Risk Aversion, Road Choice, and the One-Armed Bandit Problem

The authors theoretically analyze the use of advanced traveler information systems for driver road choice for the risk-averse who, over time, learn rationally in a simple setting. The one-armed bandit problem is studied, in which either a random or a safe road is selected by a driver day after day. The authors envisioned four different information regimes. On a random road, in order to acquire information on the road, a locally informed driver needs to select one first, while travel time is known with certainty beforehand by the visionary driver. The authors also envision two intermediary (fully and globally) information regimes. Individual benefits with respect to individual risk aversion, as well as optimal strategies, are compared by the authors after all four regimes are analyzed. Risk aversion impact on dynamic optimal strategies is also illustrated by a numerical example.

  • Availability:
  • Authors:
    • Chancelier, Jean-Philippe
    • De Lara, Michel
    • de Palma, Andre
  • Publication Date: 2007-2

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01049866
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Apr 30 2007 5:34PM