A note on modeling pedestrian-injury severity in motor-vehicle crashes with the mixed logit model
Korea Research Institute for Human Settlements, National Infrastructure & GIS Research Division, 224 Simin-Ro, Dongan-gu, Anyang-si, Gyeonggi-do 431-712, Republic of Korea. Accident; analysis and prevention
(Impact Factor: 1.65).
11/2010; 42(6):1751-8. DOI: 10.1016/j.aap.2010.04.016
Pedestrian-injury severity has been traditionally modeled with approaches that have assumed that the effect of each variable is fixed across injury observations. This assumption ignores possible unobserved heterogeneity which is likely to be particularly important in pedestrian injuries because unobserved physical health, strength, and behavior may significantly affect the pedestrians' ability to absorb collision forces. To address such unobserved heterogeneity, this research applies a mixed logit model to analyze pedestrian-injury severity in pedestrian-vehicle crashes. Using police-reported collision data from 1997 through 2000 from North Carolina, several factors were found to more than double the average probability of fatal injury for pedestrians in motor-vehicle crashes including: darkness without streetlights (400% increase in fatality probability), vehicle is a truck (370% increase), freeway (330% increase), speeding involved (360% increase), and collisions involving a motorist who had been drinking (250% increase). It was also found that the effect of pedestrian age was normally distributed across observations, and that as pedestrians became older the probability of fatal injury increased substantially. Heterogeneity in the mean of the random parameters for the freeway and pedestrian-solely-at-fault collision indicators was related to pedestrian gender, and heterogeneity in the mean of the random parameters for the traffic-sign and motorist-back-up indicators was related to pedestrian age.
Available from: Cong Chen
- "The unobserved heterogeneity could be attributed from different types of factors, including roadways (Flask et al., 2014; Haleem and Gan, 2013; Malyshkina and Mannering, 2010; Morgan and Mannering, 2011), drivers' demographic and behavior characteristics (Haleem and Gan, 2013; Islam and Mannering, 2006; Kim et al., 2013, 2010; Morgan and Mannering, 2011; Ulfarsson and Mannering, 2004), spatial and temporal variations (Malyshkina and Mannering, 2009; Malyshkina et al., 2009; Ukkusuri et al., 2011; Xiong et al., 2014; Xu and Huang, 2015), etc. For instance, Kim et al. (2010) evaluate pedestrian injury severity patterns in pedestrian-vehicle crashes considering the unobserved pedestrian heterogeneity regarding health, strength and behavior. Anastasopoulos et al. (2012a,b) investigated traffic accident rate patterns accounting for the unobserved heterogeneity effects of highway segments. "
[Show abstract] [Hide abstract]
ABSTRACT: Traffic crashes occurring on rural roadways induce more severe injuries and fatalities than those in urban areas, especially when there are trucks involved. Truck drivers are found to suffer higher potential of crash injuries compared with other occupational labors. Besides, unobserved heterogeneity in crash data analysis is a critical issue that needs to be carefully addressed. In this study, a hierarchical Bayesian random intercept model decomposing cross-level interaction effects as unobserved heterogeneity is developed to examine the posterior probabilities of truck driver injuries in rural truck-involved crashes. The interaction effects contributing to truck driver injury outcomes are investigated based on two-year rural truck-involved crashes in New Mexico from 2010 to 2011. The analysis results indicate that the cross-level interaction effects play an important role in predicting truck driver injury severities, and the proposed model produces comparable performance with the traditional random intercept model and the mixed logit model even after penalization by high model complexity. It is revealed that factors including road grade, number of vehicles involved in a crash, maximum vehicle damage in a crash, vehicle actions, driver age, seatbelt use, and driver under alcohol or drug influence, as well as a portion of their cross-level interaction effects with other variables are significantly associated with truck driver incapacitating injuries and fatalities. These findings are helpful to understand the respective or joint impacts of these attributes on truck driver injury patterns in rural truck-involved crashes.
Accident Analysis & Prevention 12/2015; 85:186-198. DOI:10.1016/j.aap.2015.09.005 · 1.87 Impact Factor
- "Unlike these studies in which the main focus was on vehicle–vehicle crashes and information on different crash types (rear end, angle, head-on, side-swipe, etc.) were available and used in segmentation, the current study focuses on a specific crash type–pedestrian crashes. Another approach that takes into account of the unobserved heterogeneity in analyzing pedestrian injury severities in pedestrian–vehicle crashes includes using a mixed logit model (Kim et al., 2010). Previous pedestrian safety studies have utilized the narratives obtained from police accident report to classify pedestrian crashes into different groups such midblock dart crashes, crashes due to pedestrian error, turning vehicle crashes, and crashes involving driver's failure to grant right of way for pedestrians (e.g. "
[Show abstract] [Hide abstract]
ABSTRACT: 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.
Accident; analysis and prevention 10/2015; 85:219-228. DOI:10.1016/j.aap.2015.09.020 · 1.65 Impact Factor
Available from: Mounir Belloumi
- "This is because the marginal effect of a variable depends on all the parameters in the model (Kim et al., 2010), so the Relative Risk ratio is used in this study to compare the risk of different groups of variables. "
[Show description] [Hide description]
DESCRIPTION: The objective of this study is to investigate the contribution of several variables involving vehicle occupant and pedestrian victim using the Negative Binomial (NB) and Multinomial Logit models (MNL) by employing the data of 7000 subjects involved in serious accidents in Tunisia between 2010 and 2013. For the NB model, empirical results show that driving in bad weather condition, road category, such as a national highway, and accidents occurring in right curve alignment, are proven to increase the frequency of accident at selected sites. Whereas, the empirical results of MNL models are of a great variety. The results of the first model of drivers show male drivers are proven to experience higher severity levels than female drivers. Added to that, accidents occurring in adverse weather increase the likelihood of fatal injuries. Another finding of the study is that day of accident, season of accident, region of the accident, time of accident, driver related factors and vehicle related factors increase the probability of severe injury. For the second model of the pedestrian, results suggest that the degree of injury severity is higher for male than female victims. Other significant factors contributing to the injury severity model are the time (day or night) of accidents and speed. The factors identified in this research are expected to help developing potential countermeasures to reduce the severity and the frequency of crashes.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.