This poster session will cover many of the aspects of how road weather impacts the roadway, drivers, and the transporting of goods and services.
A Spatial Panel Regression to Measure the Effect of Weather Events on Freight Truck Traffic
Taslima Akter, University of Arkansas, FayettevilleShow Abstract
Suman Mitra, University of California, Irvine
Sarah Hernandez, University of Arkansas, Fayetteville
Karla Diaz-Corro, University of Arkansas, Fayetteville
Severe weather conditions can have major effects on traffic volumes. Unlike passenger vehicles, which may choose not to travel during inclement weather, freight trucks adhere to delivery schedules requiring them to alter their route rather than cancel a trip. While previous studies modeled the effects of weather on total traffic volumes, very few have examined the effect of weather on truck volumes. The study of weather effects on truck volumes requires advanced modeling techniques that are able to capture effects over space. This study applies spatial panel regression techniques to develop a predictive model that relates variations in truck traffic patterns to weather conditions. Truck volume data from 18 Weigh-in-Motion (WIM) stations in Arkansas and weather data from the Long Term Pavement Performance (LTPP) InfoPave Climate Tool were used. Through the random effect spatial panel model, the significant negative value of Moran’s I explains that the change in daily truck volume is spatially dispersed. The study finds that weather variables have both direct and indirect impacts on daily truck volumes, e.g. the presence of snowfall at a station directly reduces daily truck volumes at that station by approximately 11.3% and indirectly reduces daily truck volumes at neighboring stations by 2.8%, compared to the average daily truck traffic. The model developed in this study can assist state and regional transportation agencies in developing freight-oriented programs and policies for road and winter maintenance, structural and geometric pavement design, highway life cycle analysis, and long range transportation planning.
Investigating the Impact of Rain on Crash Clearance Duration
Henrick Haule, Florida International UniversityShow Abstract
Priyanka Alluri, Florida International University
Thobias Sando, University of North Florida
MD Asif Raihan, Florida International University
Crashes are one of the major causes of traffic delays on freeways. It is essential to clear crashes as quickly as possible from highways irrespective of the prevailing weather conditions. However, rainy conditions could potentially influence the crash clearance duration. The goal of the study was to develop a methodology that will improve the accuracy pertaining to the evaluation of the impact of rain on crashes and crash clearance timeline. The specific objectives included: (a) estimate the duration of rain within the crash clearance time; and (b) evaluate the impact of crash-related, spatiotemporal, and response agencies’ attributes on the crash clearance time during rainy conditions. The crash data and rain data were extracted from SunGuide® and National Oceanic and Atmospheric Administration (NOOA) databases, respectively. The crash data were collected for the years 2014 to 2016 on a freeway network of Interstates (I-95, I-295, and I-10) and State road (SR-202) in Jacksonville, Florida. The rain data were collected hourly from 2014 to 2016. The study estimated the rain duration within the crash clearance time. Hazard-based models were used to investigate factors that influence the crash clearance time during rain. The results indicated that crash severity, extent of rain duration, time of day, day of the week, area type, and involvement of Emergency Medical Services were significant factors to the crash clearance time during rainy conditions. The study results assist incident management agencies in advancing strategies to reduce crash clearance time during adverse weather conditions.
Winter Weather Maintenance Operations and Traffic Safety Implications
Luning Zhang, Iowa State UniversityShow Abstract
Bryce Hallmark, Iowa State University
Jing Dong, Iowa State University
Many past studies relating to winter storm events have been focused on safety and mobility over the winter season as a whole. As more granular data has become available, studies have narrowed in on specific snow or winter storm events in an attempt to better quantify the impact of adverse winter weather on safety and mobility. Up to this point, few studies have attempted to quantify roadway safety by using real-time data stemming from winter maintenance vehicles, weather data, and roadway data. This study aims to identify specific winter storm events and link these to winter maintenance operation characteristics in order to determine their effect on roadway safety. The main criteria used to quantify roadway safety include traffic volumes, crash counts, roadway conditions, snowplow passes and material spreading. A list of winter storm events was identified and populated with pertinent criteria. A Negative Binomial model is estimated to describe the relationship between crash frequency and the influencing factors. It is found that the weather and road surface condition was statistically significant in influencing the crash occurrence.
