Analysis of Shared Parking Choice Behavior Based on Bayesian Network

Shared parking behavior around the business district has been deeply analyzed. The shared parking lots in the residential district which adjacent to the business district has been taken as the research situation. Using questionnaire survey, the authors analyze and filtrate five key indicators that significantly influence the parking sharing behavior, based on which, various scenarios are designed to collect the shared parking willingness. Bayesian network model is adopted to excavate the relationship between travel time, walking time difference, parking charge difference, parking purpose and the shared parking purpose. The K2 algorithm is applied to train the Bayesian network structure, and the five-fold cross-validation method is conducted to perform the parameter learning to get the conditional probabilities of shared parking under various scenarios. The optimal measure of improving the utilities of shared parking has been proposed through evidence inferring, which can provide theoretical reference for the implementation of parking sharing.

Language

  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; References; Tables;
  • Pagination: 17p

Subject/Index Terms

Filing Info

  • Accession Number: 01764341
  • Record Type: Publication
  • Report/Paper Numbers: TRBAM-21-02177
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 23 2020 11:27AM