Statistical Clustering Performance in Pavement Condition Prediction as Decision Supporting System Tool

Keywords: Statistical Clustering, Decision Supporting System, Pavement Condition Prediction, Asphalt Pavement Maintenance Management

Abstract

Mathematical methods and statistical patterns have always been considered by managers, designers and science and technology expert in order to develop technology and engineering objectives. During the development of data-gathering tools and increment of data-bases, data mining have made suitable tools in management and engineering. The assessment of roads' maintenance is highly important in order to prevent early deterioration of roads and performing maximum road capacity during the service-life. Pavement management of roads has also implemented this tool to make proper decisions and preferences of pavement repair methods, using decision tree. Through engineering management, cluster analysis is one of the basic tools of data mining and knowledge discovery and makes the decision making, easier in engineering. Data categorization is helpful for planning and is important in picking proper methods. This study was performed by using recorded data from other scientific sources considering data mining method and analyzing data with respect to statistical clustering. The results indicate that bitumen content in asphalt mix, pavement age, marshal strength and rate of passing vehicles have the most important effect on decrement of condition index of pavement, relatively. Also, the highest deterioration in asphalt happens in 5.5% and higher values of bitumen content and the progressive deteriorations take place when the pavement age exceeds 35 years.

Author Biography

Mohammad Mehdi Khabiri, Yazd University, Yazd, Iran
Associate Professor  Fields of Interest

: Pavement Material Mechanic

: Pavement Maintenance Management

: Soil Mechanic in Road and Transportation

: Traffic Safety

Published
2019-12-30
How to Cite
Khabiri, M. M. (2019). Statistical Clustering Performance in Pavement Condition Prediction as Decision Supporting System Tool . Journal of Science and Technology, 11(2), 26-35. Retrieved from https://publisher.uthm.edu.my/ojs/index.php/JST/article/view/4737