The Effect of Four-Lane to Three-Lane Conversion on the Number of Crashes and Crash Rates in Iowa Roads, June 2005

We analyze crash data collected by the Iowa Department of Transportation using Bayesian methods. The data set includes monthly crash numbers, estimated monthly traffic volumes, site length and other information collected at 30 paired sites in Iowa over more than 20 years during which an intervention experiment was set up. The intervention consisted in transforming 15 undivided road segments from four-lane to three lanes, while an additional 15 segments, thought to be comparable in terms of traffic safety-related characteristics were not converted. The main objective of this work is to find out whether the intervention reduces the number of crashes and the crash rates at the treated sites. We fitted a hierarchical Poisson regression model with a change-point to the number of monthly crashes per mile at each of the sites. Explanatory variables in the model included estimated monthly traffic volume, time, an indicator for intervention reflecting whether the site was a “treatment” or a “control” site, and various interactions. We accounted for seasonal effects in the number of crashes at a site by including smooth trigonometric functions with three different periods to reflect the four seasons of the year. A change-point at the month and year in which the intervention was completed for treated sites was also included. The number of crashes at a site can be thought to follow a Poisson distribution. To estimate the association between crashes and the explanatory variables, we used a log link function and added a random effect to account for overdispersion and for autocorrelation among observations obtained at the same site. We used proper but non-informative priors for all parameters in the model, and carried out all calculations using Markov chain Monte Carlo methods implemented in WinBUGS. We evaluated the effect of the four to three-lane conversion by comparing the expected number of crashes per year per mile during the years preceding the conversion and following the conversion for treatment and control sites. We estimated this difference using the observed traffic volumes at each site and also on a per 100,000,000 vehicles. We also conducted a prospective analysis to forecast the expected number of crashes per mile at each site in the study one year, three years and five years following the four to three-lane conversion. Posterior predictive distributions of the number of crashes, the crash rate and the percent reduction in crashes per mile were obtained for each site for the months of January and June one, three and five years after completion of the intervention. The model appears to fit the data well. We found that in most sites, the intervention was effective and reduced the number of crashes. Overall, and for the observed traffic volumes, the reduction in the expected number of crashes per year and mile at converted sites was 32.3% (31.4% to 33.5% with 95% probability) while at the control sites, the reduction was estimated to be 7.1% (5.7% to 8.2% with 95% probability). When the reduction in the expected number of crashes per year, mile and 100,000,000 AADT was computed, the estimates were 44.3% (43.9% to 44.6%) and 25.5% (24.6% to 26.0%) for converted and control sites, respectively. In both cases, the difference in the percent reduction in the expected number of crashes during the years following the conversion was significantly larger at converted sites than at control sites, even though the number of crashes appears to decline over time at all sites. Results indicate that the reduction in the expected number of sites per mile has a steeper negative slope at converted than at control sites. Consistent with this, the forecasted reduction in the number of crashes per year and mile during the years after completion of the conversion at converted sites is more pronounced than at control sites. Seasonal effects on the number of crashes have been well-documented. In this dataset, we found that, as expected, the expected number of monthly crashes per mile tends to be higher during winter months than during the rest of the year. Perhaps more interestingly, we found that there is an interaction between the four to three-lane conversion and season; the reduction in the number of crashes appears to be more pronounced during months, when the weather is nice than during other times of the year, even though a reduction was estimated for the entire year. Thus, it appears that the four to three-lane conversion, while effective year-round, is particularly effective in reducing the expected number of crashes in nice weather.

Effect of Wildlife Highway Warning Reflectors on Deer-Vehicle Accident, December 1984

