land uses, sidewalk density, lane-mile densities (by roadway class), and population and employment densities (by type). The model specification allows for region-specific heterogeneity, correlation across response types, and spatial autocorrelation via a Poisson-based multivariate conditional auto-regressive (CAR) framework and is estimated using Bayesian Markov chain Monte Carlo methods. Least-squares regression estimates of walk-miles traveled per zone serve as the exposure measure. Here, the Poisson-lognormal multivariate CAR model outperforms an aspatial Poisson-lognormal multivariate model and a spatial model (without cross-severity correlation), both in terms of fit and inference.     

On the significance of omitted variables in intersection crash modeling

Advances in safety research – trying to improve the collective understanding of motor vehicle crash causes and contributing factors – rest upon the pursuit of numerous lines of research inquiry. The research community has focused considerable attention on analytical methods development (negative binomial models, simultaneous equations, etc.), on better experimental designs (before–after studies, comparison sites, etc.), on improving exposure measures, and on model specification improvements (additive terms, non-linear relations, etc.).     

Modeling crash spatial heterogeneity: Random parameter versus geographically weighting

The widely adopted techniques for regional crash modeling include the negative binomial model (NB) and Bayesian negative binomial model with conditional autoregressive prior (CAR). The outputs from both models consist of a set of fixed global parameter estimates. However, the impacts of predicting variables on crash counts might not be stationary over space. This study intended to quantitatively investigate this spatial heterogeneity in regional safety modeling using two advanced approaches, i.e., random parameter negative binomial model (RPNB) and semi-parametric geographically weighted Poisson regression model (S-GWPR).     

Pedestrian at-fault crashes on rural and urban roadways in Alabama

The research described in this paper explored the factors contributing to the injury severity resulting from pedestrian at-fault crashes in rural and urban locations in Alabama incorporating the effects of randomness across the observations. Given the occurrence of a crash, random parameter logit models of injury severity (with possible outcomes of major, minor, and possible or no injury) for rural and urban locations were estimated. The estimated models identified statistically significant factors influencing the pedestrian injury severities. The results clearly indicated that there are differences between the influences of a variety of variables on the injury severities resulting from urban versus rural pedestrian at-fault accidents. The results showed that some variables were significant only in one location (urban or rural) but not in the other location. Also, estimation findings showed that several parameters could be modeled as random parameters indicating their varying influences on the injury severity. Based on the results obtained, this paper discusses the effects of different variables on pedestrian injury severities and their possible explanations. From planning and policy perspective, the results of this study justify the need for location specific pedestrian safety research and location specific carefully tailored pedestrian safety campaigns.     

Validating crash locations for quantitative spatial analysis: a GIS-based approach

In this paper, the spatial variables of the crash database in Hong Kong from 1993 to 2004 are validated. The proposed spatial data validation system makes use of three databases (the crash, road network and district board databases) and relies on GIS to carry out most of the validation steps so that the human resource required for manually checking the accuracy of the spatial data can be enormously reduced. With the GIS-based spatial data validation system, it was found that about 65-80% of the police crash records from 1993 to 2004 had correct road names and district board information. In 2004, the police crash database contained about 12.7% mistakes for road names and 9.7% mistakes for district boards. The situation was broadly comparable to the United Kingdom. However, the results also suggest that safety researchers should carefully validate spatial data in the crash database before scientific analysis.     

Discrete distributions when modeling the disability severity score of motor victims

Many European countries apply score systems to evaluate the disability severity of non-fatal motor victims under the law of third-party liability. The score is a non-negative integer with an upper bound at 100 that increases with severity. It may be automatically converted into financial terms and thus also reflects the compensation cost for disability. In this paper, standard and zero-altered discrete regression models are applied to model the disability severity score of victims. An application using data from Spain is provided in which the hurdle-Negative Binomial regression was the preferred method. The effects of victims’ characteristics, type of road user and recovery duration are examined. The results suggest that the expected permanent disability severity is higher for older women with long recovery periods. The results provide traffic decision makers with a model to quantify the compensation cost savings due to disability severity reductions.     

