Data-Driven Approaches for Assessing the Impact of Speed Management Strategies for Arterial Mobility and Safety
Author
Karimpour, AbolfazlIssue Date
2020Keywords
ArterialsSignal Retiming
Speed Feedback Sign
Speed Management Strategy
Traffic Mobility
Traffic Safety
Advisor
Wu, Yao-Jan
Metadata
Show full item recordPublisher
The University of Arizona.Rights
Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.Abstract
Arterials are the connector of the national transportation system to regional mobility. Arterials play a significant role in providing accessibility to residential and commercial neighborhoods. Therefore, they are essential to the regional economy and residents' quality of life. Through the Moving Ahead for Progress in the 21st Century Act (MAP-21), US Congress requires that all the state Departments of Transportation (DOTs) and Multimodal Planning Organizations (MPOs) monitor, improve, and maintain the mobility and safety performance of their jurisdiction’s road network. A simple and straightforward solution to improve arterial mobility and safety would be constructing new roads. However, due to the limited infrastructure and high construction cost, this solution is not always feasible. One viable solution to enhance the mobility and safety of arterials is using ITS technologies. Speed management strategies are one of the emerging ITS technologies that are currently been utilized by different state departments of transportation to improve the safety and mobility of their road network. This dissertation focuses on proposing comprehensive data-driven speed management strategies and evaluating their impact on the mobility and safety of signalized arterials. This dissertation consists of the following components. One important issue before conducting any traffic studies is validating traffic data quality. Low quality and incomplete traffic data will negatively impact traffic projects’ outcomes. This component of the dissertation aims to develop a data-driven hybrid model to impute the missing and incomplete values. The proposed model imputes missing values by considering the interaction, similarity, and differences of the data as well as incorporating available historical information. The application of the proposed model was used to impute missing truck travel time data in the National Performance Measures Research Dataset (NPMRDS). The analysis result showed that the proposed model is able to impute severe continuous missing data with high accuracy. The comparison results also showed the proposed model will outperform other conventional models while dealing with severe missing conditions. The clean and high-quality data achieved from the first component will be used for identifying the segments with speeding problems and then identifying appropriate speed management strategies. A considerable amount of research has demonstrated a direct relationship between speed and both crash frequency and crash severity. Therefore, speed management strategies to impose the speed limit and tackle speeding are important for transportation agencies to improve mobility and safety at the corridor level. In this component of the dissertation the effects of implementing several speed management strategies, namely speed feedback signs, periodic law enforcement, and speed feedback sign supported with periodic law enforcement on driver speed behavior and compliance was examined. To analyze the effectiveness of each strategy, nine locations in Pima County, Arizona, were selected in a cross-sectional framework. The results of this component showed that supporting SFS with periodic law enforcement could be a key speed management strategy that takes advantage of the strengths of both SFS and law enforcement. Further, the results showed the existence of periodic law enforcement could potentially modify drivers’ behaviors and increase the spatial effectiveness of speed feedback signs. When it comes to mobility and safety evaluation of speed management strategies, previous studies only focused more on the corridor level instead of breaking the evaluation into the link (the segment between two intersections) and the intersection levels. Furthermore, the majority of the studies used historical crash data to investigate the safety impact of speed management strategies. Collecting such data takes time and effort. Therefore, this component of the dissertation focuses more on evaluating the impact of speed feedback sign at both link and intersections levels. The application of this component was implemented on a west/east arterial in Pima County, AZ. The results of intersection level data analysis showed no statistically significant differences in either mean or variance of the signalized performance measures before and after disabling the speed feedback sign. Moreover, it was found that the impact of speed feedback sign on driver’s behavior is a function of their approaching speed. Finally, the benefit in dollar value per year associated with a reduction in severe crashes on the study arterial with active SFS showed promising safety enhancement. Traffic signal retiming is another effective speed management strategy that significantly impacts the overall mobility of signalized arterials. Implementing correct signal timing plans and periodically retiming them will result in direct and indirect benefits such as reduction in the delay and travel time as direct benefits, and reduction in fuel consumption, air pollution, pavement wear, and tear as indirect benefits. In this component of the dissertation, a systematic step-wise approach is proposed that can assists transportation agencies to frequently fine-tune their signal timing parameters, rather than retiming the whole corridor every three to five years. In addition, the proposed approach will allow transportation agencies to predict the intersection mobility performance prior to the field implementation. The application of the proposed approach was implemented on multiple intersections on a major corridor in Pima County, Arizona. The prediction results showed that by only fine-tuning the green split, we are able to achieve on average 10% improvement on the intersection simple delay. The outcome of this dissertation could help DOTs and MPOs to use ITS technologies, to monitor, improve, and maintain the mobility and safety performance of their jurisdiction’s road network. More specifically, the data-driven safety and mobility approaches conducted in this study could: 1-provide a framework for missing data imputation before conducting traffic studies, 2-help transportation agencies to use the high-quality data to identify the high-risk location and identify the appropriate speed management strategy, 3- analyze the impact of speed management strategies at the corridor and intersection level, and 4- help transportation agencies with a more efficient way for signal retiming, enhance arterial mobility and consequently save money and resources.Type
textElectronic Dissertation
Degree Name
Ph.D.Degree Level
doctoralDegree Program
Graduate CollegeCivil Engineering