A Marginal-cost Pricing Model for Transportation Networks with Multiple-class Users

Abstract

As the increasing social and economic activities the demand and usage of private vehicles have been constantly increasing However because road capacity is limited travel demand is periodically greater than road capacity during specific time intervals and road segments causing traffic congestion and air pollution problems In Taiwan severe traffic congestion occurs on specific locations and time periods In order to mitigate traffic congestion problem several traffic control and management schemes have been evaluated and/or implemented but the results are not satisfactory Among the traffic management strategies road (congestion) pricing policy which charges specific road users with different characteristics by various tolls have been either tested or implemented around the world and the results indicate that road (congestion) pricing policy is one of the effective methods to mitigate traffic congestion problems Generally different vehicle types have different characteristics that cause varying travel cost in terms of internal and external costs which in turn influences users’ trip planning and route choice behaviors Therefore it is needed to charge differential tolls based on different vehicles’ characteristics and their impacts on the total societal cost By doing so traffic congestion external cost could be internalized to a certain extent Traffic demand could be accordingly reduced and the ultimate goal of the road pricing policy on resolving traffic congestion problems can be possibly achieved This study solves the differential pricing problem by the marginal cost pricing theory under different vehicle types The empirical study based on the freeway No 5 system and entire freeway network in Taiwan is conducted under different test scenarios The numerical analysis results indicate that the tolls of passenger car and bus need to be larger than twice of the standard tolls (the ceiling by law) on the Hsuehshan Tunnel segment The resulted congestion tolls have an effect on reducing traffic congestion and transferring the exceeded traffic flow to alternative routes Alternatively for the other segments of the freeway No 5 system most resulted tolls are below twice of the standard tolls As for the test results of the entire freeway system the numerical analysis results indicate that congestion tolls for partial road segments are above twice of the standard tolls Therefore the ceiling by law needs to be relaxed so that the tolling policy can be respectively implemented to achieve the ultimate goal of social welfare maximization Finally this study found that using the Diagonalization and Frank-Wolfe algorithms to solve the road pricing problem is not guaranteed to obtain the global optimal solutions for some cases because the objective function of the proposed social welfare maximization-based marginal-cost pricing model is not convex which limits the popularity of the proposed congestion pricing model in practical applications The other solution algorithms need to be pursued to overcome this research limitation

Date of Award	2016 Jul 27
Original language	English
Supervisor	Shou-Ren Hu (Supervisor)