Poisson, Poisson-gamma and zero-inflated regression models of motor vehicle crashes: balancing statistical fit and theory

There has been considerable research conducted over the last 20 years focused on predicting motor vehicle crashes on transportation facilities. The range of statistical models commonly applied includes binomial, Poisson, Poisson-gamma (or negative binomial), zero-inflated Poisson and negative binomial models (ZIP and ZINB), and multinomial probability models. Given the range of possible modeling approaches and the host of assumptions with each modeling approach, making an intelligent choice for modeling motor vehicle crash data is difficult. There is little discussion in the literature comparing different statistical modeling approaches, identifying which statistical models are most appropriate for modeling crash data, and providing a strong justification from basic crash principles. In the recent literature, it has been suggested that the motor vehicle crash process can successfully be modeled by assuming a dual-state data-generating process, which implies that entities (e.g., intersections, road segments, pedestrian crossings, etc.) exist in one of two states-perfectly safe and unsafe. As a result, the ZIP and ZINB are two models that have been applied to account for the preponderance of "excess" zeros frequently observed in crash count data. The objective of this study is to provide defensible guidance on how to appropriate model crash data. We first examine the motor vehicle crash process using theoretical principles and a basic understanding of the crash process. It is shown that the fundamental crash process follows a Bernoulli trial with unequal probability of independent events, also known as Poisson trials. We examine the evolution of statistical models as they apply to the motor vehicle crash process, and indicate how well they statistically approximate the crash process. We also present the theory behind dual-state process count models, and note why they have become popular for modeling crash data. A simulation experiment is then conducted to demonstrate how crash data give rise to "excess" zeros frequently observed in crash data. It is shown that the Poisson and other mixed probabilistic structures are approximations assumed for modeling the motor vehicle crash process. Furthermore, it is demonstrated that under certain (fairly common) circumstances excess zeros are observed-and that these circumstances arise from low exposure and/or inappropriate selection of time/space scales and not an underlying dual state process. In conclusion, carefully selecting the time/space scales for analysis, including an improved set of explanatory variables and/or unobserved heterogeneity effects in count regression models, or applying small-area statistical methods (observations with low exposure) represent the most defensible modeling approaches for datasets with a preponderance of zeros.     

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.     

Statistical modelling for falls count data

Falls and their injury outcomes have count distributions that are highly skewed toward the right with clumping at zero, posing analytical challenges. Different modelling approaches have been used in the published literature to describe falls count distributions, often without consideration of the underlying statistical and modelling assumptions. This paper compares the use of modified Poisson and negative binomial (NB) models as alternatives to Poisson (P) regression, for the analysis of fall outcome counts. Four different count-based regression models (P, NB, zero-inflated Poisson (ZIP), zero-inflated negative binomial (ZINB)) were each individually fitted to four separate fall count datasets from Australia, New Zealand and United States. The finite mixtures of P and NB regression models were also compared to the standard NB model. Both analytical (F, Vuong and bootstrap tests) and graphical approaches were used to select and compare models. Simulation studies assessed the size and power of each model fit. This study confirms that falls count distributions are over-dispersed, but not dispersed due to excess zero counts or heterogeneous population. Accordingly, the P model generally provided the poorest fit to all datasets. The fit improved significantly with NB and both zero-inflated models. The fit was also improved with the NB model, compared to finite mixtures of both P and NB regression models. Although there was little difference in fit between NB and ZINB models, in the interests of parsimony it is recommended that future studies involving modelling of falls count data routinely use the NB models in preference to the P or ZINB or finite mixture distribution. The fact that these conclusions apply across four separate datasets from four different samples of older people participating in studies of different methodology, adds strength to this general guiding principle.     

Spatial analysis of fatal and injury crashes in Pennsylvania

Using injury and fatal crash data for Pennsylvania for 1996-2000, full Bayes (FB) hierarchical models (with spatial and temporal effects and space-time interactions) are compared to traditional negative binomial (NB) estimates of annual county-level crash frequency. Covariates include socio-demographics, weather conditions, transportation infrastructure and amount of travel. FB hierarchical models are generally consistent with the NB estimates. Counties with a higher percentage of the population under poverty level, higher percentage of their population in age groups 0-14, 15-24, and over 64 and those with increased road mileage and road density have significantly increased crash risk. Total precipitation is significant and positive in the NB models, but not significant with FB. Spatial correlation, time trend, and space-time interactions are significant in the FB injury crash models. County-level FB models reveal the existence of spatial correlation in crash data and provide a mechanism to quantify, and reduce the effect of, this correlation. Addressing spatial correlation is likely to be even more important in road segment and intersection-level crash models, where spatial correlation is likely to be even more pronounced.     

