VTechWorks Repository :: Browsing by Author "Jahangiri, Arash"

Browsing by Author "Jahangiri, Arash"

Now showing 1 - 16 of 16

Analyzing Highway Safety Datasets: Simplifying Statistical Analyses from Sparse to Big Data
Lord, Dominique; Geedipally, Srinivas Reddy; Guo, Feng; Jahangiri, Arash; Shirazi, Mohammadali; Mao, Huiying; Deng, Xinwei (SAFE-D: Safety Through Disruption National University Transportation Center, 2019-07)
Data used for safety analyses have characteristics that are not found in other disciplines. In this research, we examine three characteristics that can negatively influence the outcome of these safety analyses: (1) crash data with many zero observations; (2) the rare occurrence of crash events (not necessarily related to many zero observations); and (3) big datasets. These characteristics can lead to biased results if inappropriate analysis tools are used. The objectives of this study are to simplify the analysis of highway safety data and develop guidelines and analysis tools for handling these unique characteristics. The research provides guidelines on when to aggregate data over time and space to reduce the number of zero observations; uses heuristics for selecting statistical models; proposes a bias adjustment method for improving the estimation of risk factors; develops a decision-adjusted modeling framework for predicting risk; and shows how cluster analyses can be used to extract relevant information from big data. The guidelines and tools were developed using simulation and observed datasets. Examples are provided to illustrate the guidelines and tools.
Behavior-Based Predictive Safety Analytics Phase II
Miller, Andrew M.; Sarkar, Abhijit; McDonald, Tony; Ghanipoor-Machiani, Sahar; Jahangiri, Arash (Safe-D National UTC, 2023-06)
This project addressed the emerging field of behavior-based predictive safety analytics, focusing on the prediction of road crash involvement based on individual driver behavior characteristics. This has a range of applications in the areas of fleet safety management and insurance, but may also be used to evaluate the potential safety benefits of an automated driving system. This project continued work from a pilot study that created a proof-of-concept demonstration on how crash involvement may be predicted on the basis of individual driver behavior utilizing naturalistic data from the Second Strategic Highway Research Program. The current project largely focuses on understanding and identifying the risks from a driver based on their driving behaviors, personal characteristics, and environmental influences. This project analyzed large scale continuous naturalistic data as well as event data to study the role of different driving behaviors in the buildup of risk related to a safety-critical event or crash. This research can be used structure the development of real-time crash risk that accounts for those identified driver behaviors to be evaluated across the contextualized information on a roadway.
Behavior-based Predictive Safety Analytics – Pilot Study
Engström, Johan; Miller, Andrew M.; Huang, Wenyan; Soccolich, Susan A.; Machiani, Sahar Ghanipoor; Jahangiri, Arash; Dreger, Felix; de Winter, Joost (SAFE-D: Safety Through Disruption National University Transportation Center, 2019-04)
This report gives an overview of the main findings from the Behavior-based Predictive Safety Analytics – Pilot Study project. The main objective of the project was to investigate the possibilities of developing statistical models predicting individual driver crash involvement based on individual driving style, demographic and behavioral history variables, using large sets of naturalistic driving data. The project was designed as a pilot project with the objective of providing the basis for a future more comprehensive research effort. Based on Second Strategic Highway Research Program (SHRP2) data, a subset of behavior and crash data including 2,458 drivers was created for analysis. The data were analyzed to investigate to what extent these drivers were differentially involved in crashes and near crashes, to what extent this was associated with individual characteristics, and if it is possible to predict individual drivers’ crash and near crash involvement based on variables representing individual characteristics. The results clearly demonstrated the presence of differential crash and near crash involvement and showed significant associations between enduring personal factors and crash involvement. Moreover, logistic regression and random forest classifiers were relatively successful in predicting crash and near crash involvement based on individual characteristics, but the ability to specifically predict involvement in crashes was more limited.
