Efficient and Equitable Travel Demand Management Using Price and Quantity Controls

Traffic congestion is a serious problem that imposes significant costs on the economy, environment, and society. Congestion pricing as a demand management instrument has been known to be a cost-effective approach to deal with congestion. However, the issue of equity remains one of the major challenges to the successful design, acceptance, and deployment of congestion pricing. Although refunding revenues in a personalized manner has the potential to improve its acceptance by being Pareto-improving, there is limited research on methodologies to do so. An alternative approach to travel demand management termed tradable mobility credits (TMC) has been gaining attention recently. It is a type of quantity control which can avoid the flow of money from users to the regulator and has been shown to have better performance than pricing under demand and supply uncertainty. Despite these promises, several important questions remain with regard to the design and functioning of the market within the TMC schemes, an aspect critical to the effective operationalization of these schemes. The objective of this thesis is to design the efficient, equitable and Pareto improving congestion tolling for both price and quantity controls. First, we develop a market design for TMC schemes that ensures TMC is used for mobility management and avoids undesirable behavior such as hoarding, frequent selling and speculation, excessive activity at boundary (of token expiration), and negotiation cost. The developed design considers all aspects of market including token allocation, expiration, transaction fee, price adjustment and market rules governing trading. In addition, a heuristic approach to model disaggregate selling behavior is developed and the resulting simple selling strategy is derived. The developed market design addresses a growing and imminent need to develop methodologies to realistically model TMC schemes that are suited for real-world deployments. Second, we develop a bi-level optimization framework for personalized distribution to make congestion tolling (both price and quantity controls) efficient, equitable, and Pareto improving. The system optimization determines the toll policy with the objective to maximize social welfare while the user optimization can be formulated with different objectives (e.g. to achieve Pareto improvement or maximize social welfare) to determine an individual-specific distribution of revenue for pricing or mobility credits for TMC. The developed personalized congestion tolling is promising as it addresses the important issue of equity and has the potential to improve public acceptance. The performance of the designed instruments are demonstrated via microsimulation in a daily commute context between a single origin-destination pair. The simulation experiments employ a day-to-day assignment framework wherein transportation demand is modeled using a logit-mixture model with the nonlinear income effects and supply is modeled using a standard bottleneck model. The evaluation framework includes four main categories: social welfare, distributional impacts, behavior change, and level of congestion.Ph.D