Finite mixture modeling for vehicle crash data with application to hotspot identification

The application of finite mixture regression models has recently gained an interest from highway safety researchers because of its considerable potential for addressing unobserved heterogeneity. Finite mixture models assume that the observations of a sample arise from two or more unobserved components with unknown proportions. Both fixed and varying weight parameter models have been shown to be useful for explaining the heterogeneity and the nature of the dispersion in crash data. Given the superior performance of the finite mixture model, this study, using observed and simulated data, investigated the relative performance of the finite mixture model and the traditional negative binomial (NB) model in terms of hotspot identification. For the observed data, rural multilane segment crash data for divided highways in California and Texas were used. The results showed that the difference measured by the percentage deviation in ranking orders was relatively small for this dataset. Nevertheless, the ranking results from the finite mixture model were considered more reliable than the NB model because of the better model specification. This finding was also supported by the simulation study which produced a high number of false positives and negatives when a mis-specified model was used for hotspot identification. Regarding an optimal threshold value for identifying hotspots, another simulation analysis indicated that there is a discrepancy between false discovery (increasing) and false negative rates (decreasing). Since the costs associated with false positives and false negatives are different, it is suggested that the selected optimal threshold value should be decided by considering the trade-offs between these two costs so that unnecessary expenses are minimized.     

An empirical tool to evaluate the safety of cyclists: Community based,macro-level collision prediction models using negative binomial regression

Today, North American governments are more willing to consider compact neighborhoods with increased use of sustainable transportation modes. Bicycling, one of the most effective modes for short trips with distances less than 5 km is being encouraged. However, as vulnerable road users (VRUs), cyclists are more likely to be injured when involved in collisions. In order to create a safe road environment for them, evaluating cyclists’ road safety at a macro level in a proactive way is necessary. In this paper, different generalized linear regression methods for collision prediction model (CPM) development are reviewed and previous studies on micro-level and macro-level bicycle-related CPMs are summarized. On the basis of insights gained in the exploration stage, this paper also reports on efforts to develop negative binomial models for bicycle–auto collisions at a community-based, macro-level. Data came from the Central Okanagan Regional District (CORD), of British Columbia, Canada. The model results revealed two types of statistical associations between collisions and each explanatory variable: (1) An increase in bicycle–auto collisions is associated with an increase in total lane kilometers (TLKM), bicycle lane kilometers (BLKM), bus stops (BS), traffic signals (SIG), intersection density (INTD), and arterial–local intersection percentage (IALP). (2) A decrease in bicycle collisions was found to be associated with an increase in the number of drive commuters (DRIVE), and in the percentage of drive commuters (DRP). These results support our hypothesis that in North America, with its current low levels of bicycle use (<4%), we can initially expect to see an increase in bicycle collisions as cycle mode share increases. However, as bicycle mode share increases beyond some unknown ‘critical’ level, our hypothesis also predicts a net safety improvement. To test this hypothesis and to further explore the statistical relationships between bicycle mode split and overall road safety, future research needs to pursue further development and application of community-based, macro-level CPMs.     

Surface condition and safety at signalised intersections

The study reported herein focuses on assessing the effects and contributions of pavement surface condition parameters including roughness, rut depth and skid resistance on crash frequencies and rates at signalised intersections. The assessment has been performed using graphical trend analysis and negative binomial regression. For each signalised intersection in the sample, condition data are collected for the year before and after the year of surface treatment. Crash data, however, are collected for a minimum of 3 and maximum of 5 years before and after treatment years. Analyses results show that before treatment, light condition (day/night), road surface moisture condition (wet/dry) and skid resistance are the significant contributors to crash occurrence. For after-treatment; light condition, road surface moisture condition and the interaction between roughness and traffic volume are the significant contributors. The study aim and approach are presented herein together with a discussion of the main findings.     

