Modeling Risk-Taking Behavior in Queuing Networks with Advanced Traveler Information Systems

The paper formulates risk taking behavior over general queing network with advanced traveler information systems (ATIS). It supposes that drivers equipped with ATIS will make their choices based on the complete information about travel time and its possible variations, while drivers without ATIS have to choose based on their past experience and subjective understandings. The mixed behavior time dependent equilibrium problem is formulated as a nonlinear complementarity problem and a heuristic route/time swapping approach is proposed to solve the problem. Also investigated is the implementation of ATIS on a toy network under varied values of market penetration.