Bicycle Naturalistic Data Collection
Elhenawy, Mohammed; Jahangiri, Arash; Rakha, Hesham A. (Connected Vehicle/Infrastructure University Transportation Center (CVI-UTC), 2016-06-15)
Recently, bicycling has drawn more attention as a sustainable and eco-friendly mode of transportation. Between 2000 and 2011, bicycle commuting rates in the United States rose by 80% in large bicycle friendly cities (BFCs), by 32% in non-BFCs, and overall by 47%. On the other hand, about 700 cyclists are killed and nearly 50,000 are injured annually in bicycle–motor vehicle crashes in recent years in the United States. More than 30% of cyclist fatalities in the United States from 2008 to 2012 occurred at intersections, and up to 16% of bicycle-related crashes were due to cyclists’ violations at intersections. In light of these statistics, this project focused on investigating factors that affect cyclist behavior and predicting cyclist violations at intersections. Naturalistic cycling data were used to assess the feasibility of developing cyclist violation prediction models. Mixed-effects generalized regression model is used to analyze the data and identify the significant factor affecting the probability of violations by cyclists. At signalized intersections, right turn, side traffic and opposing traffic are statistically significant factors affecting the probability of red light violation. At stop-controlled intersections, the presence of other road users, left turn, right turn and warm weather are statistically significant factors affecting the probability of violations. Violation prediction models were developed for stop-controlled intersections based on kinetic data measured as cyclists approached the intersection. Prediction error rates were 0% to 10%, depending on how far from the intersection the prediction task was conducted. An error rate of 6% was obtained when the violating cyclist was at a time-to-intersection of about 2 seconds, which is sufficient for most motor vehicle drivers to respond.
Big Data Visualization and Spatiotemporal Modeling of Risky Driving
Jahangiri, Arash; Marks, Charles; Machiani, Sahar Ghanipoor; Nara, Atsushi; Hasani, Mahdie; Cordova, Eduardo; Tsou, Ming-Hsiang; Starner, Joshua (SAFE-D: Safety Through Disruption National University Transportation Center, 2020-07)
Statistical evidence shows the role of risky driving as a contributing factor in roadway collisions, highlighting the importance of identifying such driving behavior. With the advent of new technologies, vehicle kinematic data can be collected at high frequency to enable driver behavior monitoring. The current project aims at mining a huge amount of driving data to identify risky driving behavior. Relational and non-relational database management systems (DBMSs) were adopted to process this big data and compare query performances. Two relational DBMSs, PostgreSQL and PostGIS, performed better than a non-relational DBMS, MongoDB, on both nonspatial and spatial queries. Supervised and unsupervised learning methods were utilized to classify risky driving. Cluster analysis as an unsupervised learning approach was used to label risky driving during short monitoring periods. Labeled driving data, including kinematic information, were employed to develop random forest models for predicting risky driving. These models showed high prediction performance. Open source and enterprise visualization tools were also developed to illustrate risky driving moments in space and time. These tools can be used by researchers and practitioners to explore where and when risky driving events occur and prioritize countermeasures for locations in highest need of improvement.
Connected Vehicle Applications for Adaptive Overhead Lighting (On-demand Lighting)
Gibbons, Ronald B.; Palmer, Matthew; Jahangiri, Arash (Connected Vehicle/Infrastructure University Transportation Center (CVI-UTC), 2016-07-01)
The Virginia Tech Transportation Institute (VTTI) has developed an on-demand roadway lighting system and has tested the system’s effect on driver visual performance. On-demand roadway lighting can dramatically reduce energy usage while maintaining or increasing vehicle and pedestrian safety. The system developed by VTTI uses connected vehicle technology (CVT), wireless lighting controls, LED luminaires, and a stand-alone processor on the Virginia Smart Road to sense vehicles and turn on roadway lighting only when needed. During this research project, the use of on-demand, or just-in-time, lighting was investigated with respect to assessing driver distraction, and to human factors, including a driver’s ability to visually detect and recognize on-road objects and pedestrians. The developed on-demand lighting system described above utilized dedicated short range communication (DSRC), connected vehicle infrastructure (CVI), and centralized wireless lighting controls, and was used with VTTI-developed in-vehicle instrumentation and custom software. The software allowed the study of forward preview time in terms of forward lighting distance needed for drivers to detect roadside pedestrians and hazards. Visual performance testing revealed a relationship between speed and the amount of forward lighting needed to detect pedestrians and hazards on the side of the roadway, and a small, but statistically insignificant, practical difference in visual performance between on-demand lighting and continuously-on lighting conditions. A survey of participant reactions indicated that the public generally accepts on-demand lighting and does not find it distracting as long as a minimum lighting condition is met. The survey also found that participants felt the system provided a safe driving environment. The main application for an on-demand lighting system would be on roadways with little traffic at night and higher accident rates, or higher conflict areas such as intersections, pedestrian crossings, and merge areas.