he number of deer-vehicle accidents in Iowa and around the country has steadily increased during the past 30 years. This i s basically due to: ( 1 ) increased volume of traffic; 12) an expanding network of hard surface roads, especially 4 lane interstates; and (3) a general increase in deer populations. Initidtion of a 55 MPH speed limit in 1974 and gasoline shortages in 1975 reduced deer-vehicle accident rates briefly, but since 1975, rates have continued to climb. Various methods of reducinq these accidents have been attempted in other states. These include: instal lation of rc?flective devlres, deer crossing signs, fencing, underpasses, clearing right--of--waysa,n d controlled hunting to reduce deer population s i z e . These methods have met with varying degrees of success, depending on animal behavior, deet- population fluctuations, method used, topoyr-aphy, road-side vegetation, traffic patterns, and highway configuration. This project was designed to evaluate a new ntethod of reducing deer-vehicle accidents. There are qenerally 4 important aspects of deer-vehicle accidents: danger to human l i f e , vehicle damage, loss of a valuable wildlife resource, and cost of processing accident reports. In !owe, during 1983, there were over 15,OOC) reported deer--vehicle accidents and probably many more that were not reported (Gladfelter 1984). The extent of human injury or death in Iowa i s not known, but studies in southern Michigan show that human injur ies occurred in about 4% of the deer-vehicle accidents (A1 lcn and MrCullough 1976). T h i s would indicate that in Iowa there could have been 200 human injury cases from deer-vehicle accidents i n 1983. These injuries usual 1 occur from secondary collisions when motorists try to avoid a deer on the highway, and hit some other object Vehicle darnaye from these accidents can into thousands of dollars because of the high speed involved and the size of the animal. The total amount of vehicle damage occurring in Iowa is unknown, but if the average vehicle damage was between $500-$800 per accident, estimated property damage would be between $2 1/2--$4 million annually. The value of deer lost in these accidents cannot be estimated, but recreational potential of this natural resource is surely diminished for hunters and wildlife enthusiasts. Also, there ir a great deal of money spent by governmental agencies for manpower to process accident reports and remove dead animals from highways.

Effect of Video Camera-Based Remote Roadway Condition Monitoring on Snow Removal-Related Maintenance Operations. 2014

Remote monitoring through the use of cameras is widely utilized for traffic operation, but has not been utilized widely for roadway maintenance operations. The Utah Department of Transportation (UDOT) has implemented a new remote monitoring system, referred to as a Cloud-enabled Remote Video Streaming (CRVS) camera system for snow removal-related maintenance operations in the winter. The purpose of this study was to evaluate the effectiveness of the use of the CRVS camera system in snow removal-related maintenance operations. This study was conducted in two parts: opinion surveys of maintenance station supervisors and an analysis on snow removal-related maintenance costs. The responses to the opinion surveys mostly displayed positive reviews of the use of the CRVS cameras. On a scale of 1 (least effective) to 5 (most effective), the average overall effectiveness given by the station supervisors was 4.3. An expedition trip for this study was defined as a trip that was made to just check the roadways if snow-removal was necessary. The average of the responses received from surveys was calculated to be a 33 percent reduction in expedition trips. For the second part of this study, an analysis was performed on the snow removal-related maintenance cost data provided by UDOT to see if the installation of a CRVS camera had an effect in reducing expedition trips. This expedition cost comparison was performed for 10 sets of maintenance stations within Utah. It was difficult to make any definitive inferences from the comparison of expedition costs over the years for which precipitation and expedition cost data were available; hence a statistical analysis was performed using the Mixed Model ANOVA. This analysis resulted in an average of 14 percent higher ratio of expedition costs at maintenance stations with a CRVS camera before the installation of the camera compared to the ratio of expedition costs after the installation of the camera. This difference was not proven to be statistically significant at the 95 percent confident level, but indicated that the installation of CRVS cameras was on the average helpful in reducing expedition costs and may be considered practically significant. It is recommended that more detailed and consistent maintenance cost records be prepared for accurate analysis of cost records for this type of study in the future.

Effect of a Noise Wall on Snow Accumulation and Air Quality, 1985

A noise wall was investigated to assess its effect on snow accumulation and air quality. Wind tunnel studies were undertaken to evaluate (a) possible snow accumulations and (b) the dispersion of particulate concentrations (dust, smoke, and lead particles) and carbon monoxide. Full-scale monitoring of particulate concentrations and carbon monoxide was performed both before and after the noise wall was constructed. The wind tunnel experiments for snow accumulation were conducted on a model wall located in a flat, unobstructed area. A separated flow zone existed upwind of the wall and snow immediately began to accumulate over most of the separated zone. Having the noise wall in an aerodynamically rough area, such as in an urban area as this one was, substantially decreased the amount of snow collected, compared with in the wind tunnel studies, because of turbulence reducing the separation zone. The snow accumulation has not been significantly greater with the noise wall in place than it was before construction and has proven to be of no concern to date. Monitoring for particulate concentrations has shown that the noise wall has had a beneficial effect because the amount of material collected was reduced. With the noise wall in place, monitoring for carbon monoxide has indicated that (a) for equivalent emissions under conditions of high atmospheric stability and low wind speeds, the carbon monoxide levels would be lower; and (b) under conditions of low atmospheric stability and high wind speeds, the carbon monoxide levels would be higher than expected without the wall in place.