Using conditional inference forests to identify the factors affecting crash severity on arterial corridors

Introduction

The study aims at identifying traffic/highway design/driver-vehicle information significantly related with fatal/severe crashes on urban arterials for different crash types. Since the data used in this study are observational (i.e., collected outside the purview of a designed experiment), an information discovery approach is adopted for this study.

Method

Random Forests, which are ensembles of individual trees grown by CART (Classification and Regression Tree) algorithm, are applied in numerous applications for this purpose. Specifically, conditional inference forests have been implemented. In each tree of the conditional inference forest, splits are based on how good the association is. Chi-square test statistics are used to measure the association. Apart from identifying the variables that improve classification accuracy, the methodology also clearly identifies the variables that are neutral to accuracy, and also those that decrease it.

Results

The methodology is quite insightful in identifying the variables of interest in the database (e.g., alcohol/ drug use and higher posted speed limits contribute to severe crashes). Failure to use safety equipment by all passengers and presence of driver/passenger in the vulnerable age group (more than 55 years or less than 3 years) increased the severity of injuries given a crash had occurred. A new variable, ‘element’ has been used in this study, which assigns crashes to segments, intersections, or access points based on the information from site location, traffic control, and presence of signals.

Impact

The authors were able to identify roadway locations where severe crashes tend to occur. For example, segments and access points were found to be riskier for single vehicle crashes. Higher skid resistance and k-factor also contributed toward increased severity of injuries in crashes.     

An epidemiologic comparison of injuries presenting to a pediatric emergency department and local urgent care facilities

Problem

The objective of this study was to compare the epidemiology of injuries presenting to emergency department (ED) and urgent care (UC) facilities of a single, NEISS-affiliated hospital.

Method

Patient medical records (n = 36,811) were used to compare injury incidence, injury characteristics, and demographic characteristics between the ED, on-site UC, and off-site UC during 2006.

Results

ED presentations were more likely to be open wounds and motor vehicle-related compared to on-site UC presentations. ED presentations were more likely to be system wide/late effects, be made by an African American, or be paid through Medicaid compared to off-site UC presentations. On-site UC presentations were more likely to be made by an African American or be paid through Medicaid compared to off-site UC presentations.

Discussion

ED and UC injury characteristics and patient demographics differ. With no nationally-representative UC injury surveillance, current research likely underestimates injury incidence and presents skewed profiles. Impact on Industry: This article adds insight into the generalizability of ED-based injury surveillance to UC injuries.     

An economic evaluation of a participatory ergonomics process in an auto parts manufacturer

Problem

We assess the costs and consequences of a participatory ergonomics process at a Canadian car parts manufacturer from the perspective of the firm.

Method

Regression modeling was used with interrupted time series data to assess the impact of the process on several health measures. Consequences were kept in natural units for cost-effectiveness analysis, and translated into monetary units for cost-benefit analysis.

Results

The duration of disability insurance claims and the number of denied workers' compensation claims was significantly reduced. The cost-effectiveness ratio is $12.06 per disability day averted. The net present value is $244,416 for a 23-month period with a benefit-to-cost ratio of 10.6, suggesting that the process was worth undertaking (monetary units in 2001 Canadian dollars).

Discussion

Our findings emphasize the importance of considering a range of outcomes when evaluating an occupational health and safety intervention.

Impact on industry

Participatory ergonomics process can be cost-effective for a firm.     

Occupational injuries of immigrant and Finnish bus drivers

Problem

With baby boomers reaching retirement age, Western countries may need more immigrant workers to ensure productivity. Many studies have suggested a higher occupational injury frequency among immigrant workers, which could considerably reduce their contribution to society. The aim of this study was to examine whether immigrant workers have a higher injury frequency compared to Finnish workers when performing the exact same tasks under the same working conditions.

Method

A total of 176 Finnish and 130 immigrant bus drivers were asked about their occupational injuries during the past 12 months via a questionnaire. In addition, the data contained 134 injuries reported by the transport firm to an insurance company.

Results

There was no significant difference in reporting occupational injuries by self-reporting or by company-records. Because there were more accident-repeaters among Finnish drivers, their injury frequency (114) was higher than that of immigrant drivers (78).

Application/Impact

This study showed that immigrant workers did not have a higher injury frequency than other workers when they worked in the exact same conditions. Immigrant workers can work as safely as native-Finnish workers, when their working conditions and job contracts are at the same level as those of the original population. Immigrant workers can compensate for the shortage of workforce caused by an aging population.     

Marketing and alcohol-related traffic fatalities: Impact of alcohol advertising targeting minors

Problem

Alcohol-related youth traffic fatalities continue as a major public-health concern. While state and federal laws can be useful in tackling this problem, the efficacy of many laws has not been empirically demonstrated. We examined the impact of state laws prohibiting alcohol advertising to target minors. Method: Using statistics obtained from the Fatality Analysis Reporting System (FARS), youth alcohol-related, single-vehicle, driver traffic fatalities were compared by state as a function of whether the state has a law prohibiting alcohol advertising that targets minors. Results: Overall, states possessing this law experienced 32.9% fewer of the above specified traffic fatalities. Discussion and Impact on Industry: The results suggest that not only are youth drinking rates affected by alcohol advertisements targeting youth, but also drink-driving behaviors. Indeed, we estimate that if this type of legislation were adopted in the 26 states that do not prohibit targeting of minors with alcohol advertising, then 400 youth lives could be saved annually.     