Data Mining to Improve Planning for Pedestrian and Bicyclist Safety
Jahangiri, Arash; Hasani, Mahdie; Sener, Ipek Nese; Munira, Sirajum; Owens, Justin M.; Appleyard, Bruce; Ryan, Sherry; Turner, Shawn M.; Machiani, Sahar Ghanipoor (SAFE-D: Safety Through Disruption National University Transportation Center, 2019-11)
Between 2009 and 2016, the number of pedestrian and bicyclist fatalities saw a marked trend upward. Taken together, the overall percentage of pedestrian and bicycle crashes now accounts for 18% of total roadway fatalities, up from 13% only a decade ago. Technological advancements in transportation have created unique opportunities to explore and investigate new sources of data for the purpose of improving safety planning. This study investigated data from multiple sources, including automated pedestrian and bicycle counters, video cameras, crash databases, and GPS/mobile applications, to inform bicycle and pedestrian safety improvements. Data mining techniques, a new sampling strategy, and automated video processing methods were adopted to demonstrate a holistic approach that can be applied to identify facilities with highest need of improvement. To estimate pedestrian and bicyclist counts at intersections, exposure models were developed incorporating explanatory variables from a broad spectrum of data sources. Intersection-related crashes and estimated exposure were used to quantify risk, enabling identification of high-risk signalized intersections for walking and bicycling. The modeling framework and data sources used in this study will be beneficial in conducting future analyses for other facility types, such as roadway segments, and also at more aggregate levels, such as traffic analysis zones.
Developing a Framework for Prioritizing Bicycle Safety Improvement Projects using Crowdsourced and Image-Based Data
Sadeghi, Amir Reza; Jahangiri, Arash; Machiani, Sahar Ghanipoor; Hankey, Steve; Abdollahpour, Seyed Sajjad (Safe-D University Transportation Center, 2023-08)
Active transportation, including walking and cycling, has gained popularity due to the economic, environmental, and energy-efficient benefits. However, the rise of active transportation has also led to an increase in fatalities, particularly for bicyclists. A crash-risk scoring method was proposed to prioritize bicycle safety improvement projects for 50 bridges located in San Diego County. This study employs surrogate safety measures to estimate crash risk, addressing the limitations of traditional data collection methods, and incorporates transportation equity factors into the safety measure scoring method. To identify significant factors contributing to the likelihood of bicyclists exceeding 10 mph on bridges, binomial logistic regression models were employed, with three models focusing on different predictor variables. The results showed that factors such as race, home-to-work travel patterns, education levels, and crime rates influenced bicyclists' speeds on bridges. This study provides a foundation for understanding the factors associated with bicyclists' speeds on bridges and can inform future safety improvement projects in San Diego County and beyond. The findings highlight the importance of considering a range of factors to improve bicyclist safety and can ultimately lead to safer and more equitable transportation for all.
