Graduation Year

2024

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Michael Maness, Ph.D.

Committee Member

Fred Mannering, Ph.D.

Committee Member

Yu Zhang, Ph.D.

Committee Member

Changhyun Kwon, Ph.D.

Committee Member

Andrei Barbos, Ph.D.

Keywords

Congestion Pricing, Bidding, Choice Modelling, Public Opinions, Fairness

Abstract

The value of time (VOT), or willingness to pay (WTP) to reduce travel time in the context of transportation, is not homogeneous across all road users. Everyone has their own unique VOT and may have multiple values for the VOT associated with factors such as the time of day or the trip purpose. However, the traditional distance-based, time-based, or demand-based congestion pricing methods cannot reflect the true WTP for the use of tolled facilities of the road users. Recent studies have explored and proposed an innovative congestion pricing scheme that integrates auction mechanisms into tolling systems, known as auction-based tolling systems. This system requires road users to participate in an auction and bid for access to the tolled facilities. Through the employed auction mechanism, the price determined by the system could potentially reflect road users’ true WTP, as their submitted bids in an auction would likely represent their desired toll price that aligns with their needs and WTP.

This hypothetical pricing scheme has been studied in various applications within the context of traffic management such as parallel facilities, cordons/geofences, network-level traffic assignments, and bottleneck permits. However, there are research gaps found in the literature. Firstly, there is a lack of empirical evidence supporting the assumption that bidding behavior in an auction-based tolling system follows equilibrium strategy. Secondly, potential equity issues may arise in an auction due to differences in the financial capacities among road users, and there is insufficient consideration of these equity concerns in the design of auction-based tolling systems. Hence, this dissertation aims to address these two gaps.

To address the first gap, a series of economic experiments were conducted to explore how participants bid in an auction-based tolling system and to examine whether they bid closely to equilibrium strategies under various auction settings characterized by different pricing rules, disclosure of auction-related information, and incentive structures. It was observed that participants bid differently across employed pricing rules in the experiment. Specifically, in uniform price auctions, they were found to bid close to their true valuations. In discriminatory price auctions, they lied on auction-related information such as the number of available slots and participants to determine their bids, averagely bidding about 70.00 to 80.00% of their valuation. The auction performances, as reflected in maximum revenue generated, were lower under the experimentally observed bidding behavior compared to that achieved through equilibrium strategies in both uniform and discriminatory price auctions. Additionally, three distinct risk profiles were identified among participants: risk-averse, risk-neutral, and risk-seeking. These risk-profiles were found to have a significant impact on auction performance in terms of revenue generation.

The bidding behavior was further explored in more practical settings, where monetary incentives as provided in economic experiments were absent. In uniform price auctions, it was found that participants bid more conservatively in practical scenarios, specifically significantly lower than their approximate VOT. In descending price auctions, the price sensitivity relative to participants’ approximate VOT had significant impacts on their decision-making. Specifically, they were less likely to accept the offered price when it was higher than their approximate VOT. Additionally, participants who were first introduced to auction-based tolling before a fixed tolling price tended to rely more on their demographics when making choices in descending price auctions.

Although understanding road users’ bidding behavior in an auction-based tolling system provides useful insights, it alone is not sufficient to design systems that effectively meet public needs and expectations. Hence, public opinions and preferences on an auction-based tolling system and the utilization of the revenue generated from such systems were explored. A stated preferences (SP) survey was conducted to indirectly learn about public thoughts on an auction-based tolling system by collecting opinions on closely related systems such as tolling, electronic toll collection (ETC), and auctions, rather than directly asking about the auction-based system. The survey also gathered participants’ preferences on how the revenue generated should be utilized. Based on their responses, four distinct clusters were identified: Tolling and Auction Opponents, Fairness Advocates, Auction Supporters, and ETC Supporters but Auction Opponents. Members within each cluster shared similar opinions such as support for auction as an efficient and effective method of allocating items, comfort with using technology-based tolling systems like ETC, or a preference for systems that promote fairness. Regarding the utilization of revenue, participants mostly preferred to use revenue for road maintenance and investing in other transportation modes, which provide more tangible benefits to society compared to allocating revenue to non-winning users.

Download

Share

COinS