You are here

Connectionist Approach to Developing Highway Vehicles Classification for Use in Florida

Title: Connectionist Approach to Developing Highway Vehicles Classification for Use in Florida.
50 views
11 downloads
Name(s): Kwigizile, Valerian, author
Mussa, Renatus, professor directing thesis
Sobanjo, John O., committee member
AbdelRazig, Yassir AbdAlla, committee member
Department of Civil and Environmental Engineering, degree granting department
Florida State University, degree granting institution
Type of Resource: text
Genre: Text
Issuance: monographic
Date Issued: 2004
Publisher: Florida State University
Place of Publication: Tallahassee, Florida
Physical Form: computer
online resource
Extent: 1 online resource
Language(s): English
Abstract/Description: Federal, State, and local agencies use vehicle classification data for planning, design, and conducting safety and operational evaluation of highway facilities. The Federal Highway Administration (FHWA) Office of Highway Planning requires states to furnish vehicle classification data as part of the Highway Performance Monitoring System (HPMS). In conformity with the federal reporting requirements, most states use the "F" scheme to classify vehicles. Also, the mechanistic-empirical pavement design methodology being developed under the National Cooperative Highway Research Program Project 1-37A will require accurate classification of vehicles in order to develop axle load spectra information needed as the design input. "Scheme F", used by most states to classify vehicles, can also be used to develop the required load spectra. Unfortunately, the scheme is difficult to automate and is prone to errors resulting from imprecise demarcation of class thresholds. In this work, the classification problem was viewed as a pattern recognition problem in which pattern recognition techniques such as probabilistic neural networks (PNN) was used to assign vehicles to their correct classes and hence to establish optimum axle spacing thresholds. The PNN was developed, trained and applied to field data composed of individual vehicle's axle spacing, number of axles per vehicle and overall vehicle weight. The PNN reduced the error rate from 9.7 percent to 6.1 percent compared to an existing classification algorithm used by the State of Florida Department of Transportation (FDOT). The inclusion of overall vehicle weight as a classification variable further reduced the error rate from 6.1 percent to only 2.9 percent. The promising results from the neural networks were used to set up new thresholds that reduce classification error rate. The proposed vehicle classification table thresholds were validated using additional data collected from the field. The validation results indicated a significant improvement in the accuracy of vehicle classification table compared to the existing FDOT table. The developed table will enable the FDOT to consistently collect more accurate vehicle classification data by using any vendor's equipment. This will enable the state to conduct more accurate environmental impact analysis during highway design and schedule for timely highway maintenance basing on projected remaining pavement life.
Identifier: FSU_migr_etd-3029 (IID)
Submitted Note: A Thesis submitted to the Department of Civil and Environmental Engineering in partial fulfillment of the requirements for the degree of Master of Science.
Degree Awarded: Fall Semester, 2004.
Date of Defense: October 26, 2007.
Keywords: Vehicle Classification, Probabilistic Neural Networks
Bibliography Note: Includes bibliographical references.
Advisory Committee: Renatus Mussa, Professor Directing Thesis; John O. Sobanjo, Committee Member; Yassir AbdAlla AbdelRazig, Committee Member.
Subject(s): Civil engineering
Environmental engineering
Persistent Link to This Record: http://purl.flvc.org/fsu/fd/FSU_migr_etd-3029
Owner Institution: FSU

Choose the citation style.
Kwigizile, V. (2004). Connectionist Approach to Developing Highway Vehicles Classification for Use in Florida. Retrieved from http://purl.flvc.org/fsu/fd/FSU_migr_etd-3029