Developing an Intelligent Transportation Management Center (ITMC) with a Safety Evaluation Focus for Smart Cities
Salehipour, Sina; Jahangiri, Arash; Paolini, Christopher P.; Machiani, Sahar Ghanipoor; Bergcollins, Django (Safe-D University Transportation Center, 2024-01)
In the context of smart cities, ensuring transportation safety is a complex task that involves understanding the impact of new technologies, measuring the effectiveness of safety measures, and identifying high-risk locations. However, recent advances in communication and big data analytics have made it possible to address these challenges in a more efficient manner. Traditional transportation management centers (TMCs) are limited in their ability to analyze large amounts of data for safety evaluation. To overcome this limitation, this project aims to develop an intelligent transportation management center (ITMC) that utilizes automated video analysis to assess safety. By leveraging Intelligent Transportation Systems (ITS) technologies and big data analytics, the proposed ITMC can proactively evaluate safety at signalized intersections. Unlike conventional methods that rely on crash data, the ITMC uses safety surrogate measures (SSMs) to identify near-crash situations and calculate proactive risk. In this study, the results obtained from a machine vision model were used along with the Post Encroachment Time (PET) safety surrogate measure (SSM) to assess safety proactively at a selected signalized intersection. The study utilized the latest YOLO series model, YOLOX, for deep learning to detect and classify road users in video frames from four intersection traffic cameras.
Evaluating the Safe Routes to School (SRTS) Transportation Program in Socially Vulnerable Communities in San Diego County, California
Fernandez, Gabriela; Etaati, Bita; Mercado, Andrick; Jahangiri, Arash; Machiani, Sahar Ghanipoor; Tsou, Ming-Hsiang; Mejia, Christian (Safe-D University Transportation Center, 2023-05)
Child safety concerns are among the strongest impediments to children walking or biking to school, but some students must walk or bike due to financial or other circumstances. These travel modes are more than twice as common among students from low-income households than students from higher income households. The Safe Routes to School (SRTS) program fosters opportunities for students to walk and bike to school safely and routinely. This study provides insights into the SRTS program’s effectiveness and potential to improve walking and biking safety in socially vulnerable communities by evaluating the program’s impact on schools in the Chula Vista Elementary School District, a vulnerable area in San Diego County. (i) A linear regression model was used to assess the program’s impact on each school, and a logistic regression model was employed to identify factors influencing students’ walking behavior. (ii) An SRTS web-based interactive tool (ArcGIS Experience) was developed to identify traffic incident hot spots and facilitate future routing improvements. (iii) A virtual reality (VR) road safety training tool for children was developed, and a case study at Feaster Charter Elementary School was conducted to assess its effectiveness. Twenty-six students played the VR game before and after watching traffic safety educational videos, and observations from the VR session were recorded. (iv) The outreach and deliverables from this study strengthened community collaboration across San Diego County.
A Holistic Work Zone Safety Alert System through Automated Video and Smartphone Sensor Data Analysis
Akhavian, Reza; Jahangiri, Arash; Shahnavaz, Farid; Salehipour, Sina (Safe-D National UTC, 2022-12-16)
This project was inspired by a major gap identified in the literature pertaining to work zone safety monitoring systems that leverage advanced technologies for tracking workers, identifying hazardous situations, and alerting workers of danger. Existing systems target safety hazards that are either external to the work zone (e.g., accidents due to vehicular intrusions) or workers’ internal physical/physiological states (e.g., human-factor ergonomics such as improper or prolonged use of vibrating hand tools). This project presents a holistic approach in which visual and wearable sensor data are used for safety monitoring and alert generation to offer a practical mitigation strategy to both external and internal safety risks. With a major focus on feasibility of adoption and facilitating maintenance, smartphones were used in this project to provide a ubiquitous platform for data collection and communication. A mobile application was developed to generate an alert when unsafe vibration levels were reached in proximity to a high vibration power tool such as a jackhammer. Additionally, visual data collected from surveillance cameras were analyzed to detect speeding vehicles approaching the work zone. In either of these situations, a worker with the application running on their smartphone would be alerted of the internal or external safety hazard.
