published journal article
Phd Dissertation
This dissertation focuses on the development of optimization methods and approximation algorithms for combinatorial auctions, particularly with application to the contract procurement problem in freight transportation. Combinatorial auctions are auctions in which a set of heterogeneous items are sold simultaneously and in which bidders can bid for their preferred combinations of items. They involve many difficult optimization problems both for auction hosts and bidders and have received significant attention from computer scientists, operations researchers and economists recently. Large shippers (typically manufacturing companies or retailers) have begun to use this method to procure services from trucking companies and logistics services providers. This dissertation first analyzes the economic impact of combinatorial auction-based procurement methods both on shippers and carriers using a simulation study and reveals that both parties can benefit from this economically efficient price discovery mechanism. While the majority of prior research has been from an auctioneer’s perspective, we demonstrate that bidders have even more complicated optimization problems in combinatoiral auctions. The bid construction problem, that is, how bidders should identify and construct beneficial bids, is very hard and remains an open question. This dissertation investigates this problem and proposes an optimization based approximation method that involves solving an NP-hard problem only once, yielding significant improvements in computational efficiency. Further, the current state of trucking and third party logistics industries are examined. The trucking industry is very competitive and small carriers are operating under thin margins. This dissertation addresses these issues by proposing an auction based collaborative carrier network in which participating carriers can identify inefficient lanes from daily operations quickly and exchange them with partners under an auction protocol. This system is proved to be Pareto efficient. Further, decision problems are discussed regarding how carriers should identify inefficient operations and how to make and select bids. This represents an effort to use advanced auction mechanisms to enhance the carriers’ operational efficiencies.
published journal article
research report
The increasing diversity of vehicle type holdings and growing demand for BEVs and PHEVs have serious policy implications for travel demand and air pollution. Consequently, it is important to accurately predict or estimate the preference for vehicle holdings of households as well as the vehicle miles traveled by vehicle body- and fuel-type to project future VMT changes and mobile source emission levels. Leveraging the 2019 California Vehicle Survey data, this report presents the application of a utility-based model for multiple discreteness that combines multiple vehicle types with usage in an integrated model, specifically the MDCEV model. The model results suggest the important effects of household demographics, residence location, and built environment factors on vehicle body type and powertrain choice and usage. Further the predictions associated with changes inbuilt environment factors like population density can inform the design of land-use and transportation policies to influence household vehicle holdings and usage that can in turn impact travel demand and air quality issues in California.View the NCST Project Webpage
published journal article
Background. Exposure to high air temperature in late pregnancy is increasingly recognized as a risk factor for preterm birth (PTB). However, the combined effects of heatwaves with air pollution and green space are still unexplored. In the context of climate change, investigating the interaction between environmental factors and identifying communities at higher risk is important to better understand the etiological mechanisms and design targeted interventions towards certain women during pregnancy. Objectives. To examine the combined effects of heatwaves, air pollution and green space exposure on the risk of PTB. Methods. California birth certificate records for singleton births (2005–2013) were obtained. Residential zip code-specific daily temperature during the last week of gestation was used to create 12 definitions of heatwave with varying temperature thresholds and durations. We fit multi-level Cox proportional hazard models with time to PTB as the outcome and gestational week as the temporal unit. Relative risk due to interaction (RERI) was applied to estimate the additive interactive effect of air pollution and green space on the effect of heatwaves on PTB. Results. In total, 1 967 300 births were included in this study. For PM2.5, PM10 and O3, we found positive additive interactions (RERIs >0) between heatwaves and higher air pollution levels. Combined effects of heatwaves and green space indicated negative interactions (RERIs <0) for less intense heatwaves (i.e. shorter duration or relatively low temperature), whereas there were potential positive interactions (RERIs >0) for more intense heatwaves. Conclusion. This study found synergistic harmful effects for heatwaves with air pollution, and potential positive interactions with lack of green space on PTB. Implementing interventions, such as heat warning systems and behavioral changes, targeted toward pregnant women at risk for high air pollution and low green space exposures may optimize the benefits of reducing acute exposure to extreme heat before delivery.
working paper
Aggression and the automobile have had a long standing association, yet research on aggressive behavior has neglected the roadway context. This chapter reviews existing work which has included archival analysis, field interview studies, personality research, and field experiments. Among the recurrent themes have been the relationship between aggressivity in driving to accident liability and to violence in the larger social context. Validity issues in road aggression research are discussed, and a typology of roadway aggression is presented. The typology maps a range of contemporary forms, most of which have never been investigated scientifically and have received sparse academic attention otherwise, despite having high social and scientific relevance. Disinhibiting influences that heighten the probability of roadway aggression are discussed.
