Network Sensor Location Problem based on Genetic Algorithm for Minimizing Variability of Traffic Information

Traffic information such as speed, occupancy, and flow is essential factors of transportation planning and operation. It follows that determining the optimal location of a sensor for collecting the traffic information is one of the key problems in the design of an urban transport system. The aim of this paper is to develop a model for the network sensor location problem (NSLP) to minimize the variability of traffic information for the entire network. In order to reflect the discrete property of the collected information, this study develops an NSLP model based on the discrete distribution of speed data. The genetic algorithm is selected as the optimization method for the multiple sensor locations for this study, as it is particularly suited to deal with the discrete data and for solving the computationally expensive problems. The specific genetic algorithm fit for the model was also developed by our research. To assess the applicability of the model, speed data collected from the dedicated short-range communication (DSRC) system in Daegu city are used. Based on the findings, it is possible to quantify the effects of sensor locations on the accuracy of data estimation for the entire network and determine the optimal sensor locations with computational efficiency. It is also shown that the value of variability decreases sharply as the number of sensors increases. The application of such a model will enhance the investment efficiency and improve the accuracy of information in transportation.

  • Supplemental Notes:
    • This paper was sponsored by TRB committee ABJ35 Standing Committee on Highway Traffic Monitoring.
  • Corporate Authors:

    Transportation Research Board

    ,    
  • Authors:
    • Yang, Jae Hwan
    • Hwang, Hyunjun
    • Kho, Seung-Young
    • Kim, Dong-Kyu
  • Conference:
  • Date: 2019

Language

  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; Maps; References; Tables;
  • Pagination: 6p

Subject/Index Terms

Filing Info

  • Accession Number: 01697757
  • Record Type: Publication
  • Report/Paper Numbers: 19-01402
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 7 2018 9:36AM