Methodology of CO2 emission evaluation in the life cycle of office building façades

The construction industry is one of the greatest sources of pollution because of the high level of energy consumption during its life cycle. In addition to using energy while constructing a building, several systems also use power while the building is operating, especially the air-conditioning system. Energy consumption for this system is related, among other issues, to external air temperature and the required internal temperature of the building. The façades are elements which present the highest level of ambient heat transfer from the outside to the inside of tall buildings. Thus, the type of façade has an influence on energy consumption during the building life cycle and, consequently, contributes to buildings' CO₂ emissions, because these emissions are directly connected to energy consumption. Therefore, the aim is to help develop a methodology for evaluating CO₂ emissions generated during the life cycle of office building façades. The results, based on the parameters used in this study, show that façades using structural glazing and uncolored glass emit the most CO₂ throughout their life cycle, followed by brick façades covered with compound aluminum panels or ACM (Aluminum Composite Material), façades using structural glazing and reflective glass and brick façades with plaster coating. On the other hand, the typology of façade that emits less CO₂ is brickwork and mortar because its thermal barrier is better than structural glazing façade and materials used to produce this façade are better than brickwork and ACM. Finally, an uncertainty analysis was conducted to verify the accuracy of the results attained.     

Continuous high-frequency monitoring of estuarine water quality as a decision support tool: a Dublin Port case study

High-frequency, continuous monitoring using in situ sensors offers a comprehensive and improved insight into the temporal and spatial variability of any water body. In this paper, we describe a 7-month exploratory monitoring programme in Dublin Port, demonstrating the value of high-frequency data in enhancing knowledge of processes, informing discrete sampling, and ultimately increasing the efficiency of port and environmental management. Kruskal–Wallis and Mann–Whitney tests were used to show that shipping operating in Dublin Port has a small–medium effect on turbidity readings collected by in situ sensors. Turbidity events are largely related to vessel activity in Dublin Port, caused by re-suspension of sediments by vessel propulsion systems. The magnitudes of such events are strongly related to water level and tidal state at vessel arrival times. Crucially, measurements of Escherichia coli and enterococci contamination from discrete samples taken at key periods related to detected turbidity events were up to nine times higher after vessel arrival than prior to disturbance. Daily in situ turbidity patterns revealed time-dependent water quality “hot spots” during a 24-h period. We demonstrate conclusively that if representative environmental assessment of water quality is to be performed at such sites, sampling times, informed by continous monitoring data, should take into account these daily variations. This work outlines the potential of sensor technologies and continuous monitoring, to act as a decision support tool in both environmental and port management.     

Evaluation of a high visibility enforcement project focused on passenger vehicles interacting with commercial vehicles

Introduction

In 2004, Washington State applied NHTSA's High Visibility Enforcement model used in the Click It or Ticket seat belt campaign in an attempt to reduce unsafe driving behaviors around commercial motor vehicles (CMVs). The program was called Ticketing Aggressive Cars and Trucks (TACT). This paper details the methods used to evaluate the program's effectiveness and the results of the evaluation.

Method

Four high-crash interstate highway corridors, each approximately 25 miles in length, were selected. Two of these corridors received TACT media messages and increased enforcement over an 18-month period while two comparison corridors did not receive any increased media or enforcement.

Results

A total of 4,737 contacts were made with drivers during the two enforcement waves, and 72% of these contacts led to a citation. Drivers at the intervention sites who said they saw or heard any of the TACT messages increased from 17.7% in the pre period to a high of 67.3% in the post periods. Drivers at the intervention sites also reported increased exposure to the core message of leaving more space when passing trucks (14% pre to 40% post period). The percentage of drivers who said they leave more room when passing trucks than when passing cars rose from 16% in the pre period to 24% in the post period at the intervention sites, while comparison sites showed no change. Over 150 hours of video recorded by law enforcement officers in unmarked vehicles were utilized to examine violation rates and severity of violations before and after the intervention campaigns. Statistical analyses showed that violation rates were reduced significantly at the intervention sites (between 23% and 46%), while remaining constant at the comparison sites. Analyses of the video data also showed that the seriousness of the residual violations at the intervention sites decreased.

Conclusions

Overall, the evaluation results provide a consistent picture of the effectiveness of the TACT pilot project. Success was demonstrated at every step - messages were received and understood, knowledge was changed in the intended direction, self reported driving behavior around large trucks improved, and observed driving behaviors confirmed the self reports.

Impact on Industry

After this initial success in Washington State, the TACT model will continue to be implemented and evaluated by FMCSA in an attempt to validate the program. Based on the results of this study and the consistent positive results found for other sTEP projects, it is likely that TACT will show continued success in a variety of settings and will help reduce the number and severity of crashes involving CMVs. Future research should attempt to use many of the methods described here to further validate the methods for not only evaluations of TACT programs, but also for any other highway safety programs that require measurements of the program's effectiveness.     

Traffic speeds associated with implementation of 80 mph speed limits on West Texas rural interstates

Problem

In 2006 Texas raised the daytime speed limit for passenger vehicles on segments of I-10 and I-20 from 75 to 80 mph. Methods: Traffic speeds were measured before and 3, 12, and 16 months after the limit was changed. Results: During the 16-month period following the speed limit increase, mean speeds of passenger vehicles on I-20 increased by 9 mph relative to the comparison road, where no speed limit change occurred and traffic speeds declined. On I-10 mean speeds increased by 4 mph relative to the comparison road. Limiting the analysis to the month before the speed limit change and 1 year later, the proportion of drivers exceeding 80 mph was 18 times higher on I-20 and 2 times higher on I-10. Discussion: The smaller speed increases on I-10 may be related to its proximity to the U.S. border with Mexico. Highly visible border patrol activity coincided with posting of the higher speed limit. Long-term monitoring in other states suggests that traffic speeds in Texas are likely to continue to increase. Impact on Industry: The present study adds to the wealth of evidence that increased speed limits lead to increased travel speeds. The primary countermeasures to reduce the risk of speed-related crashes include highly visible police traffic enforcement and the use of speed cameras accompanied by publicity.