working paper |
Aug 1990
working paper |
May 1990
working paper |
Sep 1991
working paper |
Jun 1991
working paper |
Sep 1991
working paper
This working paper reports initial results of a set of analyses investigating differences between non-truck involved and truck-involved crashes. Data was selected from the California Department of Transportation TASAS accident reporting system. Results indicate that there are differences in primary collision factors (as assigned by the California Highway Patrol through issuance of traffic citations) between truck-involved and non-truck-involved crashes in the vicinity of freeway interchanges. “Speeding’ in truck-involved crashes is cited at approximately one-half its rate for non-truck-involved crashes. Additionally, there is no statistical difference in the appearance of the “Uninvolved Motorist” TASAS party designation between truck and non-truck-involved crashes.
published journal article
One approach to making transportation more sustainable is to transition away from a car-oriented society. Unfortunately, our understanding of the factors that prompt households to voluntarily forgo their motor vehicles is limited. The 2008 Mobility in Germany (MiD) and the 2012 California Household Travel Survey (CHTS) provide an opportunity to start filling this gap by teasing out what built environment and socio-economic variables impact the likelihood that a household is carless (voluntarily or not) in Germany and in California, two car-dependent societies with different carless rates. Results from our generalized structural equation models show that in both Germany and California, households who reside in denser neighborhoods, closer to transit stations, and who have a lower income or fewer children, are more likely to be voluntarily carless. However, households with more education are more likely to be voluntarily carless in Germany, whereas the reverse is true in California. Moreover, employment density and public transit have a higher impact on voluntary carlessness in Germany than in California. Our results also show that different socio-economic groups have substantially different residential location preferences in Germany and in California. These differences may be explained by cultural preferences, historical differences in land use and transportation policies, and by the higher cost of owning a motor vehicle in Germany.
Preprint Journal Article
The environmental and resilience concerns and the potential mass adoption of electric vehicles (EVs) call for managing and operating power-mobility systems (PMS) as a whole. Agent-based and network flow-based modeling methods have proven useful for potential model-based PMS management and operation, but they tend to be heavy in computation, lack intuitive implications of overall PMS dynamics, and are inflexible to the change of the problem scope and resolution. These challenges call for a complementary modeling approach to facilitate model-based large-scale urban PMS decision-making. To this end, we propose a dynamic macroscopic modeling framework for high-level strategic decision-making. The framework contains three main components – power production and delivery, energy exchange interface between power grids and EVs, and EV-involved mobility systems. A case study shows that this approach facilitates tractable and coherent model development with low computational requirements and shows promising benefits of coordinated PMS.
published journal article |
Sep 2023
published journal article |
Jan 2025
published journal article |
Jul 2022
Phd Dissertation
A major source of urban freeway delay in the United States is non-recurring congestion caused by incidents such as accidents, disabled vehicles, spilled loads, temporary maintenance and construction activities, signal and detector malfunctions, and other special and unusual events that disrupt the normal flow of traffic. The automated detection of freeway incidents is an important function of a freeway traffic management center. Early detection of incidents is vital for formulating effective response strategies such as timely dispatch of emergency services and incident removal crews, control and routing of traffic around the incident location, and provision of real-time traffic information to motorists. A number of incident detection algorithms, based on conventional approaches, have been developed over the past several decades, and a few of them are being deployed at urban freeway systems in major cities. These conventional algorithms have met with varying degree of success in their detection performance. In this research, a new incident detection technique based on an artificial neural network approach has been proposed. The objective of this research was to demonstrate the use of artificial neural network models for automated detection of lane-blocking incidents on urban freeways. The study focused on the application of neural network models in classifying traffic surveillance data obtained from inductive loop detectors, and the use of the classified output to detect an incident. Three types of neural network models were developed to detect lane-blocking incidents: the multi-layer feed-forward neural network, self-organizing feature map and adaptive resonance theory 2. The models were developed with simulation data from a study site and tested with both simulation and field data at the study site and other locations. The multi-layer feed-forward neural network was found to have the highest potential among the four models to achieve a better incident detection performance. This network consistently detected most of the lane-blocking incidents and gave a false alarm rate lower than the conventional algorithms currently in use. The results have demonstrated the potential of artificial neural network models in improving incident detection performance over currently available techniques.
research report |
Sep 1993