Macro-level collision prediction models for evaluating neighbourhood traffic safety

2006 ◽  
Vol 33 (5) ◽  
pp. 609-621 ◽  
Author(s):  
Gordon R Lovegrove ◽  
Tarek Sayed
Keyword(s):  
Linear Regression ◽  
Traffic Safety ◽  
Road Safety ◽  
Prediction Models ◽  
Demand Management ◽  
Macro Level ◽  
Safety Planning ◽  
Transportation Demand ◽  
Regression Modelling ◽  
Collision Prediction

This study describes the development of macro-level (i.e., neighbourhood or traffic zone level) collision prediction models using data from 577 neighbourhoods across the Greater Vancouver Regional District. The objective is to provide a safety planning decision-support tool that facilitates a proactive approach to community planning which addresses road safety before problems emerge. The models are developed using the generalized linear regression modelling (GLM) technique assuming a negative binomial error structure. The resulting models relate traffic collisions to neighbourhood characteristics such as traffic volume, demographics, network shape, and transportation demand management. Several models are presented for total or severe collisions in rural or urban zones using measured and (or) modelled data. It is hoped that quantifying a predictive traffic safety – neighbourhood planning relationship will facilitate improved decisions by community planners and engineers and, ultimately, facilitate improved neighbourhood traffic safety for residents and other road users.Key words: neighbourhood safety, macro-level collision prediction models, road safety, safety planning, transportation demand management, sociodemographic, generalized linear regression modelling.

Traffic Safety Implications of Travel Demand Management Policies

2014 ◽  
pp. 115-140
Author(s):  
Ali Pirdavani ◽  
Tom Bellemans ◽  
Tom Brijs ◽  
Bruno Kochan ◽  
Geert Wets
Keyword(s):  
Spatial Correlation ◽  
Traffic Safety ◽  
Travel Demand ◽  
Prediction Models ◽  
Demand Management ◽  
Primary Objective ◽  
Travel Demand Management ◽  
Crash Prediction ◽  
Crash Prediction Models ◽  
Crash Types

Travel Demand Management (TDM) consists of a variety of policy measures that affect the transportation system’s effectiveness by changing travel behavior. Although the primary objective to implement such TDM strategies is not to improve traffic safety, their impact on traffic safety should not be neglected. The main purpose of this study is to investigate differences in the traffic safety consequences of two TDM scenarios: a fuel-cost increase scenario (i.e. increasing the fuel price by 20%) and a teleworking scenario (i.e. 5% of the working population engages in teleworking). Since TDM strategies are usually conducted at a geographically aggregated level, crash prediction models that are used to evaluate such strategies should also be developed at an aggregate level. Moreover, given that crash occurrences are often spatially heterogeneous and are affected by many spatial variables, the existence of spatial correlation in the data is also examined. The results indicate the necessity of accounting for the spatial correlation when developing crash prediction models. Therefore, Zonal Crash Prediction Models (ZCPMs) within the geographically weighted generalized linear modeling framework are developed to incorporate the spatial variations in association between the Number Of Crashes (NOCs) (including fatal, severe, and slight injury crashes recorded between 2004 and 2007) and a set of explanatory variables. Different exposure, network, and socio-demographic variables of 2200 traffic analysis zones in Flanders, Belgium, are considered as predictors of crashes. An activity-based transportation model is adopted to produce exposure metrics. This enables a more detailed and reliable assessment while TDM strategies are inherently modeled in the activity-based models. In this chapter, several ZCPMs with different severity levels and crash types are developed to predict the NOCs. The results show considerable traffic safety benefits of conducting both TDM scenarios at an average level. However, there are certain differences when considering changes in NOCs by different crash types.

Integrated Modeling Approach for Non-Motorized Mode Trips and Fatal Crashes in the Framework of Transportation Safety Planning

2018 ◽  
Vol 2672 (32) ◽  
pp. 49-60 ◽  
Author(s):  
Jaeyoung Lee ◽  
Mohamed Abdel-Aty ◽  
Qing Cai ◽  
Ling Wang ◽  
Helai Huang
Keyword(s):  
Traffic Safety ◽  
Prediction Models ◽  
Census Bureau ◽  
Integrated Modeling ◽  
Transportation Safety ◽  
Safety Planning ◽  
Contributing Factors ◽  
Modeling Approach ◽  
Exposure Variable ◽  
Safety Models

