Document Type

Dissertation

Degree

Doctor of Philosophy (PhD)

Major/Program

Civil Engineering

First Advisor's Name

Atorod Azizinamini

First Advisor's Committee Title

Committee chair

Second Advisor's Name

Mohammed Hadi

Second Advisor's Committee Title

committee member

Third Advisor's Name

Armin Mehrabi

Third Advisor's Committee Title

committee member

Fourth Advisor's Name

Wallied Orabi

Fourth Advisor's Committee Title

committee member

Fifth Advisor's Name

B M Golam Kibria

Fifth Advisor's Committee Title

committee member

Keywords

Work Zone Safety, Crash Severity, Crash Frequency, Machine Learning

Date of Defense

10-6-2020

Abstract

The attributes of work zones have significant impacts on the risk of crash occurrence. Therefore, identifying the factors associated with crash severity and frequency in work zone locations is of important value to roadway safety. In addition, the significant loss of workers’ lives and injuries resulting from work zone crashes indicates the emergent need for a comprehensive and in-depth investigation of work zone crash mechanisms.

The cost of work zone crashes is another issue that should be taken into account as work zone crashes impose millions of dollars on society each year. Applying innovative construction methods like Accelerated Bridge Construction (ABC) dramatically decreases on-site construction duration and thus improves roadway safety. This safe and cost-effective procedure for building new bridges or replacing/rehabilitating existing bridges in just a few weeks instead of months or years may prevent crashes and avoid injuries as a result of work zone presence.

The application of machine learning techniques in traffic safety studies has seen explosive growth in recent years. Compared to statistical methods, MLs are more accurate prediction models due to their ability to deal with more complex functions. To this end, this study focuses on three major areas: crash severity at construction work zones with worker presence, crash frequency at bridge locations, and assessment of the associated costs to calculate the contribution of safety to the benefit-cost ratio of ABC as compared to conventional methods.

Some key findings of this study can be highlighted as in-depth investigation of contributing factors in conjunction with the results from statistical and machine learning models, which can provide a more comprehensive interpretation of crash severity/frequency outcomes. The demonstration of work zone crashes needs to be modeled separately by time of day for severity analysis with a high level of confidence. Investigation of the contributing factors revealed the nonlinear relationship between crash severity/frequency and contributing factors. Finally, the results showed that the safety benefits from a case study in Florida consisted of 43% of the total ABC implementation cost. This indicates that the safety benefits of ABC implementation consist of a considerable portion of its benefit-cost ratio.

Identifier

FIDC009190

ORCID

0000-0002-2289-0257

Previously Published In

Mokhtarimousavi, S., Anderson, J.C., Azizinamini, A., Hadi, M., 2019. Improved support vector machine models for work zone crash injury severity prediction and analysis. Transportation Research Record 2673 (11), 680-692. DOI:10.1177/0361198119845899.

Share

COinS
 

Rights Statement

Rights Statement

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).