Date of Award
Doctor of Philosophy (PhD)
Committee Member 1
Committee Member 2
Committee Member 3
Greenhouse gas (GHG) emissions, known as a major cause of climate change, have been emitted by the combustion of fossil fuels over the past few decades. The transportation sector contributes significantly to global GHG emissions. Inspired by the successful implementation of tradable credit schemes (TCSs) in pollution control programs, this dissertation focuses on multi-period TCSs to minimize vehicular emissions. In this scheme, a central authority (CA) allocates travel credits to travelers (credit allocation scheme) and then, charges them to travel on each link (credit charging scheme). Travelers are able to trade credits amongst themselves in the market. To address the long-term planning goals of the CA, the dissertation proposes the concept of a multi-period TCS framework. This framework enables the CA to achieve steady progress toward system-level goals, i.e., reducing traffic congestion and GHG emissions, over the long-term planning horizon. First, a TCS-based multi-period equilibrium modeling framework is developed to address the planning problem of a CA that seeks to achieve system-level goals by varying the credit supply and the link usage credit charging schemes across the various periods of the planning horizon. Further, the CA seeks stable credit prices across these periods to provide them as information to travelers in an operational context. Based on this information, bank interest rate and their travel needs, travelers determine their actions in terms of the consumption or sale of credits in the current period or the transfer of credits to future periods. It is proved that the credit price volatility is dampened by the ability to transfer credits. Since a TCS is subject to market manipulation and the artificial control of credit price, a transfer fee, which is shown to be an effective instrument to control hoarding among travelers, is proposed.
Using the proposed multi-period TCS framework, the dissertation develops different system optimal (SO) TCS designs, as bi-level models, to derive credit allocation and charging schemes to achieve system-level goals. In the first SO multi- period TCS design, the CA minimizes the vehicular emissions in the upper level over the long-term planning horizon. This enables the CA to plan the trajectory of vehicular emissions during the planning horizon. This trajectory can be used to predetermine the emissions standard for each period to use in the second SO multi-period TCS design, which aims to minimize total system travel time, in the upper level, over the planning horizon. These designs include bounds on increases in travel costs, allowing travelers to better adapt to the TCS implementation. The lower-level models are the equilibrium conditions in which travelers minimize their costs under the obtained multi-period TCS in the upper level. To enhance realism in capturing the equilibrium conditions under the multi-period TCS, this dissertation factors different travelers’ characteristics and bank interest rates. In making route choices, travelers factor value of time (VOT) and tradeoff credit consumption and travel costs. Hence, travelers’ heterogeneity in terms of VOT is factored. It is shown that if the CA does not factor VOT in SO TCS design, it leads to a socially inequitable policy in practice. Further, the heterogeneity of travelers in terms of perceived future credit prices is factored. Travelers decide to consume or transfer credits in each period based on several factors, including future credit prices. However, due to the uncertainty in traffic network demand/supply forecasts over the long-term horizon, the CA cannot provide an accurate forecast of future credit prices a priori. It is shown that as the difference between travelers’ perceptions of future CPs and the actual CPs set by the CA for each period increases, the effectiveness of the SO TCS design in minimizing total system travel time decreases; this has implications for traffic congestion management. Fourth, the dissertation investigates the robust design of multi-period TCS to account for travel demand uncertainty and achieve system-level goals. To minimize vehicular emissions, the CA leverages the TCS to promote zero-emissions vehicles (ZEVs), which circumvents the need for current subsidy-based incentive policies. The incentive to shift to ZEVs is fostered by allocating more credits and charging fewer credits to ZEV travelers compared to other travelers. To factor the uncertainty in travel demand forecasts, this research proposes a robust multi-period TCS design that minimizes the worst-case vehicular emissions, i.e. maximum vehicular emissions, under different traffic network demand scenarios. It is shown that the robust TCS design increases reliability in achieving system-level goals, compared to the SO TCS design that does not factor travel demand uncertainty. Finally, the dissertation analyzes the ability of a TCS to manage morning commute congestion while factoring the market loss aversion of commuters. It uses a single bottleneck model in a discrete time setting and classifies commuters into groups based on VOT heterogeneity, schedule delay penalty, and desired arrival time. The existence and uniqueness of equilibrium departure rates, credit prices, and travel disutility are investigated under TCS. The SO TCS design is formulated by applying linear programming duality to derive the TCS parameters. It is demonstrated that credit prices and the total value of traded credits approach zero as commuters’ loss aversion increases. The research findings suggest that if the market loss aversion behavior of commuters is not considered, the SO TCS design can lead to an inequitable scheme in which some commuters experience high travel disutility, fostering public opposition. This dissertation contributes significantly to the literature by developing a multi-period TCS framework, from the long-term planning perspective of the CA, that fosters consistency in credit prices across periods, attains long-term system-level objectives, and manages traffic demand over the planning horizon to achieve these goals. It investigates the impacts of different aspects of traveler behavior that affect the progress of the system toward achieving the system-level goals. The proposed robust design of multi-period TCS enables the CA to promote ZEVs and optimizes the system against the worst-case vehicular emissions. Finally, it develops a general modeling framework to investigate the ability of a TCS to manage morning commute congestion within an operational framework, which can also lead to the reduction of vehicular emissions.
Miralinaghi, Seyedmohammad, "Multi-period Tradable Credit Schemes for Transportation and Environmental Applications" (2018). Open Access Dissertations. 1860.