Modelling area-wide count outcomes with spatial correlation and heterogeneity: an analysis of London crash data

Count models such as negative binomial (NB) regression models are normally employed to establish a relationship between area-wide traffic crashes and the contributing factors. Since crash data are collected with reference to location measured as points in space, spatial dependence exists among the area-level crash observations. Although NB models can take account of the effect of unobserved heterogeneity (due to omitted variables in the model) among neighbourhoods, such models may not account for spatial correlation areas. It is then essential to adopt an econometric model that takes account of both spatial dependence and uncorrelated heterogeneity simultaneously among neighbouring units. In studying the spatial pattern of traffic crashes, two types of spatial models may be employed: (i) classical spatial models for higher levels of spatial aggregation such as states, counties, etc. and (ii) Bayesian hierarchical models for all spatial units, especially for smaller scale area-aggregations. Therefore, the primary objectives of this paper is to develop a series of relationships between area-wide different traffic casualties and the contributing factors associated with ward characteristics using both non-spatial models (such as NB models) and spatial models and to identify the similarities and differences among these relationships. The spatial units of the analysis are the 633 census wards from the Greater London metropolitan area. Ward-level casualty data are disaggregated by severity of the casualty (such as fatalities, serious injuries, and slight injuries) and by severity of the casualty related to various road users. The analysis implies that different ward-level factors affect traffic casualties differently. The results also suggest that Bayesian hierarchical models are more appropriate in developing a relationship between area-wide traffic crashes and the contributing factors associated with the road infrastructure, socioeconomic and traffic conditions of the area. This is because Bayesian models accurately take account of both spatial dependence and uncorrelated heterogeneity.     

Spatial panel analyses of alcohol outlets and motor vehicle crashes in California: 1999–2008

Although past research has linked alcohol outlet density to higher rates of drinking and many related social problems, there is conflicting evidence of density's association with traffic crashes. An abundance of local alcohol outlets simultaneously encourages drinking and reduces driving distances required to obtain alcohol, leading to an indeterminate expected impact on alcohol-involved crash risk. This study separately investigates the effects of outlet density on (1) the risk of injury crashes relative to population and (2) the likelihood that any given crash is alcohol-involved, as indicated by police reports and single-vehicle nighttime status of crashes. Alcohol outlet density effects are estimated using Bayesian misalignment Poisson analyses of all California ZIP codes over the years 1999–2008. These misalignment models allow panel analysis of ZIP-code data despite frequent redefinition of postal-code boundaries, while also controlling for overdispersion and the effects of spatial autocorrelation. Because models control for overall retail density, estimated alcohol-outlet associations represent the extra effect of retail establishments selling alcohol. The results indicate a number of statistically well-supported associations between retail density and crash behavior, but the implied effects on crash risks are relatively small. Alcohol-serving restaurants have a greater impact on overall crash risks than on the likelihood that those crashes involve alcohol, whereas bars primarily affect the odds that crashes are alcohol-involved. Off-premise outlet density is negatively associated with risks of both crashes and alcohol involvement, while the presence of a tribal casino in a ZIP code is linked to higher odds of police-reported drinking involvement. Alcohol outlets in a given area are found to influence crash risks both locally and in adjacent ZIP codes, and significant spatial autocorrelation also suggests important relationships across geographical units. These results suggest that each type of alcohol outlet can have differing impacts on risks of crashing as well as the alcohol involvement of those crashes.     

A Change Point Method for Monitoring Generalized Linear Profiles in Phase I

The Phase I applications of the statistical profile monitoring have recently been extended to the case when the response variable is binary. We are motivated to undertake the current research in an attempt to try to provide a unified framework for the Phase I control in the context of statistical profile monitoring that can be used to tackle a large class of response variables, such as continuous, count, or categorical response variables. The unified framework is essentially based on applying the change point model to the class of generalized linear models. The proposed Phase I control chart is assessed and compared with the existing charts under binomial and Poisson profiles. Some diagnostic procedures are also discussed. A real data set obtained from a test lab concerning the dispersion of carbon black filler in a rubber mix is used to demonstrate how the proposed chart can be used in practical applications. Some future research directions are also discussed. Copyright © 2014 John Wiley & Sons, Ltd.     

嵌入式RFID系统识别率关键因子的分析

为提高RFID系统识别率,优化设计RFID系统,需要确定影响RFID系统识别率的关键因子.基于关键过程输入变量、关键过程输出变量的模型,提出了一种分析影响RFID系统识别率关键因子的试验设计方法,并确定了关键因子为标签与读写器平面夹角、读写器天线数量、标签与读写器距离.通过测试,分别在标签嵌入汽车轮胎未硫化和硫化两种情况下,进一步确定影响嵌入式RFID系统识别率的各关键因子.最后,应用MINITAB统计软件,采用二项logistic的回归分析方法,分析了各关键因子对识别率的贡献情况.统计分析结果有利于优化设计RFID系统,提高RFID系统识别率.