Control Strategy for Vehicular and Pedestrian Midblock Crossing Movements

On major corridors where mid-block pedestrian crossing exists, traffic flows are frequently interrupted by pedestrian crossing movements. The pedestrian crossings operated by manual control may cause vehicular movements experience unexpected stops and excess travel delay. Specifically in developing countries, the signal timing and pedestrian crossing operations are intuitively judged by individual police patrols or crossing guards. The objectives of this study are (i) to examine the decision mechanism of the crossing guards; (ii) to develop the simulation models that replicate the crossing guard control at midblock pedestrian crossings and evaluate their operational performances; and (iii) to determine the most reasonable and viable control strategy that balances the mobility of vehicular and pedestrian flows. This study proposes a fuzzy logic control strategy to analyse the optimal traffic signal setting for pedestrian crossing signals along an urban corridor. The fuzzy logic signal control is used to characterize subjective and imprecise decision made by police patrols. It is applicable to deal with a complex traffic condition. The study tests the proposed scheme with the real-world congested corridor in Chiang Mai, Thailand. Traffic data were collected from the field during the peak period; and the opinions of the judgment rules were gathered from the police patrols. The study used a microscopic traffic simulation to evaluate the measures of effectiveness of the proposed control strategy and compare its performance with the conventional pedestrian crossing signal including fixed-time and demand-actuated signal operation. The expected results of this study are (i) an alternative approach to control traffic signals for both vehicular and pedestrian flows; and (ii) a simulation-based environment that can evaluate vehicular and pedestrian performance measures.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01642426
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jul 27 2017 2:15PM