Identifying High-Risk Intersections for Walking and Bicycling Using Multiple Data Sources in the City of San Diego
Hasani, Mahdie; Jahangiri, Arash; Sener, Ipek Nese; Munira, Sirajum; Owens, Justin M.; Appleyard, Bruce; Ryan, Sherry; Turner, Shawn M.; Machiani, Sahar Ghanipoor (Hindawi, 2019-06-16)
Over the last decade, demand for active transportation modes such as walking and bicycling has increased. While it is desirable to provide high levels of safety for these eco-friendly modes of travel, unfortunately, the overall percentage of pedestrian and bicycle fatalities increased from 13% to 18% of total road-related fatalities in the last decade. In San Diego County, although the total number of pedestrian and bicyclist fatalities decreased over the same period of time, a similar trend with a more drastic change is observed; the overall percentage of pedestrian and bicycle fatalities increased from 19.5% to 31.8%. This study aims to estimate pedestrian and bicyclist exposure and identify signalized intersections with highest risk for walking and bicycling within the city of San Diego, California, USA. Multiple data sources such as automated pedestrian and bicycle counters, video cameras, and crash data were utilized. Data mining techniques, a new sampling strategy, and automated video processing methods were adopted to demonstrate a holistic approach that can be applied to identify facilities with highest need of improvement. Cluster analysis coupled with stratification was employed to select a representative sample of intersections for data collection. Automated pedestrian and bicycle counting models utilized in this study reached a high accuracy, provided certain conditions exist in video data. Results from exposure modeling showed that pedestrian and bicyclist volume was characterized by transportation network, population, traffic generators, and land use variables. There were both similarities and differences between pedestrian and bicycle models, including different spatial scales of influence by mode. Additionally, the study quantified risk incorporating injury severity levels, frequency of victims, distance crossed, and exposure into a single equation. It was found that not all intersections with the highest number of pedestrian and bicyclist victims were identified as high-risk after exposure and other factors such as crash severity were taken into account.
Investigating Violation Behavior at Intersections using Intelligent Transportation Systems: A Feasibility Analysis on Vehicle/Bicycle-to-Infrastructure Communications as a Potential Countermeasure
Jahangiri, Arash (Virginia Tech, 2015-10-06)
The focus of this dissertation is on safety improvement at intersections and presenting how Vehicle/Bicycle-to-Infrastructure Communications can be a potential countermeasure for crashes resulting from drivers' and cyclists' violations at intersections. The characteristics (e.g., acceleration capabilities, etc.) of transportation modes affect the violation behavior. Therefore, the first building block is to identify the users' transportation mode. Consequently, having the mode information, the second building block is to predict whether or not the user is going to violate. This step focuses on two different modes (i.e., driver violation prediction and cyclist violation prediction). Warnings can then be issued for users in potential danger to react or for the infrastructure and vehicles so they can take appropriate actions to avoid or mitigate crashes. A smartphone application was developed to collect sensor data used to conduct the transportation mode recognition task. Driver violation prediction task at signalized intersections was conducted using observational and simulator data. Also, a naturalistic cycling experiment was designed for cyclist violation prediction task. Subsequently, cyclist violation behavior was investigated at both signalized and stop-controlled intersections. To build the prediction models in all the aforementioned tasks, various Artificial Intelligence techniques were adopted. K-fold Cross-Validation as well as Out-of-Bag error was used for model selection and validation. Transportation mode recognition models contributed to high classification accuracies (e.g., up to 98%). Thus, data obtained from the smartphone sensors were found to provide important information to distinguish between transportation modes. Driver violation (i.e., red light running) prediction models were resulted in high accuracies (i.e., up to 99.9%). Time to intersection (TTI), distance to intersection (DTI), the required deceleration parameter (RDP), and velocity at the onset of a yellow light were among the most important factors in violation prediction. Based on logistic regression analysis, movement type and presence of other users were found as significant factors affecting the probability of red light violations by cyclists at signalized intersections. Also, presence of other road users and age were the significant factors affecting violations at stop-controlled intersections. In case of stop-controlled intersections, violation prediction models resulted in error rates of 0 to 10 percent depending on how far from the intersection the prediction task is conducted.
