Genetic algorithm fuzzy clustering using GPS data for defining level of service criteria of urban streets

Smruti Sourava Mohapatra, P.k Bhuyan and K.V.Krishna Rao

European Transport \ Trasporti Europei, 2012, issue 52, 8

Abstract: Developing countries like India need to have proper Level of Service (LOS) criteria for various traffic facilities as this helps in planning, design of transportation projects and also allocating resources to the competing projects. The LOS analysis for urban street followed in India is an adaptation of HCM-2000 methodology but the methodology is relevant for developed countries having homogenous traffic flow. In this research an attempt has been made to establish a framework to define LOS criteria of urban street in Indian context keeping in mind the geometric and surrounding environmental characteristics. Defining LOS criteria is basically a classification problem for which cluster analysis is a suitable technique can be applied. In this research a hybrid algorithm comprising of Genetic Algorithm (GA) and Fuzzy C-mean is utilized. As input to the clustering algorithm GA-Fuzzy a lot of speed data is required. From literature review GPS is found to be a suitable tool for collecting second by second speed data and GIS is suitable in handling large amount of speed data. The clustering algorithm is used twice in this study. First the GA-Fuzzy algorithm was used to classify Free Flow Speed (FFS) data into number of classes in order to get the FFS ranges of different urban street classes. To determine the optimal number of cluster using FFS data five cluster validation parameters are considered. After getting the FFS ranges for different urban street classes the same GA-Fuzzy algorithm is used on average travel speed data collected during both peak and off-peak hours to determine the speed ranges of different LOS categories. From this analysis the free flow speed ranges for different urban street classes and the speed ranges for different LOS categories are defined and the values are found to be lower than that suggested by HCM-2000. The coherence of the clustering result in classification of urban streets into four classes and speed values into six LOS categories is agreed with the physical and surrounding environmental characteristics of road segments under the study area. From this analysis it is also found that good LOS can’t be expected from urban street segment for which physical and surrounding environmental characteristics are not good.

Keywords: Urban Streets; Level of Service; GPS; Genetic Algorithm; Fuzzy C-mean clustering; Cluster Validation (search for similar items in EconPapers)
Date: 2012
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10077/8187 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sot:journl:y:2012:i:52:p:8

Access Statistics for this article

European Transport \ Trasporti Europei is currently edited by Romeo Danielis

More articles in European Transport \ Trasporti Europei from ISTIEE, Institute for the Study of Transport within the European Economic Integration
Bibliographic data for series maintained by Romeo Danielis ( this e-mail address is bad, please contact ).