A Reliability-Based Stochastic System Optimum Congestion Pricing Model Under ATIS with Endogenous Market Penetration and Compliance Rate

摘要

This paper presents a reliability-based stochastic system optimum congestion prici1 ngmodel with endogenous market penetration and compliance rate in an Advanced TravelerInformation Systems (ATIS) environment. All travelers are divided into two classes. The firstguided travelers are referred to as the equipped travelers who follow ATIS advice, while thesecond unguided travelers are referred to as the unequipped travelers and the equipped travelerswho do not follow the ATIS advice (also referred to as non-complied travelers). We assume thattravelers will take travel time, congestion pricing, and network travel time reliability into accountwhen making travel route choice decisions. In order to arrive on time, travelers need to allow for asafety margin to their trip. Those guided travelers have lower perception variations on disutilitythan those unguided travelers. The market penetration of ATIS is determined by a continuousincreasing function of the information benefit, and the ATIS compliance rate of equipped travelersis given as the probability of the actually experienced travel costs of guided travelers being lessthan or equal to those of unguided travelers. The whole Origin-Destination (OD) pair traveldemand is determined by the expected minimal perceived travel cost between the OD pair. Theproposed model is formulated as an equivalent variational inequality problem. Numericalexamples illustrate that the stochastic system optimum congestion pricing can help improve thereliability of transport networks. Furthermore, the traditional congestion pricing model may driveneither the traffic network toward the system optimum nor reduce total network travel cost.