Modeling and Assessing Crossing Elimination as a Strategy to Reduce Evacuee Travel Time
Jahangiri, Arash (Virginia Tech, 2012-12-13)
During evacuations, emergency managers and departments of transportation seek to facilitate the movement of citizens out of impacted or threatened areas. One strategy they may consider is crossing elimination, which prohibits certain movements at intersections, that may be permissible under normal operating conditions. A few previous studies examined this strategy in conjunction with contra-flow operations, but fewer have considered crossing elimination by itself. This study helps fill the existing gap in knowledge of the individual effects of crossing elimination. A bi-level model that iterates between optimization and simulation is developed to determine the optimal configuration of intersection movements from a set of pre-specified possible configurations for intersections in a given area. At the upper level, evacuees' travel time is minimized and at the lower level, traffic is assigned to the network with the traffic assignment-simulation software DynusT. The overall model is solved with a simulated annealing heuristic and applied to a real case study to assess the impact of crossing elimination. Three scenarios are developed and examined using the solution method proposed in this research. These scenarios are developed using combinations of two elements: (1) Evacuee destination distributions, and (2) Evacuee departure time distributions. Results showed about 3-5 percent improvement in total evacuee travel time can be achieved in these scenarios. Availability of through movements at intersections and existing merging points in movement configurations are the two factors influencing the selection of movement configurations.
Predicting Vehicle Trajectories at Intersections using Advanced Machine Learning Techniques
Jazayeri, Mohammad Sadegh; Jahangiri, Arash; Machiani, Sahar Ghanipoor (SAFE-D: Safety Through Disruption National University Transportation Center, 2021-05)
The ability to accurately predict vehicle trajectories is essential in infrastructure-based safety systems that aim to identify critical events such as near-crash situations and traffic violations. In a connected environment, important information about these critical events can be communicated to road users or the infrastructure to avoid or mitigate potential crashes. Intersections require special attention in this context because they are hotspots for crashes and involve numerous and complex interactions between road users. In this project, we developed an advanced machine learning method for trajectory prediction using B-spline curve representations of vehicle trajectories and Inverse Reinforcement Learning (IRL). B-spline curves were used to represent vehicle trajectories, and a neural network model was trained to predict the coefficients of these curves. Small perturbations of these predicted coefficients were used to create candidate trajectories. These candidate trajectories were then ranked according to a reward function that was obtained by training an IRL model on the (spline smoothed) vehicle trajectories and the surroundings of the vehicles. In our experiments we found that the neural network model outperforms a Kalman filter baseline and the addition of the IRL ranking module further improves the performance of the overall model.
Safety Impact Evaluation of a Narrow-Automated Vehicle-Exclusive Reversible Lane on an Existing Smart Freeway
Machiani, Sahar Ghanipoor; Jahangiri, Arash; Melendez, Benjamin; Katthe, Anagha; Hasani, Mahdie; Ahmadi, Alidad; Musial, Walter B. (SAFE-D: Safety Through Disruption National University Transportation Center, 2021-02)
This study fills the gap in the limited research on the effect of emerging Automated Vehicle (AV) technology on infrastructure standards. The main objective of this research is to evaluate implications of an innovative infrastructure solution, exclusive AV lanes, for safe and efficient integration of AVs into an existing transportation system. Examining a real-world case study, this project investigates implications of adding a narrow reversible AV exclusive lane to the existing configuration of the I-15 expressway in San Diego, resulting in a 9-foot AV reversible lane, and in both directions of travel, two 12-feet lanes for HOV and HOT vehicles. Given the difference between the operation of AVs and human-driven vehicles and reliance of AVs on sensors as opposed to human capabilities, the question is whether we can provide exclusive and narrower roadways for AVs while maintaining proper safety and mobility? To accomplish the project’s goal, the research team conducted a series of research approaches including a literature review, an AV manufacturers product review, expert interviews, a consumer questionnaire review, a crash data analysis, and a traffic simulation analysis. Recommendations and guidelines from the results of the study may be used for practitioners and professional organizations involved or interested in AV development.

Browsing by Author "Jahangiri, Arash"

Results Per Page

Sort Options