Nueral Networks to Predict Visibility: Application at Link Level
Venkata Duddu, University of North Carolina, CharlotteShow Abstract
Srinivas Pulugurtha, University of North Carolina, Charlotte
Ajinkya Mane, University of North Carolina, Charlotte
Christopher Godfrey, University of North Carolina, Asheville
Extreme weather condition such as dense fog and heavy rain can directly affect visibility and, hence, the level of safety on a road. Fog is a highly localized phenomenon, and it would be difficult and expensive to install visibility sensors every few kilometers along roads. Therefore, the objective of this paper is to develop a model to predict fog / lower visibility condition using machine learning technique such as neural network, and, apply at link-level. Five years of meteorological data was collected from all the weather stations in and around the state of North Carolina (NC). Back-propagation neural network was developed and validated to predict the visibility condition and apply at link-level. The results obtained indicate that the neural network has better prediction capability for visibility during normal foggy condition when compared to visibility during very dense/thick fog condition.
Multimodal Connected Vehicle Technology for Improved Winter Travel
Yaqin He, Wuhan University of TechnologyShow Abstract
Michelle Akin, Washington State University
Xianming Shi, Washington State University
Winter weather impacts the safety, mobility and economic productivity of surface transportation systems. Snow and ice on roads reduces friction and visibility, contributes to accidents and decreases traffic speed, flow, volume and capacity. Connected Vehicle (CV) technology is well-suited to address multiple safety and mobility impacts of winter weather. Accurate and real time traffic information, route operation information and road weather information are essential for commuters to take a high level-of-service trip. A nationwide survey of transit operation managers or supervisors was conducted to assess priorities for transit applications of CV technologies to improve safety and mobility during winter. A summary of findings from the survey includes: 1) snow build-up at stops, route delays, and route changes/cancellations are significant impacts; 2) most transit agencies use road weather information and forecasts to improve transit operations, 3) the transit CV applications thought to have the most potential for improving winter transit travel are pedestrian warning and left turn assistance warning, and 4) the greatest concerns for CV technology are increased driver distraction, safety consequences of equipment failure, and system performance in poor weather. Two applications of CV technology for improved multimodal winter travel are presented in the Concept of Operations section: 1) CV applications in winter transit operations--- mobile road weather-related and traffic flow route-specific data to be used by transit operators to provide real-time and forecast route information. 2) CV applications in multimodal commuter winter travel advisories and warnings---provide the road segment weather information and transit route information to commuters.
Use of Topography, Weather Zones, and Semivariogram Parameters to Optimize RWIS Station Density Across Large Spatial Scales
Simita Biswas, University of AlbertaShow Abstract
Mingjian Wu, University of Alberta
Stephanie J. Melles, Ryerson University
Tae Kwon, University of Alberta
A Road Weather Information System (RWIS) is a combination of advanced technologies that collect, process, and disseminate road weather and condition information. This information is used by road maintenance authorities to make operative decisions that improve safety and mobility during inclement weather events. Many North American transportation agencies have invested millions of dollars to deploy RWIS stations to indeed improve the monitoring coverage of winter road surface conditions. However, the design of these networks often varies by region, and it is not entirely clear how many stations are necessary to provide adequate monitoring coverage under different conditions: substantial gaps in knowledge about optimal designs remain. To fill these gaps, an investigation was done to determine how optimized RWIS station densities relate to topographic and weather characteristics. A series of geostatistical semivariogram models were constructed and compared using topographic position index (TPI) and weather severity index (WSI) to measure relative topographic variation and weather severity, respectively. The geostatistical approach was then applied to map the optimum number of RWIS stations across several topographic and weather zones. The study area captured varying environmental characteristics, including regions with flat or varied terrain and warm or cold regions. This study suggests that RWIS data collected from a specific region can be used to estimate the number of stations required for regions with similar zonal characteristics. The outcome of this study can be used as a decision-making tool for RWIS network expansion, thus maximizing RWIS network monitoring capability using topographic and weather-related zonal classifications.