Cognitive Learning Approach for Travel Demand Modeling: Estimation Results

The paper reports progress in the development of an agent-based model of cognitive learning, which simulates spatial perception updating in connection with daily travel behavior based on the principle of Bayesian perception updating. This model is embedded in a multi agent-based model of activity-travel scheduling and choice behavior. The aim of this paper is to empirically estimate the proposed model using data on individuals’ landmark recognition in a field survey. The main findings of the study show that the model fits the data satisfactorily and results are reasonable. The comparison between the proposed Bayesian model and a more basic binary logit model shows that the model improves when prior probabilities are taken into account, which provides evidence for the proposed Bayesian model.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01636093
  • Record Type: Publication
  • Files: TRIS
  • Created Date: May 17 2017 4:49PM