In recent decades, considerable efforts have been made to incorporate traffic safety into long-term transportation plans (LTTPs), a process which is often termed transportation safety planning (TSP). Although some researchers have attempted to integrate transportation plans and safety by adopting transportation planning data (e.g., trip generation) for estimating traffic crash frequency at the macroscopic level, no studies have attempted to develop trip and safety models in one structure simultaneously. A Bayesian integrated multivariate modeling approach is suggested for estimating trips and crashes of non-motorized modes (i.e., walking and cycling). American Housing Survey (AHS) data were collected from the U.S. Census Bureau and were used for the proposed approach. In the first part of the proposed model, the probabilities of choosing walking and cycling modes were estimated, and the estimated probabilities were converted to trips by multiplying the number of sampled households. In the second part, the estimated trips were fed into crash prediction models (or safety performance functions) as an exposure variable. The modeling result revealed many contributing factors for pedestrian/bicycle trips and crashes. Possible shared unobserved features between pedestrian and bicycle trips, and between pedestrian and bicycle crashes, were accounted for by adopting a multivariate structure. In addition, it was found that the crash models with the estimated exposures outperform those with the observed exposures. It is expected that the integrated modeling approach for trips and crashes in this study will provide great insights into the future directions of TSP.

Developing Macrolevel Collision Prediction Models to Evaluate Bicycle Safety in Vancouver, British Columbia, Canada

2017 ◽  
Vol 2659 (1) ◽  
pp. 25-32
Author(s):  
Bianca Popescu ◽  
Tarek Sayed
Keyword(s):  
British Columbia ◽  
Traffic Congestion ◽  
Road Safety ◽  
Prediction Models ◽  
Safety Management ◽  
Safety Issues ◽  
Bicycle Safety ◽  
Collision Prediction ◽  
High Traffic Volume

To encourage greener cities while reducing the impacts of the transportation system—such as impacts on climate change, traffic congestion, and road safety—governments have been investing in sustainable modes of transportation, such as cycling. A safe and comfortable cycling environment is critical to encourage bicycle trips because cyclists are usually subject to greater safety risks. Engineering approaches to road safety management have traditionally addressed road safety by reacting to existing collision records. For bicycle collisions, which are rare events, a proactive approach is more appropriate. This study described the use of bicycle-related macrolevel (i.e., neighborhood or zonal-level) collision prediction models as empirical tools in road safety diagnosis and planning. These models incorporated an actual bicycle exposure indicator (the number of bicycle kilometers traveled). The macrolevel bicycle–vehicle collisions models were applied at the zonal level to a case study of Vancouver, British Columbia, Canada. Collision-prone zones in Vancouver were identified, and the highest-ranked zones were diagnosed to identify bicycle safety issues and to recommend potential safety countermeasures. The findings from this study suggest that the safety issues may be a result of high density and commercial land use type, coupled with a high traffic volume, particularly on arterial routes, and high bicycle volumes on routes with mixed vehicle and bicycle traffic. The case study demonstrated the use of the models to enhance bicycle safety proactively.

Development of Macro-Level Crime and Collision Prediction Models to Support Data-Driven Approach to Crime and Traffic Safety (DDACTS)

2018 ◽  
Vol 2672 (33) ◽  
pp. 56-66 ◽  
Author(s):  
Emmanuel A. Takyi ◽  
Seun Daniel Oluwajana ◽  
Peter Y. Park
Keyword(s):  
Law Enforcement ◽  
Traffic Safety ◽  
Empirical Bayes ◽  
Negative Binomial ◽  
Prediction Models ◽  
Data Driven ◽  
Macro Level ◽  
Goodness Of Fit Test ◽  
Violent Crimes ◽  
Fatal Injury

The number of violent crimes and fatal-injury collisions concerns many jurisdictions. Traditional enforcement tactics are often reactive, relying on historical crime and collision data to select locations for law enforcement. Advanced law enforcement tactics take a proactive approach. Such tactics include Data-Driven Approaches to Crime and Traffic Safety (DDACTS), which uses predicted numbers of crimes and collisions to identify locations for law enforcement. This DDACTS study was conducted in Regina, Saskatchewan, Canada. The research developed macro-level prediction models to predict violent crimes and collisions in each traffic analysis zone (TAZ) in Regina. The zonal nature of the analysis is important for overcoming confidentiality and privacy issues associated with violent crimes and fatal-injury collisions. Fifty-four input variables were used to describe each TAZ’s crimes, collisions, socio-demographic, road inventory, traffic, and land use characteristics. The analysis used negative binomial regression coupled with the empirical Bayes method (a popular approach in transportation, but relatively new to crime mapping) to develop two statistical models that predict the long-term mean value for the number of violent crimes/collisions per zone. Cumulative residual plots were used as the main goodness-of-fit test. The findings are summarized on a map showing the top ten hotzones for violent crimes, the top ten hotzones for fatal-injury collisions, and the zones where the crime and collisions zones overlap. The overlapping zones are the DDACTS zones. By focusing law enforcement in the DDACTS zones, it may be possible to reduce violent crimes and fatal-injury collisions simultaneously and use limited resources more cost effectively.

Sign in / Sign up

Export Citation Format

Share Document